{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":247111689,"defaultBranch":"vsearch","name":"cassandra","ownerLogin":"datastax","currentUserCanPush":false,"isFork":true,"isEmpty":false,"createdAt":"2020-03-13T16:10:11.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/573369?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1715962043.0","currentOid":""},"activityList":{"items":[{"before":"88150881f0e982c605be8de9503909566a889add","after":"80c39c140b55f4b23799d7a707c5292892b3094f","ref":"refs/heads/CNDB-9099-rebase-ds-trunk-vsearch","pushedAt":"2024-05-17T22:33:34.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"djatnieks","name":"dan jatnieks","path":"/djatnieks","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2855940?s=80&v=4"},"commit":{"message":"Squashed commit of the following:\n\ncommit 4c4c1ff802c7667e1289a4e730cc34bf6b3bd007\nAuthor: Jonathan Ellis \nDate: Wed Jul 5 12:03:22 2023 -0500\n\n add test for creating index after data is added\n\ncommit 0acaae364c19ed7556b66b5576762ecc131af5fd\nAuthor: Jeremiah D Jordan \nDate: Wed Jul 5 09:23:01 2023 -0500\n\n Revert \"enable segment compaction by default for vector indexes -- we should prioritize read performance\"\n\n This reverts commit a4fa0728337ebf236ad03f6ad708cf51c3dc08c5.\n\ncommit e3ddb6426de4e89b5fe61ed742b69257865c2d3c\nAuthor: Jonathan Ellis \nDate: Wed Jun 28 10:54:49 2023 -0500\n\n VECTOR-64 set sstable_growth to 0.5 by default\n\ncommit 3ca75046e6802ce5836a654459b01948c83d1d7b\nMerge: a4fa072833 10a2a31ae8\nAuthor: Jonathan Ellis \nDate: Wed Jun 28 10:54:18 2023 -0500\n\n merge ds-trunk\n\ncommit a4fa0728337ebf236ad03f6ad708cf51c3dc08c5\nAuthor: Jonathan Ellis \nDate: Wed Jun 28 10:24:41 2023 -0500\n\n enable segment compaction by default for vector indexes -- we should prioritize read performance\n\ncommit 08153b38549d700505d064c25d7a2070c2ff07aa\nMerge: 6a063b4f84 1a45fdc5f9\nAuthor: Jonathan Ellis \nDate: Wed Jun 28 10:18:38 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit 6a063b4f8458f119d4b99bd6b78cbc8832f27f13\nAuthor: Jonathan Ellis \nDate: Wed Jun 28 10:13:51 2023 -0500\n\n VECTOR-63 add cassandra.sai.hnsw.allow_custom_parameters defaulting to false to control setting maximum_node_connections and construction_beam_width, which can cause memory usage to explode if used incorrectly\n\ncommit 1a45fdc5f92e9c82dba6e9e56a5e341806070b13\nAuthor: Jeremiah D Jordan \nDate: Mon Jun 26 15:57:10 2023 -0500\n\n Add the option to configure the file_cache_size_in_mb in a -D like it is in 6.8-cndb.\n\ncommit d2e45dab5b57f20fbdd6250d6dc186c83d783ed3\nAuthor: Jonathan Ellis \nDate: Thu Jun 22 16:13:00 2023 -0500\n\n VECTOR-62 don't flush a graph that only contains deleted vectors\n\ncommit 4a9afe205c20b1f863e89d4e02138557d225c3fc\nAuthor: Jonathan Ellis \nDate: Thu Jun 22 16:11:11 2023 -0500\n\n r/m obsolete comment\n\ncommit a925dc22696bcbebcad17a460441e8b84ff97660\nMerge: 7c7d160dd9 2b22984fe1\nAuthor: Jeremiah D Jordan \nDate: Thu Jun 22 09:03:08 2023 -0500\n\n Merge remote-tracking branch 'datastax/ds-trunk' into vsearch\n\ncommit 7c7d160dd90624b0e365287ef0a2f0b7ef94da53\nMerge: a14b387372 7a5e374cea\nAuthor: Jonathan Ellis \nDate: Tue Jun 20 21:42:51 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit a14b387372ef3f220ece68b019acaafaf0d56f42\nAuthor: Jonathan Ellis \nDate: Tue Jun 20 17:52:13 2023 -0500\n\n move new options to jvm17-server.options\n\ncommit dc67582585a90628e026a693497793b7d8a298ed\nAuthor: Jonathan Ellis \nDate: Tue Jun 20 17:47:27 2023 -0500\n\n copy jvm11 options to jvm17\n\ncommit b6717410f925486a858c4c9414e417eed50950d7\nAuthor: Jonathan Ellis \nDate: Tue Jun 20 17:30:48 2023 -0500\n\n enable simd\n\ncommit 92b9e6e2f136254fb1f8e50732809ea529617f48\nAuthor: Jonathan Ellis \nDate: Mon Jun 19 15:31:09 2023 -0500\n\n make it run under jdk 20\n\ncommit 3e985953637819ac8504955552ce0feba29bdd40\nAuthor: Jonathan Ellis \nDate: Tue Jun 20 17:13:56 2023 -0500\n\n update to lucene with simd\n\ncommit 7a5e374cea785678670444121ad36be6170dd59b\nMerge: 5e5ae4de82 e9ddd5f0d9\nAuthor: Jeremiah D Jordan \nDate: Tue Jun 20 15:00:24 2023 -0500\n\n Merge remote-tracking branch 'datastax/ds-trunk' into vsearch\n\ncommit 5e5ae4de824a7e838bbdfdbabc92e8b256911e3f\nAuthor: Jonathan Ellis \nDate: Mon Jun 19 10:57:18 2023 -0500\n\n we weren't calculating length correctly (supposed to match ordinals range) but it didn't matter b/c length() is not called on the search path where these classes are used. make this explicit by throwing in length implementation.\n\ncommit c0d4055a0bc6914d0220fbd4912d6355e9f6a1d9\nMerge: f2c9a7cb59 c605f8c9f0\nAuthor: Jonathan Ellis \nDate: Mon Jun 19 10:20:02 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit f2c9a7cb59b113a349992011b739079106274c1e\nAuthor: Jonathan Ellis \nDate: Mon Jun 19 10:19:34 2023 -0500\n\n VECTOR-61 remove deletedOrdinals collection that was not threadsafe wrt add and remove operations; create postingsByOrdinal map that allows Bits operations to look up the postings, instead\n\ncommit 9744a13b405cbeb1b68894bb08cc24796aedd8f7\nAuthor: Jonathan Ellis \nDate: Mon Jun 19 09:13:24 2023 -0500\n\n add LVT.testMultiplePostings (works fine)\n\ncommit c605f8c9f037e724544b21fe2aadcf0d8503d86f\nAuthor: Zhao Yang \nDate: Sat Jun 17 12:44:36 2023 +0800\n\n skip replica-filtering-protection for ANN requests\n\ncommit 71935bd07c9218ba70f1ffde7b6db1e88009e529\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 20:20:06 2023 -0500\n\n forgot that maxBruteForceRows has to stay non-final for Byteman\n\ncommit b264afc07727305ce88ab2f18b8a2643d2510bd3\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 20:10:19 2023 -0500\n\n VECTOR-54\n - validateIndexable also checks for very large vectors (that explode to NaN when we take the cosine)\n - check query vectors with the same criteria as vectors to index\n\ncommit 66c31a6dcfcbf81b7657ac9c4119803e026d150a\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 20:07:41 2023 -0500\n\n if CFS.apply catches an InvalidRequestException, keep the same type when we re-throw so it gets back to the client as intended\n\ncommit e0bdfdbe421a1cca30746604cda9ef5b992002bb\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 17:33:37 2023 -0500\n\n add [failing] emptyIndexTest\n\ncommit 75cba96f90730b32ea4a39375c72aaff4ffd6221\nAuthor: Jonathan Ellis \nDate: Thu Jun 15 16:52:46 2023 -0500\n\n re-use bitsets across searches\n\ncommit 2af2ce8a321f7fea268d81c393da7f8f6d886776\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 17:05:52 2023 -0500\n\n more asserts that graph is in sane state when we write it\n\ncommit a63492686d13e12b799acdf50c6f0c20dd680f21\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 17:13:23 2023 -0500\n\n update lucene\n\ncommit 4c6b43e3606b8c9eb5e4ec4cf0817dc9b061c555\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 16:24:33 2023 -0500\n\n add debugging information when node is not found on level\n\ncommit 81c8e730bd4ca314a81bc13e30b197603a8eb005\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 11:57:24 2023 -0500\n\n add similarityWithAnn test\n\ncommit 87c8da9a5dddd10215a84c07a09dbaaecaa522b8\nAuthor: Jonathan Ellis \nDate: Wed Jun 14 18:26:02 2023 -0500\n\n add tests to confirm that zero-length vectors are rejected\n\ncommit 2510c1b987d846821e7095fbc2c368fa51d9a15d\nMerge: a3d9e33554 e86f91c568\nAuthor: Jonathan Ellis \nDate: Mon Jun 12 17:09:22 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit a3d9e3355470f6400328cd7ac1eb07ff06f8e00b\nAuthor: Jonathan Ellis \nDate: Mon Jun 12 17:08:43 2023 -0500\n\n use ArrayList in CVV instead of HashMap since we know the keys are consecutive\n\ncommit e86f91c568f8a82e83e61bd0dd768478a7584cfc\nMerge: 77c1bc11f3 8a7a6d9c4f\nAuthor: Jeremiah D Jordan \nDate: Mon Jun 12 09:25:06 2023 -0500\n\n Merge remote-tracking branch 'datastax/ds-trunk' into vsearch\n\ncommit 77c1bc11f31e4ff001c7eff1e6fe50812e136a45\nAuthor: Jeremiah D Jordan \nDate: Mon Jun 12 09:24:56 2023 -0500\n\n Revert \"Revert this after CNDB-6974 is fixed.\"\n\n This reverts commit 8e04280312583a18085e7e7b9d31810790b039f4.\n\ncommit cae30e9879c463d80de2a2b7e4a642979803c13a\nAuthor: Jeremiah D Jordan \nDate: Mon Jun 12 09:23:39 2023 -0500\n\n Revert \"Revert this after CNDB-6974 is fixed.\"\n\n This reverts commit 67a2b7eba3c957a993726d37706643979e92a3a3.\n\ncommit 46d1a45de84f40311f193beba3ca5fc3fe7450d7\nAuthor: Jonathan Ellis \nDate: Mon Jun 12 09:03:03 2023 -0500\n\n replace our TODO entries with VSTODO to make them easier to find\n\ncommit 415b3f29e8e7f1526e28655308a1937a1175e575\nMerge: 2fd8b848be 481d29721a\nAuthor: Jonathan Ellis \nDate: Mon Jun 12 09:02:16 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit 481d29721a7993e9babd44f792be55aca925d41a\nAuthor: Zhao Yang \nDate: Mon Jun 12 21:26:43 2023 +0800\n\n Fix VectorMemtableIndex to handle max token and fix Segment#intersects (#670)\n\n * Fix VectorMemtableIndex to handle max token and min/max bound\n\n * Fix Segment#intersects to compare bound instead of token and add tests for range search\n\n * make brute force rows per query for VectorMemtableIndex\n\n * apply feedback on Segment#intersects\n\n * add comments to VectorMemtableIndex#search\n\n * Fix SegmentTest\n\ncommit 2fd8b848be82e62999e64b57cb9800d03de8e953\nAuthor: Jonathan Ellis \nDate: Mon Jun 12 07:51:18 2023 -0500\n\n clean up a couple REVIEWME\n\ncommit 3c066e07be6188c3f48c51c02cfc99b395e2a6bc\nAuthor: Zhao Yang \nDate: Sun Jun 11 14:26:42 2023 +0800\n\n fix flaky VectorDistributedTest\n\ncommit 6c719a752e90e031916dddf6cc7b7db23627bc38\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:48:01 2023 -0500\n\n fix use of bruteForceRows -- should be larger of limit,maxBruteForceRows\n\ncommit 0a2279d7f010bcb68053bb14ea2974295b21d1c4\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:46:34 2023 -0500\n\n use maxBruteForceRows when deciding whether to skip ANN\n\ncommit 1859f7355231ce7972e3d4b7af70e5d2967da516\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:27:03 2023 -0500\n\n simplify partitionKeySearchTest using euclidean distance\n\ncommit 4f40e1c2d9b1c30675eeaee7d800a8113cce97c0\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:09:24 2023 -0500\n\n typo\n\ncommit 7dbe85ac9a248aa6a0985839df1f40ceb2c08e6c\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:08:23 2023 -0500\n\n rename methods that return Bits but had bitset in their names\n\ncommit 2861bfdba8f2638a1e3f9e81df9cad3878d0e544\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:06:51 2023 -0500\n\n simplify skipANN\n\ncommit fbfe4bcdd03a66f4dedfa91ce9d06206f89827cd\nAuthor: Zhao Yang \nDate: Sat Jun 10 16:36:56 2023 +0800\n\n optimize VectorIndexSearcher#searchPosting\n - return empty posting if key range is not found in current sstable\n - return empty posting if all row ids are shadowed\n - skip ANN if matching row ids are less than limit\n\ncommit 168911b70888a3dda0ddf4f615af1b7ff54f43e3\nAuthor: Zhao Yang \nDate: Sat Jun 10 20:54:41 2023 +0800\n\n fix data visibility issue during flush: make sure SSTableAddedNotification is sent before MemtableDiscardedNotification\n\ncommit c5dd47e3ce1c7276703fa3f89a9932a3b2cd43b7\nAuthor: Zhao Yang \nDate: Sat Jun 10 12:31:53 2023 +0800\n\n fix primaryKeySearchTest and partitionKeySearchTest to use correct selector and expected results\n\ncommit 2e636529c63d0d726868d6df9b3e11d15d3871d8\nAuthor: Zhao Yang \nDate: Sun Jun 11 08:31:39 2023 +0800\n\n Vector-48: index#update is not triggered by partition/range deletion,… (#665)\n\n * Vector-48: index#update is not triggered by partition/range deletion, we have to fix VectorMemtableIndex to include shadowed primary keys\n\n - during flush, append ordinals that are removed by partition/range deletion into deletedOrdinals set\n\n * add comments to VP\n\n * revert redundant variables\n * simplify\n\n ---------\n\n Co-authored-by: Jonathan Ellis \n\ncommit daa2623b880e4b82d99204fe718db22e52ae3b42\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 13:54:38 2023 -0500\n\n use mmapped builder in OnDiskHnswGraphTest\n\ncommit 553778a22cfa8757daf3b3ec0e2cfa6fc7ba29cf\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 08:34:41 2023 -0500\n\n fix logic in partitionKeySearchTest, test still fails\n\ncommit a7f5a2f824085691da80ef62e37a430e5637c31c\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 17:57:34 2023 -0500\n\n ignore invalid vectors during build against existing data, instead of failing the build\n\ncommit 2af203646a2e57fca0a2521c31ba4269b0d8f326\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 17:43:47 2023 -0500\n\n switch from ignoring zero vectors to throwing IRE\n\ncommit 4c98fbb447a275cbf557ae0adf8f4f8a30fba17f\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 17:15:57 2023 -0500\n\n Revert \"if there are no rows in the range to search, return EmptyPostingsList instead of giving the hnsw search an empty Bits (which is a pathological case for the search)\"\n\n This reverts commit 6e5eab787a3add985dc30eca82c5dee2646f5564.\n\ncommit da99598a5d11dfa4ecac3b1eeeb9d19a4bad21e5\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 17:09:30 2023 -0500\n\n don't attempt to add zero vector to cosine indexes\n\ncommit a77792447d33600f69dd0a75a280a2b6dac51389\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 16:43:34 2023 -0500\n\n add failing partitionKeySearchTest\n\ncommit b061dd1c77f30c67324da4f94ea8b5a22d50fba7\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:52:08 2023 -0500\n\n inline the test ops\n\ncommit 6e5eab787a3add985dc30eca82c5dee2646f5564\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:51:09 2023 -0500\n\n if there are no rows in the range to search, return EmptyPostingsList instead of giving the hnsw search an empty Bits (which is a pathological case for the search)\n\ncommit 518c055dfeabacf0fa0863e40bf814058f2ad981\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:28:57 2023 -0500\n\n cleanup\n\ncommit e922b03ffe5436dc1341dbe20ee7aaca9901058f\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:27:53 2023 -0500\n\n failing tests for primary key search\n\ncommit 428e4b713e25a5820d02ca790c30e9f1477c0f44\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:21:59 2023 -0500\n\n cleanup\n\ncommit 5319bcdf3b694e861509b2a9f549f39bf51cd253\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:20:02 2023 -0500\n\n move testInvalidColumnNameWithAnn to VectorInvalidQueryTest\n\ncommit d8b64845e59c0bd3658a16df21dcc75564417e3b\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:12:40 2023 -0500\n\n upgrade lucene to reduce Integer boxing on build path\n\ncommit a62356e49c0a32f61732b8525d2bb655f46dd767\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:11:55 2023 -0500\n\n cleanup\n\ncommit b754eb7f021d10d0ae8de83ebdf2ce2f19ce59d8\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 08:29:34 2023 -0500\n\n replace CHM with NBHMLong to avoid boxing\n\ncommit 53d232ad2bcddefd72b10542f703cf36728c05f5\nAuthor: Piotr Kołaczkowski \nDate: Fri Jun 9 12:17:46 2023 +0200\n\n STAR-550 Handle SAI AbortedOperationException\n\n AbortedOperationException is thrown by SAI when index search\n hits a timeout. Now instead of allowing this exception to\n bubble up to the top and be eventually logged as error, we catch\n and swallow it in the InboundSink after creating the query response.\n Additionally now we also we set a proper error code (TIMEOUT)\n in the response, so the client has a hint on what happened.\n\ncommit 5e0c5eb3c54b58002162ce3e39d0891643e1f663\nAuthor: Matt Fleming \nDate: Fri Jun 9 08:44:10 2023 +0100\n\n CNDB-7037: Check that memtable exists before flushing and avoid IOOBE (#664)\n\n For offline services such as the compactor it's possible to not have\n live memtables. Account for this during flushing after removing indexes to\n avoid triggering an IndexOutOfBoundsException:\n\n Error happened while updating the schema\n java.lang.IndexOutOfBoundsException: Index: -1\n \tat java.base/java.util.Collections$EmptyList.get(Collections.java:4483)\n \tat org.apache.cassandra.db.lifecycle.View.getCurrentMemtable(View.java:106)\n \tat org.apache.cassandra.db.ColumnFamilyStore.forceFlush(ColumnFamilyStore.java:1115)\n \tat org.apache.cassandra.db.SystemKeyspace.forceBlockingFlush(SystemKeyspace.java:552)\n \tat org.apache.cassandra.db.SystemKeyspace.setIndexRemoved(SystemKeyspace.java:576)\n \tat org.apache.cassandra.index.SecondaryIndexManager.markIndexRemoved(SecondaryIndexManager.java:837)\n \tat org.apache.cassandra.index.SecondaryIndexManager.removeIndex(SecondaryIndexManager.java:420)\n\ncommit c241f7d41ee46c28f60d691611f32e071dac6684\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 19:19:09 2023 -0500\n\n read neighbors using .intBuffer since we know that the HNSW searcher always reads all the neighbors\n\ncommit 32f021a4f0623e57d82d70c54350eab6f0f57db7\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 18:08:30 2023 -0500\n\n VECTOR-51 leave vectors as ByteBuffers during compaction to avoid 2x memory usage incurred by keeping them around as float[] as well\n\ncommit 32d8cbbd19301228530d3d84f7a7e4120b4d4422\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 17:37:08 2023 -0500\n\n r/m node cache from query metrics since there is nothing the operator can act on there\n\ncommit 890df8e6c36c5c9e1314dbcc0a681415401120dd\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 14:15:25 2023 -0500\n\n add more information to exception when reading row offsets goes wrong\n\ncommit 5c7fe0cfed1edb10f81b134f8949dd1b5abb2781\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 13:26:17 2023 -0500\n\n revise ordinals cache as follows:\n // cache full levels including neighbors up to neighborsRamBudget, starting with the top level,\n // but always cache all levels above the bottom two levels -- this will be ~1% of the graph.\n // then on L1, cache at least the offsets\n // L0 we do not cache since we only need one extra seek (no bsearch) to read the neighbors offset\n\ncommit e131014c5cba59624e0ed9fac142abe5340e2fc0\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 13:53:10 2023 -0500\n\n add deletes test\n\ncommit 7c5a880f007535d9f03b7c7d9d3e84dcc79ef3d7\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 13:34:42 2023 -0500\n\n must use graph.size for bitset size -- not correct to use max rowid, since deleted ordinals will not have a rowid but will still be in the graph\n\ncommit 0654fbb61aec86deae69edf83fb1cf653ed7df65\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 08:44:28 2023 -0500\n\n info -> debug for validatePerIndexComponents\n\ncommit b289db2362b928f29b6c5c6289e09b33f0be1e88\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 18:08:18 2023 -0500\n\n update lucene\n\ncommit 347dae644b0d395014d8faf2affd48a6b2546e96\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 18:23:41 2023 -0500\n\n undo burntest logging config change\n\ncommit 87a6a84f38d0a0423acd9a0dda67d386876681b5\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 15:20:06 2023 -0500\n\n add debug logging\n\ncommit 20013bc865c7fb2c34111c98362aaf997dc724dc\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 14:50:39 2023 -0500\n\n add a bit more information to exception when we fail in index construction from disk\n\ncommit 86389ceae8188b8e427d8a528fada2f913f1f633\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 12:25:31 2023 -0500\n\n reduce test vector count from \"all of them\" to 200\n\ncommit 20177a1c84ba34babc82d8c1aa4e3436321b5d9f\nMerge: 2e2fce09ee f2c697ac46\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 12:21:20 2023 -0500\n\n Merge branch 'wip' into vsearch\n\ncommit f2c697ac46348301260c5947ade4ebed7dee90ee\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 12:16:28 2023 -0500\n\n multipleSegmentsMultiplePostingsTest\n\ncommit 59e4a1c53e1c19a7ebba4a0834229e61f9e4d3ce\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 12:16:04 2023 -0500\n\n cleanup\n\ncommit 2e2fce09eeb5bef4780cecf7c7b5e492a810f67a\nAuthor: Zhao Yang \nDate: Wed Jun 7 23:33:00 2023 +0800\n\n fix OnDiskOrdinalsMap to seek to segment offset before reading (#661)\n\ncommit 9502f1040229f2be6619f7e4497dc25ca5126b39\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 10:04:38 2023 -0500\n\n fix build\n\ncommit 3535444b57366fe705e0fd8c1ca095e60ed2a706\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 09:19:41 2023 -0500\n\n add write-only workload\n\ncommit a112b0965cb298f293b6e384bafc92a1d5a1fb8f\nAuthor: Zhao Yang \nDate: Wed Jun 7 21:15:53 2023 +0800\n\n VECTOR-44: improve in-memory partition-restricted query perf (#660)\n\n * VECTOR-44: improve in-memory partition-restricted query perf\n - using post-filter top-k processor instead of ANN: 14x improvement on partition-restricted query in LongVectorTest\n\ncommit cf1ec3be50ddbe40ae99dfc34eea51b9fc5c2648\nAuthor: Zhao Yang \nDate: Wed Jun 7 21:06:37 2023 +0800\n\n Vector-45: fix bitsetForShadowedPrimaryKeys to skip shadowed primary keys outside of current sstable/segment (#659)\n\n * VECTOR-45: fix bitsetForShadowedPrimaryKeys to skip shadowed primary keys outside of current sstable/segment\n\n * return empty iterator if results are empty instead of ReorderingRangeIterator\n\ncommit 87217b4f49935fbbd13c73f4d0aadb84836696a7\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 07:31:46 2023 -0500\n\n don't make areL0ShardsEnabled final, it breaks mocks\n\ncommit 2bb3a299df43561369544b662674872575975b7c\nAuthor: Matt Fleming \nDate: Wed Jun 7 10:56:08 2023 +0100\n\n VECTOR-42: Check columns exist to avoid NPE during ANN expr binding (#657)\n\n Using an non-existent column with an ANN expression triggers an NPE like\n so,\n\n java.lang.NullPointerException: null\n \tat org.apache.cassandra.cql3.ArrayLiteral.forReceiver(ArrayLiteral.java:43)\n \tat org.apache.cassandra.cql3.ArrayLiteral.prepare(ArrayLiteral.java:54)\n \tat org.apache.cassandra.cql3.Ordering$Raw$Ann.bind(Ordering.java:178)\n \tat org.apache.cassandra.cql3.Ordering$Raw.bind(Ordering.java:139)\n\n Use TM.getExistingColumn() which throws InvalidRequestException if the\n column is undefined.\n\ncommit f29d4528fc43f34889e560acafa810eee1b88ba9\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 21:18:47 2023 -0500\n\n restore query timeouts\n\ncommit 6e5734e52b41ab1e355d202ad0433440828fbb75\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 19:08:52 2023 -0500\n\n looking at performance over time in LongVectorTest\n\ncommit f96af8f0b9185163a49156bdab107801d2588bf0\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 16:00:23 2023 -0500\n\n log level = info for burn tests\n\ncommit 1247cf372927cc185be5c9767a06fdd15f102dbe\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 17:48:00 2023 -0500\n\n write the in-memory deleted ordinals to disk at the start of the postings component\n\ncommit e319d71bd6b0d265f616c807dcaf4a9d9fc38aef\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 16:36:27 2023 -0500\n\n don't allocate unnecessary objects on the happy path of no tombstones\n\ncommit 6d146a8460a17d146de756aab2ccbbf2abdb967a\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 16:19:56 2023 -0500\n\n VECTOR-23: force UCS to not shard L0\n\ncommit 79d6e177093312bc1b951d6740ad45ce8a6c5875\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 16:01:05 2023 -0500\n\n mark AutoCloseable\n\ncommit 5ec2f7d7f792bfa2dcf582ea4eed3fa8745fdaff\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 14:37:15 2023 -0500\n\n Support string literals as vectors\n\n Co-authored-by: Andrés de la Peña \n\ncommit 73a53c37536835e462e3cf17c452027c7aa591ed\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 12:25:39 2023 -0500\n\n (take 3, this time with the code changes) fix race conditions across concurrent inserts + searches in memtable\n\ncommit 51f4f419d31916a81088f171b158f1e002b5c800\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 12:25:39 2023 -0500\n\n (take 2) fix race conditions across concurrent inserts + searches in memtable\n\ncommit 94097a1f7cf1831c03e52b1f539bdc2e42fa038b\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 12:35:32 2023 -0500\n\n Revert \"fix race conditions across concurrent inserts + searches in memtable\"\n\n This reverts commit c4d41b492190fb2644f83bf4902acf71d1e4f891.\n\ncommit c4d41b492190fb2644f83bf4902acf71d1e4f891\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 12:25:39 2023 -0500\n\n fix race conditions across concurrent inserts + searches in memtable\n\ncommit 22143a8c5db160ef0fa7687ff3e8cb227de160f0\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 10:28:30 2023 -0500\n\n fix AOOB in bruteForceRows logic\n\ncommit b5624f62c88378e03260399440edf4d335ccc3af\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 10:12:53 2023 -0500\n\n call deserializeFloatArray (which gives float[]) instead of deserialize (which gives List)\n\ncommit 13ad40718737b54652c875d81ec20992b5828777\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 08:36:32 2023 -0500\n\n create CassandraHnswGraphBuilder with concurrent and serial implementations, so that single-threaded compaction doesn't have to pay the concurrency overhead\n\ncommit b83f4e4007e3d597963023b8ae32f4d0934b792d\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 08:00:44 2023 -0500\n\n add ASL header to injections.md to make CI happy\n\ncommit 16074336b1322d4bade40eb71562ca50221399dc\nMerge: 227c6c13a3 1ff6992354\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 07:48:04 2023 -0500\n\n Merge branch 'VECTOR-37' into vsearch\n\ncommit 227c6c13a3cfe42484fc560d629d54a93f8265e0\nAuthor: Jaroslaw Grabowski \nDate: Tue Jun 6 13:02:37 2023 +0200\n\n CNDB-7007 return expired tables level from getLevels\n\ncommit 312d07de8d297cdede39017354b678f2fb1b1006\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 07:31:26 2023 -0500\n\n randomize our brute force threshold, which will get the actual index scans exercised more\n\ncommit 0b897961869965629893c94250b7d4b1fb7c0f86\nAuthor: Zhao Yang \nDate: Tue Jun 6 13:44:01 2023 +0800\n\n Reference ANN sstable indexes in case of ann hybrid search (#653)\n\n * Reference ANN sstable indexes in case of ann hybrid search\n\n * simplify\n\n ---------\n\n Co-authored-by: Jonathan Ellis \n\ncommit 1ff6992354e8411f9f7c48a76f0aadb4a00f2789\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 21:55:14 2023 -0500\n\n per-query hnsw metrics\n\ncommit d3d6ae107737020eb0ca3b00c117d5bc84027dc0\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 21:16:14 2023 -0500\n\n reduce test size to prevent Jenkins OOM\n\ncommit 7765ac73fd29c9664b6f8e946c067d6f1b0a9a03\nMerge: 5a04dbcf67 c5cd09cebf\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 21:15:44 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit c5cd09cebf0c63f7bafcd2f4548395efe68b12cd\nAuthor: Zhao Yang \nDate: Tue Jun 6 09:38:42 2023 +0800\n\n Skip warning log about receiving a range that is not owned by the current replica, see VECTOR-30\n\ncommit 5a04dbcf6738d7681732e28518d6eb37f74357b8\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 11:21:15 2023 -0500\n\n add injections.md from bdp repo\n\ncommit baf9cce301a39f9506bd60c5a75d2011150f2fea\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 10:02:02 2023 -0500\n\n VECTOR-36 fix looking up indexes and SSTableContext by SSTableReader, because the SSTR will be removed from internal structures once it's compacted\n\ncommit e7fdc8454e750e2ee886c4ab27fb09abb46303fa\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 09:53:51 2023 -0500\n\n limitToTopResults should skip rows that are not in the current segment (or more preciesely, have no vectors that were indexed)\n\ncommit 2f64695a90651d9e76eee0a3399032ea241c983c\nAuthor: Zhao Yang \nDate: Mon Jun 5 14:55:20 2023 +0800\n\n Vector 6 take 2: reapply Vector-6 commit and fix NPE in SelectStatement (#648)\n\n * Revert \"Revert VECTOR-6\"\n\n This reverts commit 63a6ff609a02e4f7d9f5499f1af48ca5e58f63a8.\n\n * Vector-6 take 2:\n - fix NPE in SelectStatement by skipping reversed() for null ColumnComparator\n - fix getOrderingColumns() to return LinkedHashMap to preverse ordering columns' order\n - fix VectorLocalTest compilation\n - remove debug log in CassandraOnHeapHnsw\n\ncommit 9da51b315d207417de4f3e005616e275c62c74cb\nAuthor: Zhao Yang \nDate: Mon Jun 5 12:38:21 2023 +0800\n\n fix SAI test failures in vsearch branch (#649)\n\n - BatchlogEndpointFilterTest\n - rangeRestrictedTest\n - SegmentFlushTest\n - SegmentMergerTest\n - OperationTest\n - RangeIntersectionIteratorTest\n\ncommit dc53d7db5f0f8fe1c780f6d2b7580794f7b6b5fb\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 13:35:11 2023 -0500\n\n create views for ordinals map so we don't have to open a new Reader for each method call\n\ncommit c6385022ff2949c23e0dd0161a043cf4061f050c\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 08:19:32 2023 -0500\n\n add assert sstableContext != null\n\ncommit f340e7590db578b71201326e82006f72dd491047\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 07:37:36 2023 -0500\n\n on-disk Searcher bits should not need to be growable\n\ncommit fd5cc7f989a47d03689dd29495b974697ad1338b\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 07:37:18 2023 -0500\n\n add asserts\n\ncommit 6436e5f4ad49881a62c1442692edfedfe31968dc\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 07:21:50 2023 -0500\n\n cleanup\n\ncommit 32c8e005740d3c7ec0b1008091728e12260aecdd\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 13:57:18 2023 -0500\n\n comment\n\ncommit 091d96832ff9a68f6ea7857eafd2ec7ee0c09a0b\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 13:32:24 2023 -0500\n\n r/m unused search method returning iterators over PrimaryKey\n\ncommit 4e78a4b429b0ba6690871df9a91c8ffd8d2e53fa\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 11:25:30 2023 -0500\n\n we can use the slightly more lightweight RangeConcatIterator instead of RangeUnionIterator when combining results from different Segments\n\ncommit e7a9186bf3426ad025e809481e47db85a7bf7190\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 11:22:35 2023 -0500\n\n r/m obsolete FIXME\n\ncommit c76ee9ea0cab5b70c535de3f50b81cf333bf94a2\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 07:12:43 2023 -0500\n\n add testAppendedGraphs\n\ncommit d00bdbfb6acf9f08d2bdfa2ef06e833e567cf9eb\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 06:57:37 2023 -0500\n\n move write-to-File to test class\n\ncommit b10fd1100daf4cde0291b3b4ef06f5ceb3ca650c\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 06:46:01 2023 -0500\n\n fix Ref leaks in test\n\ncommit 0656e558bb51172df4e0cde1ebe92946d4dcec8b\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 06:26:05 2023 -0500\n\n fix tests to use View\n\ncommit 92797ced03ed26f9a5cce398504d6f0de519b899\nAuthor: Zhao Yang \nDate: Sat Jun 3 19:19:38 2023 +0800\n\n VECTOR-35: fix vector on-disk writer to append segments (#647)\n\ncommit 22f5c01860367586d0cc39def7d975a6d6bae784\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 18:54:41 2023 -0500\n\n add \"Vector indexes only support ANN queries\" check\n\ncommit bfc0b056dbacd0a1eca86020e901f545dca7670e\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 17:34:24 2023 -0500\n\n split invalid requests into separate test class\n\ncommit f66768c35920101382c1975c9d0ba0e3301b7c61\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 17:28:56 2023 -0500\n\n add specific error message when trying to do ANN without an index\n\ncommit 63a6ff609a02e4f7d9f5499f1af48ca5e58f63a8\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 12:31:22 2023 -0500\n\n Revert VECTOR-6\n\n This reverts commits:\n a37009c187edeba68389d239dc1b9f40519b1187\n 5565690fbe1056d4c159ddbe233fa22c7695320a\n e7733bb8f858a16b082b8a5c64d0322db6f6271a\n\ncommit 3d496662b259470b505d88212074c436660c27ad\nMerge: fab67bb134 4bdae7e362\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 11:05:53 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit 4bdae7e36213f5112efce085696dd24aaa0adfad\nAuthor: Mike Adamson \nDate: Fri Jun 2 14:36:01 2023 +0100\n\n Stabilise random tests using word2vec model vectors\n\ncommit 2ddc3ec6b11f6810f6c4cbaf6b62bfe46d2d1b6c\nMerge: 8e04280312 1f9179002a\nAuthor: Jeremiah D Jordan \nDate: Fri Jun 2 10:05:15 2023 -0500\n\n Merge remote-tracking branch 'datastax/ds-trunk' into vsearch\n\ncommit 8e04280312583a18085e7e7b9d31810790b039f4\nAuthor: Jeremiah D Jordan \nDate: Fri Jun 2 09:51:06 2023 -0500\n\n Revert this after CNDB-6974 is fixed.\n\ncommit fab67bb13499691f813c893cd39a3bbd406653d2\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 08:59:12 2023 -0500\n\n vector cache defaulting to 1MB per segment\n\ncommit 67a2b7eba3c957a993726d37706643979e92a3a3\nAuthor: Jeremiah D Jordan \nDate: Thu Jun 1 17:49:35 2023 -0500\n\n Revert this after CNDB-6974 is fixed.\n\ncommit 4538956b31d8c40d2e9b603b3b0a3392343e1853\nMerge: 7518680a4a af4c83aef4\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 16:17:14 2023 -0500\n\n Merge branch 'ds-trunk' into vsearch\n\ncommit 7518680a4a17cb2589cf06a9175befc10c9eab1a\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 16:14:33 2023 -0500\n\n re-use float[] across calls to vectorValue to avoid allocation overhead (credit to Jake for the idea)\n\ncommit d20bd8aedeb6232358dd16f901e2708f217b8108\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 15:35:17 2023 -0500\n\n optimize vectorValue() with direct access to the mmap-ed region\n\ncommit 24572025af5b7893279b5b03c58555b682f3abdb\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 14:33:55 2023 -0500\n\n decomposeVector does not modify the underlying buffer, so no need to duplicate() here\n\ncommit efef35d47a24222fae300d1cbd5a801e9ed79442\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 14:29:30 2023 -0500\n\n specialize deserializeFloatArray for ByteBuffer and FloatBuffer. this saves about 3% total CPU on search workloads\n\ncommit 63c27f5550d6c3e22d960b9a42b9097029e3b760\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 11:04:23 2023 -0500\n\n default target size of 5GB\n\ncommit 69c55c8d07c6e1d58b4829c153f5ebf8a4343669\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 07:20:38 2023 -0700\n\n make the on-disk hnsw code threadsafe using FileHandle.createReader (#641)\n\ncommit b194cae2e56133b946d0231768254064a47f01b6\nMerge: 7ee6422084 23c2891e7a\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 06:22:17 2023 -0500\n\n Merge branch 'ds-trunk' into vsearch\n\ncommit 7ee642208485c9e2ae62aa662dbbddc7b295bae1\nMerge: c286ec0ee2 a37009c187\nAuthor: Jonathan Ellis \nDate: Wed May 31 20:41:48 2023 -0500\n\n Merge branch 'VECTOR-6' into vsearch\n\ncommit c286ec0ee231ededded18e81357edebe72064c59\nAuthor: Jonathan Ellis \nDate: Wed May 31 20:40:21 2023 -0500\n\n add testLargeGraph\n\ncommit a37009c187edeba68389d239dc1b9f40519b1187\nAuthor: Zhao Yang \nDate: Thu Jun 1 08:50:07 2023 +0800\n\n cleanup unused code\n\ncommit 6246dbe3a970a0f4672701632003cdf576705c24\nAuthor: Jonathan Ellis \nDate: Wed May 31 16:43:42 2023 -0500\n\n comment\n\ncommit c7420440c7d3c81eac96e716148d570ae3cceb1d\nAuthor: Jonathan Ellis \nDate: Wed May 31 16:43:37 2023 -0500\n\n encapsulate shadowedPrimaryKey better\n\ncommit a1a25c33b8bec9d78354dd2c1d992444db2bf76f\nAuthor: Jonathan Ellis \nDate: Wed May 31 15:57:52 2023 -0500\n\n make hnsw cache size configurable, and make the default 128KB (roughly the size of the bloom filter for a 1GB sstable)\n\ncommit 55747d2351e915645697fa8dec404c749540456f\nAuthor: Andrés de la Peña \nDate: Wed May 31 19:28:47 2023 +0100\n\n Replace FunctionParameter.sameAsFirst by FunctionParameter.sameAs, allowing type inferrence in both directions for vector functions\n\ncommit 5565690fbe1056d4c159ddbe233fa22c7695320a\nAuthor: Jonathan Ellis \nDate: Wed May 31 12:59:42 2023 -0500\n\n cleanup and fix\n\ncommit e7733bb8f858a16b082b8a5c64d0322db6f6271a\nAuthor: Zhao Yang \nDate: Wed May 24 16:01:26 2023 +0800\n\n VECTOR-6: return vector results to client in ANN order instead of token order\n\ncommit 771067d1475c94484da178236928d2bad78e00b1\nAuthor: Mike Adamson \nDate: Wed May 31 18:33:55 2023 +0100\n\n Fix annOrderingMustHaveLimit test\n\ncommit 8dd76a73541e70da0da022dd463e6711a315f971\nAuthor: Mike Adamson \nDate: Wed May 31 18:30:49 2023 +0100\n\n Improve the validation of ORDER BY ANN OF (#639)\n\n * Improve the validation of ORDER BY ANN OF\n\n * Change to hasNonClusteredOrdering and improve limit message\n\ncommit 1ef6f2e3418db19af23f0eae24cd344d72290455\nAuthor: Jonathan Ellis \nDate: Fri May 26 18:34:25 2023 -0500\n\n add vector similarity functions\n\ncommit f4e3a39a8ece623647ba7891981e7c5ee10ed96b\nAuthor: Jonathan Ellis \nDate: Wed May 31 09:39:45 2023 -0500\n\n pull in the smallest set of changes possible from 93e0ae9a to get FunctionParameter.sameAsFirst\n\ncommit 0ef2614346c9426a966004df220221f74353de70\nAuthor: Jonathan Ellis \nDate: Wed May 31 06:22:41 2023 -0700\n\n Add support for updates + deletes (#636)\n\ncommit 5da9fefe1e18f733bb8ffd0a35ef2e1ac3cbd975\nAuthor: Jonathan Ellis \nDate: Tue May 30 16:32:02 2023 -0500\n\n rename SegmentOrdering.reorderOneComponent to limitToTopResults to align with MemtableOrdering\n\ncommit caf6eb2154ddf57c88d69c6246404b65e183057d\nAuthor: Mike Adamson \nDate: Tue May 30 22:41:29 2023 +0100\n\n Fix V1SearchableIndex.reorderOneComponent to use segments (#635)\n\ncommit 72cfa94e9389d0078d64aee60d26b8b01c6ca22f\nAuthor: Piotr Kołaczkowski \nDate: Mon May 29 20:17:44 2023 +0200\n\n VECTOR-14: Move ANN OF expressions from WHERE to ORDER BY\n\n Before:\n SELECT * FROM table WHERE ANN OF ;\n\n After:\n SELECT * FROM table ORDER BY ANN OF ;\n\ncommit 226266ef124fc220819d328ed8547c5d86626c4b\nAuthor: Mike Adamson \nDate: Tue May 30 18:58:16 2023 +0100\n\n Implement getInstance(TypeParser) for VectorType. Fix losing data at startup.\n\ncommit 2bbe88ad88e2243f12dc8df789ea9bdf208ff74b\nMerge: dd0ffb64f3 a3b8661746\nAuthor: Jonathan Ellis \nDate: Tue May 30 12:37:58 2023 -0500\n\n merge ds-trunk\n\ncommit dd0ffb64f3e54e2ce1af75da52fe3ece66a49f3f\nAuthor: Jonathan Ellis \nDate: Tue May 30 11:40:20 2023 -0500\n\n reduce startup log noise at info level\n\ncommit 1ef17101def1aacbb84f3434633a9b706291aa4b\nAuthor: Jonathan Ellis \nDate: Tue May 30 11:34:26 2023 -0500\n\n add partialUpdateTest\n\ncommit f816046a17809b06ad754a16c0f089c0d26909f9\nAuthor: Jonathan Ellis \nDate: Mon May 29 12:57:12 2023 -0500\n\n fix recall computation in test\n\ncommit 1c1442c36baafdf4e1641a9c0288dbd520cf7c0e\nAuthor: Jonathan Ellis \nDate: Mon May 29 12:34:02 2023 -0500\n\n also apply tolerance for inexact results to searchWithKey, but only for size > 10\n\ncommit 39df5112c6a551b53649ad98164f2bde93cbdc10\nMerge: d598206329 8ca4bba861\nAuthor: Jonathan Ellis \nDate: Mon May 29 12:27:24 2023 -0500\n\n merge ds-trunk\n\ncommit 8ca4bba8617e8d0fa4f48de1c479de66e3a77dd3\nAuthor: Jonathan Ellis \nDate: Mon May 29 12:19:32 2023 -0500\n\n rename MemtableIndex -> TrieMemtableIndex to make merge to vsearch easier\n\ncommit d5982063292b6a91b156da32454abacd36fa6980\nMerge: ae96e0f423 1488a5f0b9\nAuthor: Jonathan Ellis \nDate: Mon May 29 09:03:24 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit ae96e0f42393eab3547e2314f1acdb1ff766f2a6\nAuthor: Jonathan Ellis \nDate: Mon May 29 09:02:12 2023 -0500\n\n check for results within a percentage (I went with 5%) of the expected; the A in ANN means we shouldn't expect to find 100% of matches unless the graph is tiny\n\ncommit 1488a5f0b9c65473bb6a69c9e18c045d931a1fe3\nAuthor: Zhao Yang \nDate: Mon May 29 09:20:28 2023 +0800\n\n Vector-19: fix PrimaryKey min/max prefix ByteComparable and add reversed lookup for row id (#627)\n\n * Vector-19: fix PrimaryKey min/max prefix ByteComparable and add reversed lookup for row id\n\ncommit 14c41f5f913dc7f7f4cfc64b932bb6b9df475f19\nAuthor: Jonathan Ellis \nDate: Wed May 17 17:35:16 2023 -0500\n\n use incremental bytes used estimate from hnsw to avoid recomputing full ramBytesUsed on every call to add\n\ncommit f8d2aaad6e96693f532d745f71e0eaabb1ddf934\nAuthor: Jeremiah D Jordan \nDate: Fri May 26 12:02:54 2023 -0500\n\n Update cqlsh for new syntax\n\ncommit f0a882ae1171d41bc9c6ccf224ce3be80f7128e8\nAuthor: Mike Adamson \nDate: Fri May 26 16:24:38 2023 +0100\n\n Only allow VectorType to accept float\n\ncommit a29712b1365cdfac9a709e35325e85f7fa6c61c8\nAuthor: Mike Adamson \nDate: Fri May 26 16:17:51 2023 +0100\n\n Fix max term size for vectors at 16k is SSTableIndexWriter\n\ncommit 363d3f4e35c1a4df0f257557309bb042fd3f3ec7\nAuthor: Mike Adamson \nDate: Fri May 26 15:23:02 2023 +0100\n\n Merge CASSANDRA-18504 vector grammar (#628)\n\n - Now uses vector to describe vector\n - This commit does not bring in the whole 18504 patch\n only the essential grammar and type parts of the\n patch\n\ncommit 18c4e35d4ce14cda7a3c03398e16edf368ad9e6a\nMerge: 16158a5b48 b0f71ee2c3\nAuthor: Jonathan Ellis \nDate: Thu May 25 16:24:41 2023 -0500\n\n Merge branch 'VECTOR-3' into vsearch\n\ncommit b0f71ee2c371d1003cf241c3aedd7437385bcecb\nAuthor: Jonathan Ellis \nDate: Thu May 25 15:51:27 2023 -0500\n\n cleanup\n\ncommit 2d840eae74c079229729767731d3719d12ca1931\nMerge: c01cacf300 197a3207b1\nAuthor: Jonathan Ellis \nDate: Thu May 25 15:47:39 2023 -0500\n\n Merge branch 'vsearch' (early part) into VECTOR-3\n\ncommit 16158a5b48665f57bde649e87a09d078298932f5\nAuthor: Jonathan Ellis \nDate: Thu May 25 15:45:51 2023 -0500\n\n optimize ramBytesUsed\n\ncommit 197a3207b16ec97bae4924883247e2cd6a2923bd\nAuthor: Jonathan Ellis \nDate: Thu May 25 13:34:19 2023 -0500\n\n update lucene\n\ncommit a7bfcc7a6f31090c3c8443d299dabb7acd437ccd\nAuthor: Jonathan Ellis \nDate: Thu May 25 08:05:41 2023 -0500\n\n optimize ConcurrentVectorValues.write\n\ncommit 456fb08af0b922922ea0e4f24def0184cf4b32c6\nAuthor: Jonathan Ellis \nDate: Wed May 24 21:25:52 2023 -0500\n\n fix NPE better\n\ncommit c01cacf30025d181a2d86c4632042449396333bc\nAuthor: Zhao Yang \nDate: Thu May 25 08:53:38 2023 +0800\n\n Return negative ordinal if row id is not found; add failing test for null vector\n\ncommit 009fadf2ed889bcc6748f500b7fa4b1d43e910df\nAuthor: Jonathan Ellis \nDate: Wed May 24 18:00:23 2023 -0500\n\n search() errors out when an empty graph is passed to it, so special-case that\n\ncommit 8c9972d7453fad46f55e6fe5e7538442235e7fd6\nAuthor: Jonathan Ellis \nDate: Wed May 24 17:46:50 2023 -0500\n\n use serializer dimensions instead of trying to cache it from the first vector added, because we might need it before vectors are added (if someone tries to search an empty graph)\n\ncommit 8423541c4997462b65fe8f1231b0baf15dab3297\nAuthor: Jonathan Ellis \nDate: Wed May 24 17:42:30 2023 -0500\n\n fix NPEs when nulls are inserted\n\ncommit d4510e0f1934f03f66d89b7c585b3d7aac17dda2\nAuthor: Mike Adamson \nDate: Wed May 24 15:54:38 2023 +0100\n\n Rebuild HNSW graph on flush since we don't know ahead of time where UCS will want to split the range boundaries\n\ncommit 6b925531011798285a6b25d70961643092e2febe\nAuthor: Mike Adamson \nDate: Mon May 22 14:58:22 2023 +0100\n\n disable index segment compaction by default\n\ncommit 3480a80e3303c13b8c04f718f83f87489315e381\nAuthor: Jonathan Ellis \nDate: Wed May 24 14:46:03 2023 -0500\n\n move chatty logs to debug level\n\ncommit 05c5758244f6f6d4fe30fbcefbb13932f68f8891\nAuthor: Jonathan Ellis \nDate: Wed May 24 13:59:01 2023 -0500\n\n r/m fixme obsoleted by #624\n\ncommit b09db4e0425c92491e3e9c73ebdbdb84508075fd\nAuthor: Jonathan Ellis \nDate: Wed May 24 13:56:46 2023 -0500\n\n pre-size the intersection builder\n\ncommit 6d62d8c2546cda1f487c14dcde1303e14a18c804\nAuthor: Jonathan Ellis \nDate: Wed May 24 13:45:14 2023 -0500\n\n fix nested boolean expressions\n\ncommit b8301a86d37513b67545883548c5670131281dc1\nAuthor: Zhao Yang \nDate: Wed May 24 14:19:41 2023 +0800\n\n Vector-3: support partition/range restricted query\n\ncommit 8ba56fad62d21da228e3a61ed7d5971441572cf9\nAuthor: Jonathan Ellis \nDate: Wed May 24 10:35:16 2023 -0500\n\n update lucene version hash\n\ncommit a4768c583eb1a029f629e5ebda274ab2fcce3130\nAuthor: Jeremiah D Jordan \nDate: Tue May 23 10:45:02 2023 -0500\n\n Revert the revert of \"Use published lucene fork jar.\"\"\n\n This reverts commit 7a8ff12860a97d451045c9dd1978a3599a0f4679.\n\ncommit 063a176caeec0f1e298d044490aa29ecdd468811\nAuthor: Zhao Yang \nDate: Wed May 24 22:03:32 2023 +0800\n\n Union the results from multiple segments per index (#624)\n\n - Union the results from multiple segments per index\n - Revert intersection behavior to pick 2 most selective indexes if there is no ANN\n - Fix SSTableRowIdPostingList to return END_OF_STREAM if next row is END_OF_STREAM\n - Skip TermTree for vector index\n - Fixes:\n - VectorMemtableIndexTest#randomQueryTest\n - SegmentMergerTest\n - SingleNodeQueryFailureTest#testFailedRangeIteratorOnMultiIndexesQuery\n - SelectiveIntersectionTest\n - QueryTimeoutTest\n\ncommit 643f34c2b654b80573ba92381e386b4286f7fb63\nAuthor: Jonathan Ellis \nDate: Wed May 24 08:25:35 2023 -0500\n\n switch all the hnsw internals to use float[] so we don't need to keep both representations around\n\ncommit 7a8ff12860a97d451045c9dd1978a3599a0f4679\nAuthor: Jonathan Ellis \nDate: Wed May 24 08:06:28 2023 -0500\n\n Revert \"Use published lucene fork jar.\"\n\n This reverts commit 53c6c90920dd2a48fc3334a6cddb215d3f4ebfcd.\n\ncommit 8731eefe4122a5af0b3723a3b07884a201616e7a\nMerge: e50d5e955b 53c6c90920\nAuthor: Jonathan Ellis \nDate: Wed May 24 08:05:18 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit e50d5e955bfde6357a5bb892b2ea70d46e8cc77d\nAuthor: Jonathan Ellis \nDate: Wed May 24 08:02:52 2023 -0500\n\n cache float[] from ByteBuffer\n\ncommit 53c6c90920dd2a48fc3334a6cddb215d3f4ebfcd\nAuthor: Jeremiah D Jordan \nDate: Tue May 23 18:29:00 2023 -0500\n\n Use published lucene fork jar.\n\ncommit f2567cd2fb2a683e7c0fee23945baac3ceca6ad4\nAuthor: Jonathan Ellis \nDate: Tue May 23 14:56:25 2023 -0500\n\n attempting to add back IN support but with limited success, even \"basicOrTest\" gets parsed as child nodes\n\ncommit 9e1f018f5c0d5832446fc6a0338aaf95c8323303\nAuthor: Jonathan Ellis \nDate: Tue May 23 14:21:23 2023 -0500\n\n reduce log level of some of the chattiest locations\n\ncommit efaf501fe3158da42dcd874b06014acb127f5cd9\nAuthor: Jonathan Ellis \nDate: Tue May 23 13:55:40 2023 -0500\n\n clean up RangeIntersection Builder overloads\n\ncommit e7127d9d32392347f7010de6267521742f831217\nAuthor: Jonathan Ellis \nDate: Tue May 23 13:46:27 2023 -0500\n\n complete generification of RangeIterator in src/ (but tests are still incomplete)\n\ncommit c07ded6420aa25b9dfa2b3c0785178b08d286b26\nAuthor: Jonathan Ellis \nDate: Tue May 23 13:04:16 2023 -0500\n\n TermIterator is more or less replaced by CheckpointingIterator\n\ncommit 14d03ae99c69c2419dc0cca3291b4f5eaefbc054\nAuthor: Jonathan Ellis \nDate: Tue May 23 14:45:18 2023 -0500\n\n update lucene jar\n\ncommit a15b964d56dcba285c6871d354755d7aa489af1d\nAuthor: Jonathan Ellis \nDate: Tue May 23 12:48:41 2023 -0500\n\n disk and memory recalls should be the same (since disk graph should be exactly equivalent to in-memory)\n\ncommit f03680bdf8581fd13a2d772e8265e3c8a713d247\nMerge: 868a6b8f11 fd1293dbaf\nAuthor: Jonathan Ellis \nDate: Tue May 23 12:08:46 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit 868a6b8f11a25c20b7e59a93d65cc7e2b021a432\nAuthor: Jonathan Ellis \nDate: Tue May 23 12:04:11 2023 -0500\n\n use brute force if we expect to perform fewer comparisons that way than with a graph search\n\ncommit 2ed32bf53acaf8180130bef1aa5cebd9b6d32d13\nAuthor: Jonathan Ellis \nDate: Tue May 23 11:59:11 2023 -0500\n\n refactor AnnKeyRangeIterator -> ReorderingRangeIterator\n\ncommit d28bad9b2df3fc84b117f9ed0a04314ecb9c0a02\nAuthor: Jonathan Ellis \nDate: Tue May 23 11:55:15 2023 -0500\n\n rename reorderOneComponent to limitToTopResults\n\ncommit fd1293dbaf82f054593b3150249ac00110b286cf\nAuthor: Jeremiah D Jordan \nDate: Tue May 23 10:45:02 2023 -0500\n\n Update python driver to one with supports the Vector type\n\ncommit 6603f7601671357928948dfb0921b0c731bb96d3\nAuthor: Jonathan Ellis \nDate: Tue May 23 11:06:12 2023 -0500\n\n extract ReorderingPostingList\n\ncommit 94b4ff20ccc3a51e966b4798f5364a2d9ed4b966\nAuthor: Jonathan Ellis \nDate: Tue May 23 10:29:33 2023 -0500\n\n deserialize doesn't modify buffer so duplicate is unnecessary\n\ncommit 840053d212985a109cb5242c19e8a5876fdec7b3\nAuthor: Jonathan Ellis \nDate: Tue May 23 10:26:57 2023 -0500\n\n use LongHeap instead of generic PQ\n\ncommit aac90dc3e0fdb745e465f044f2624228e705be10\nAuthor: Jonathan Ellis \nDate: Tue May 23 09:47:22 2023 -0500\n\n r/m unused defer flag from toPrimaryKeyIterator / toSSTableRowIdsIterator\n\ncommit 1fd71c1cb7f2e785237b36c6e42842fb326d2380\nAuthor: Jonathan Ellis \nDate: Tue May 23 09:26:03 2023 -0500\n\n remove unused method marked \"to remove\"\n\ncommit b96c5a9efccf3226acf2f2a08ca149d0acc0b21d\nMerge: 54f41e32d3 ba8aa07dd0\nAuthor: Jonathan Ellis \nDate: Tue May 23 09:24:44 2023 -0500\n\n Merge remote-tracking branch 'zhao/vsearch-row-id-iterator-for-reordering' into vsearch\n\ncommit 54f41e32d3359941ddf70d34fc8e68a842b9091d\nAuthor: Jonathan Ellis \nDate: Tue May 23 08:14:10 2023 -0500\n\n update lucene to fix NPE in ram usage\n\ncommit ba8aa07dd09f52a72e5dffccb950a21ec37a1c11\nAuthor: Zhao Yang \nDate: Tue May 23 16:51:36 2023 +0800\n\n intersect on row-ids before fetching primary key for multiple non-ann indexes or single index\n\ncommit 2a69547167f8dc1b40e8a26b71cdf43a45779280\nAuthor: Zhao Yang \nDate: Tue May 23 09:44:00 2023 +0800\n\n Use sstable-row-id iterator to re-order ann index\n\n - add searchSSTableRowIds to searcher\n\ncommit 04b9a606d9a1aaeeeab3ec0981a5f1300cc2a423\nAuthor: Zhao Yang \nDate: Tue May 23 00:30:12 2023 +0800\n\n Make RangeIterator generic\n\ncommit 4ff716a7411d3ad80009d4fb97bf53af51dd5eba\nAuthor: Jonathan Ellis \nDate: Mon May 22 14:57:34 2023 -0500\n\n r/m obsolete FIXMEs\n\ncommit 175ad0edce9cef3fb29f230fb64fb1913328d101\nAuthor: Zhao Yang \nDate: Tue May 23 00:30:12 2023 +0800\n\n fix vector distributed test: use 1 token and ignore multi-ann-index test (#622)\n\ncommit 78adf77d9e6b7f20ece71fe665a5d1aec51f79b6\nAuthor: Mike Adamson \nDate: Mon May 22 14:26:48 2023 +0100\n\n Copy lucene snapshot to build folder in build-resolver.xml\n\ncommit 5172315dcb51e05a44af224ddfe944bca29005d0\nAuthor: Jonathan Ellis \nDate: Mon May 22 08:17:41 2023 -0500\n\n declare lucene dependency to come from lib/\n\ncommit d4ee45b8675f79b4a1e06f97d0ce618ab08f030a\nAuthor: Jonathan Ellis \nDate: Sun May 21 05:40:30 2023 -0500\n\n add caching and test code\n\ncommit 5102d96e2fb3460f1e541cc1b481e224d112eac1\nMerge: 6afcfabe88 87d39a7b25\nAuthor: Jonathan Ellis \nDate: Sat May 20 11:15:14 2023 -0500\n\n merge cep-vsearch\n\ncommit 6afcfabe883c5627c96eda8eb488c51c06e17c91\nMerge: a993dbe49a fa85a191c5\nAuthor: Jonathan Ellis \nDate: Sat May 20 09:15:13 2023 -0500\n\n Merge commit 'fa85a191' into vsearch\n\ncommit a993dbe49a75dde4b9d5528e0816b81ae427a1dd\nAuthor: Jonathan Ellis \nDate: Fri May 19 22:00:31 2023 -0700\n\n FIXME not sure why version is messed up post merge, this hacks around it\n\ncommit aadf168e47a4271cc3faeb62ec7044cbfe193893\nMerge: c217120935 5bc4d4b42e\nAuthor: Jonathan Ellis \nDate: Sat May 20 08:52:48 2023 -0500\n\n merge 5bc4d4b42e\n\ncommit c2171209350cc307f3b5fcc73a6ec09423d203df\nAuthor: Jonathan Ellis \nDate: Fri May 19 22:00:31 2023 -0700\n\n FIXME not sure why version is messed up post merge, this hacks around it\n\ncommit 2e44491d3db6a881324126490e6e1511d6082f39\nMerge: 68f85f7009 0ff4566080\nAuthor: Jonathan Ellis \nDate: Sat May 20 08:15:16 2023 -0500\n\n merge 0ff4566080\n\ncommit 68f85f700995b274ec0d64654ed2692c7fc8bf61\nMerge: c8843779be 2ae60e7411\nAuthor: Jonathan Ellis \nDate: Sat May 20 07:50:16 2023 -0500\n\n Merge commit '2ae60e74' into vsearch\n\ncommit c8843779be9fe0c3edd047754bb08f633cff5646\nMerge: 3cb57548bb 93616d080c\nAuthor: Jonathan Ellis \nDate: Sat May 20 07:43:18 2023 -0500\n\n Merge commit '93616d080c539e880e37cd22fe5f27396a7a2594' into vsearch\n\ncommit 87d39a7b252b3f3cc1f4aae0f149085f6bc83abc\nAuthor: Jonathan Ellis \nDate: Fri May 19 17:24:55 2023 -0700\n\n fix intersection count, and add comments to KRI api\n\ncommit 4fdc58fced9ca350333675cb19a42169bebeaccf\nAuthor: Jonathan Ellis \nDate: Fri May 19 15:27:34 2023 -0700\n\n test recall\n\ncommit 26fb1006430513a37f267a4dfb81ef473516ac7b\nAuthor: Jonathan Ellis \nDate: Fri May 19 14:19:08 2023 -0700\n\n add tests\n\ncommit 1cbf78bbf71fff2d597ddd199fd3a758313c7bf4\nAuthor: Jonathan Ellis \nDate: Fri May 19 14:02:36 2023 -0700\n\n comments\n\ncommit 65e41490f041535f1a48a726032e28f6c8febcff\nAuthor: Jonathan Ellis \nDate: Fri May 19 13:57:04 2023 -0700\n\n rm AnnResult\n\ncommit 06b10a28c3f6f19c76b056526cdabb4c9aff1528\nAuthor: Jonathan Ellis \nDate: Fri May 19 13:54:08 2023 -0700\n\n cleanup commented-out code\n\ncommit e33236ab8b9da7a025f4c59406c52ecbc3b5ae4f\nAuthor: Jonathan Ellis \nDate: Fri May 19 09:52:50 2023 -0700\n\n convert suspicious one-off binary search in OnDiskHnswGraph to use DiskBinarySearch\n\ncommit fbd30bce13e87c63b129db26c5c123ae9bf8f9ed\nAuthor: Jonathan Ellis \nDate: Fri May 19 09:31:16 2023 -0700\n\n perform disk-based binary search for ordinals\n\ncommit 77bb1b5cfcc743224bdb01d21563365caa184c21\nAuthor: Jonathan Ellis \nDate: Fri May 19 08:22:36 2023 -0700\n\n move OnDiskOrdinalsMap to its own class, and move write() code into it\n\ncommit 8a1fd9f0a38f4c16b72d5bc7b2aeb3ee07bacc7a\nAuthor: Jonathan Ellis \nDate: Thu May 18 21:35:28 2023 -0700\n\n add ordinal to row mapping\n\ncommit db32d9c7b077133a213ac3a0da161f163c4310a1\nAuthor: Jonathan Ellis \nDate: Wed May 17 20:43:49 2023 -0500\n\n mostly-working reordering\n\ncommit 62ba004d79aa7511f519ebb56fa8af2cb424e21e\nAuthor: Jonathan Ellis \nDate: Wed May 17 20:13:10 2023 -0500\n\n turn QueryView inside out; perform intersections first for each sstable, then union the results\n\ncommit d65039a6ccd8364e8eb8ef31933cdf8dca298437\nAuthor: Jonathan Ellis \nDate: Wed May 17 18:37:07 2023 -0500\n\n update MIM to give individual iterators back\n\ncommit a190f1e917b93b6cabd15347dec55b1683807b1f\nAuthor: Jonathan Ellis \nDate: Wed May 17 18:59:13 2023 -0500\n\n refactor CheckpointingIterator to just take an iterator to wrap, and index references to close if something goes wrong\n\ncommit 307fe1c20b400e17048a356f73822ef51f66f341\nAuthor: Jonathan Ellis \nDate: Wed May 17 18:47:20 2023 -0500\n\n rename IndexSearchResultIterator -> CheckpointingIterator\n\ncommit 690751b216afbe3e77ed0f89ce04d16e5c6e985d\nMerge: 4b6d0f2fb5 fa85a191c5\nAuthor: Jonathan Ellis \nDate: Wed May 17 10:17:49 2023 -0500\n\n Merge branch 'cep-7-sai' into cep-vsearch\n\ncommit 4b6d0f2fb531f6832d2b24465b0982683e741dfe\nAuthor: Zhao Yang \nDate: Wed May 17 16:45:42 2023 +0800\n\n Add vector top-k filter at replica side before returning to coordinator (#618)\n\n * added QueryPlan#postIndexQueryProcessor to filter top-k at replica side before sending response to coordinator\n\ncommit 5bc4d4b42e77180df050bdbd9f5fd18fbb37332c\nAuthor: Jonathan Ellis \nDate: Tue May 16 07:38:45 2023 -0500\n\n failing test was failing b/c test was wrong\n\ncommit f37e40f5c5d99aeaafb33f41be0bd10bff6b5756\nMerge: ae1bffa49b 1c41a095e1\nAuthor: Jonathan Ellis \nDate: Mon May 15 21:24:47 2023 -0500\n\n merge\n\ncommit ae1bffa49b4c8757a3d9110ae065b8cb70d8e416\nMerge: bbc4f72b8d 381e04aaa1\nAuthor: Jonathan Ellis \nDate: Mon May 15 21:22:16 2023 -0500\n\n Merge commit '381e04aaa132cf7a255efad3f792852d9ec1729d' into cep-vsearch\n\ncommit 1c41a095e130bce4c7c215650f9ef3f817ae635e\nAuthor: Jonathan Ellis \nDate: Mon May 15 21:11:47 2023 -0500\n\n fix computeNext to return endOfData when done\n\ncommit b845395ad7e5f4670597cd171f28b98c941d03a7\nAuthor: Jonathan Ellis \nDate: Mon May 15 21:11:15 2023 -0500\n\n cleanup\n\ncommit 58df11dad894aa5fb36bfbdaaf4c7dc1987fcc94\nAuthor: Jonathan Ellis \nDate: Mon May 15 20:59:12 2023 -0500\n\n failing tests\n\ncommit ea5977620ebb07d220ad51800c9b90f470494079\nAuthor: Jonathan Ellis \nDate: Mon May 15 20:58:52 2023 -0500\n\n FIXME temporarily remove checkstyle\n\ncommit 715790922379ad7b5dbdea5d3358f24542762889\nAuthor: Jonathan Ellis \nDate: Mon May 15 20:49:10 2023 -0500\n\n cleanup\n\ncommit 19e47d363c856f81bda3323be02bfe18f60b51c0\nAuthor: Jonathan Ellis \nDate: Mon May 15 17:55:52 2023 -0500\n\n update NeighborSet usage\n\ncommit 920abafce00c94bbc7d6b0c86ba78a2a9d3f9540\nAuthor: Jonathan Ellis \nDate: Mon May 15 17:53:43 2023 -0500\n\n cleanup\n\ncommit d58db09004427506efe03e81364cadf44fc914c4\nAuthor: Jonathan Ellis \nDate: Mon May 15 15:46:06 2023 -0500\n\n pull in latest hnsw optimizations\n\ncommit bbc4f72b8d7d99db003103103c90d78db4ee04d1\nAuthor: Jonathan Ellis \nDate: Mon May 15 10:02:40 2023 -0500\n\n use IndexOutputWriter in other components\n\ncommit 381e04aaa132cf7a255efad3f792852d9ec1729d\nAuthor: Jonathan Ellis \nDate: Mon May 15 09:32:25 2023 -0500\n\n switch other components to also use IndexOutputWriter\n\ncommit 5951690f5472502bcb093179e8905212c1fcf815\nAuthor: Mike Adamson \nDate: Mon May 15 14:48:16 2023 +0100\n\n Fix CassandraHnswGraphWriter to use correct output writer (#616)\n\ncommit bc5b956607fa77d53fcab9ad338776af69610feb\nAuthor: Zhao Yang \nDate: Mon May 15 17:35:42 2023 +0800\n\n Distributed vector search (#613)\n\n * distributed vector search\n\n - query all replicas selected by consistency level at once with full request range\n\n - filter top-K results at coordinator in QueryPlan#postProcessor\n\n - skip short-read-protection, read repair and replica filtering protection because replica response will be top-k\n\n - fail ANN query without limit or limit exceeding MAX_TOP_K\n\n - make vector search max_top_k configurable and default to 1k\n\ncommit 6d8c94535354f162f60099deecb0523f9f71bb99\nAuthor: Jonathan Ellis \nDate: Sun May 14 10:01:08 2023 -0500\n\n fix imports for checkstyle\n\ncommit 68c20dc9ceeaf0c2169718aa4a1dce9c0f92f003\nAuthor: Jonathan Ellis \nDate: Sat May 13 19:23:58 2023 -0500\n\n add the index and ordinal mapping to the set of components that the system knows about\n\ncommit ba7efaac142c8b3181881f86020f659837c90463\nAuthor: Jonathan Ellis \nDate: Sat May 13 17:57:22 2023 -0500\n\n cleanup\n\ncommit 5ae06fe31840e5ba1899086f3b8eb8194f825db7\nAuthor: Jonathan Ellis \nDate: Sat May 13 17:23:30 2023 -0500\n\n update to latest lucene snapshot that addresses all known concurrency bugs\n\ncommit f996edd63f66cf7b688f8e014b31f15450dd1d3d\nAuthor: Jonathan Ellis \nDate: Fri May 12 16:08:40 2023 -0500\n\n fix for cqlsh by Bret McGuire\n\ncommit fa85a191c5e0bd508da584412648308888769cb9\nAuthor: Andrés de la Peña \nDate: Mon Apr 24 16:39:26 2023 +0100\n\n Allow CQL queries on multiple indexes without ALLOW FILTERING\n\n patch by Andrés de la Peña; reviewed by Berenguer Blasi for CASSANDRA-18217\n\ncommit e2f3d2150ab2bc5e038f2661b8e2337d3b5cd4bf\nAuthor: Jonathan Ellis \nDate: Thu May 11 14:34:22 2023 -0500\n\n imports\n\ncommit 68e59bc715de6d1f69d69530310655a598a54200\nAuthor: Mike Adamson \nDate: Thu May 11 17:20:52 2023 +0100\n\n Use correct bind types for vector\n\ncommit c57119254b7e065433804e36ed6ead71d53984ea\nAuthor: Jonathan Ellis \nDate: Wed May 10 23:06:19 2023 -0500\n\n avoid calling neighborSet.size() in write path\n\ncommit f91e9ea40fa4bfc07c0bb280b7c479d473956da3\nAuthor: Jonathan Ellis \nDate: Wed May 10 23:05:11 2023 -0500\n\n add more asserts. countNeighbors is failing\n\ncommit 1bd6789304a60762f0ef497c2137dd96f806a262\nAuthor: Jonathan Ellis \nDate: Wed May 10 22:40:01 2023 -0500\n\n lock out updates to the graph while we're writing it to disk\n\ncommit da7cb2984f96c06a5890ed2da21c42d946d7c754\nAuthor: Jonathan Ellis \nDate: Wed May 10 22:06:58 2023 -0500\n\n imports\n\ncommit 3676a86b821c6c05d9c5df33f4e4fbc9eeb1f591\nAuthor: Jonathan Ellis \nDate: Wed May 10 22:03:14 2023 -0500\n\n compact by building graph in memory, like it did before\n\ncommit fa1ebd013bc03def3d211932fe2a98d1bb4442d9\nAuthor: Jonathan Ellis \nDate: Wed May 10 16:15:58 2023 -0500\n\n rewrite save-to-disk without Lucene code, to support multiple rows having the same vector value\n\n bonus: we don't need to rebuild an index that we already have in memory\n\ncommit 3b328324d7c3afcda34a95c4b9fcbb9fe017384c\nAuthor: Jonathan Ellis \nDate: Wed May 10 17:00:02 2023 -0500\n\n upgrade to latest lucene snapshot\n\ncommit 1f4e768e9ec3327a56d143fa17c86…","shortMessageHtmlLink":"Squashed commit of the following:"}},{"before":"0bd0bfd2f8a8d0578a5e078122a20b6507b85c1b","after":"9a2f6ac4d0b0e1443a85ba1fa39062a8909de700","ref":"refs/heads/CNDB-9099-rebase-ds-trunk","pushedAt":"2024-05-17T21:42:09.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"djatnieks","name":"dan jatnieks","path":"/djatnieks","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2855940?s=80&v=4"},"commit":{"message":"Fix the remaining issues with ZCS","shortMessageHtmlLink":"Fix the remaining issues with ZCS"}},{"before":"9b98eb1b306f3f222ce797e8bb2c67b5ce3e9fc7","after":null,"ref":"refs/heads/jvector-300beta10","pushedAt":"2024-05-17T17:08:02.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"jkni","name":"Joel Knighton","path":"/jkni","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5061375?s=80&v=4"}},{"before":"478ed37f6322128b5837ec826096bd35d56be25e","after":"46f2b46cd08355f8e6cb8e098d845e6712bee414","ref":"refs/heads/vsearch","pushedAt":"2024-05-17T17:08:02.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"jkni","name":"Joel Knighton","path":"/jkni","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5061375?s=80&v=4"},"commit":{"message":"Pick up bug fix affecting larger-than-memory vector index construction in JVector 3.0.0-beta.10 (#1124)","shortMessageHtmlLink":"Pick up bug fix affecting larger-than-memory vector index constructio…"}},{"before":null,"after":"9b98eb1b306f3f222ce797e8bb2c67b5ce3e9fc7","ref":"refs/heads/jvector-300beta10","pushedAt":"2024-05-17T16:07:23.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jkni","name":"Joel Knighton","path":"/jkni","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5061375?s=80&v=4"},"commit":{"message":"Pick up bug fix affecting larger-than-memory vector index construction in JVector 3.0.0-beta.10","shortMessageHtmlLink":"Pick up bug fix affecting larger-than-memory vector index constructio…"}},{"before":null,"after":"571de1cb4fdc565386187d27ac950639c059a9e4","ref":"refs/heads/fix-frozen-types-syntax-errors","pushedAt":"2024-05-17T16:01:32.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"adelapena","name":"Andrés de la Peña","path":"/adelapena","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5229055?s=80&v=4"},"commit":{"message":"Fix management of syntax errors in frozen types","shortMessageHtmlLink":"Fix management of syntax errors in frozen types"}},{"before":null,"after":"c56657fc4d8a649764040997a58814e0a691b9ea","ref":"refs/heads/prevent-double-close-jl","pushedAt":"2024-05-17T14:46:08.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jacek-lewandowski","name":"Jacek Lewandowski","path":"/jacek-lewandowski","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6516951?s=80&v=4"},"commit":{"message":"add a test and eliminate a leak in case of a failure","shortMessageHtmlLink":"add a test and eliminate a leak in case of a failure"}},{"before":"cc528e8642c18f683b64edcff4f4df6e29ef878d","after":"7193858911b4f75646ce2ffb3865cec8b9e420cf","ref":"refs/heads/todo-cleanup","pushedAt":"2024-05-17T14:32:33.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jbellis","name":"Jonathan Ellis","path":"/jbellis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/42158?s=80&v=4"},"commit":{"message":"add explicit cleanup() to force connected-ness\n\nexecute queries both with and without brute force enabled, highlighting intermittent failure [before the fix] on non-brute-force path","shortMessageHtmlLink":"add explicit cleanup() to force connected-ness"}},{"before":"6e1bd6a26d099abc50fbb54486c5f949a610dcd3","after":"cc528e8642c18f683b64edcff4f4df6e29ef878d","ref":"refs/heads/todo-cleanup","pushedAt":"2024-05-17T14:04:41.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jbellis","name":"Jonathan Ellis","path":"/jbellis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/42158?s=80&v=4"},"commit":{"message":"add explicit cleanup() to force connected-ness\n\nexecute queries both with and without brute force enabled, highlighting intermittent failure [before the fix] on non-brute-force path","shortMessageHtmlLink":"add explicit cleanup() to force connected-ness"}},{"before":"59f774d6e76d281050100005dc159281390fc911","after":"6e1bd6a26d099abc50fbb54486c5f949a610dcd3","ref":"refs/heads/todo-cleanup","pushedAt":"2024-05-16T21:37:33.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jbellis","name":"Jonathan Ellis","path":"/jbellis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/42158?s=80&v=4"},"commit":{"message":"add explicit cleanup() to force connected-ness\n\nexecute queries both with and without brute force enabled, highlighting intermittent failure [before the fix] on non-brute-force path","shortMessageHtmlLink":"add explicit cleanup() to force connected-ness"}},{"before":null,"after":"88150881f0e982c605be8de9503909566a889add","ref":"refs/heads/CNDB-9099-rebase-ds-trunk-vsearch","pushedAt":"2024-05-16T21:21:00.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"djatnieks","name":"dan jatnieks","path":"/djatnieks","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2855940?s=80&v=4"},"commit":{"message":"Squashed commit of the following:\n\ncommit 4c4c1ff802c7667e1289a4e730cc34bf6b3bd007\nAuthor: Jonathan Ellis \nDate: Wed Jul 5 12:03:22 2023 -0500\n\n add test for creating index after data is added\n\ncommit 0acaae364c19ed7556b66b5576762ecc131af5fd\nAuthor: Jeremiah D Jordan \nDate: Wed Jul 5 09:23:01 2023 -0500\n\n Revert \"enable segment compaction by default for vector indexes -- we should prioritize read performance\"\n\n This reverts commit a4fa0728337ebf236ad03f6ad708cf51c3dc08c5.\n\ncommit e3ddb6426de4e89b5fe61ed742b69257865c2d3c\nAuthor: Jonathan Ellis \nDate: Wed Jun 28 10:54:49 2023 -0500\n\n VECTOR-64 set sstable_growth to 0.5 by default\n\ncommit 3ca75046e6802ce5836a654459b01948c83d1d7b\nMerge: a4fa072833 10a2a31ae8\nAuthor: Jonathan Ellis \nDate: Wed Jun 28 10:54:18 2023 -0500\n\n merge ds-trunk\n\ncommit a4fa0728337ebf236ad03f6ad708cf51c3dc08c5\nAuthor: Jonathan Ellis \nDate: Wed Jun 28 10:24:41 2023 -0500\n\n enable segment compaction by default for vector indexes -- we should prioritize read performance\n\ncommit 08153b38549d700505d064c25d7a2070c2ff07aa\nMerge: 6a063b4f84 1a45fdc5f9\nAuthor: Jonathan Ellis \nDate: Wed Jun 28 10:18:38 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit 6a063b4f8458f119d4b99bd6b78cbc8832f27f13\nAuthor: Jonathan Ellis \nDate: Wed Jun 28 10:13:51 2023 -0500\n\n VECTOR-63 add cassandra.sai.hnsw.allow_custom_parameters defaulting to false to control setting maximum_node_connections and construction_beam_width, which can cause memory usage to explode if used incorrectly\n\ncommit 1a45fdc5f92e9c82dba6e9e56a5e341806070b13\nAuthor: Jeremiah D Jordan \nDate: Mon Jun 26 15:57:10 2023 -0500\n\n Add the option to configure the file_cache_size_in_mb in a -D like it is in 6.8-cndb.\n\ncommit d2e45dab5b57f20fbdd6250d6dc186c83d783ed3\nAuthor: Jonathan Ellis \nDate: Thu Jun 22 16:13:00 2023 -0500\n\n VECTOR-62 don't flush a graph that only contains deleted vectors\n\ncommit 4a9afe205c20b1f863e89d4e02138557d225c3fc\nAuthor: Jonathan Ellis \nDate: Thu Jun 22 16:11:11 2023 -0500\n\n r/m obsolete comment\n\ncommit a925dc22696bcbebcad17a460441e8b84ff97660\nMerge: 7c7d160dd9 2b22984fe1\nAuthor: Jeremiah D Jordan \nDate: Thu Jun 22 09:03:08 2023 -0500\n\n Merge remote-tracking branch 'datastax/ds-trunk' into vsearch\n\ncommit 7c7d160dd90624b0e365287ef0a2f0b7ef94da53\nMerge: a14b387372 7a5e374cea\nAuthor: Jonathan Ellis \nDate: Tue Jun 20 21:42:51 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit a14b387372ef3f220ece68b019acaafaf0d56f42\nAuthor: Jonathan Ellis \nDate: Tue Jun 20 17:52:13 2023 -0500\n\n move new options to jvm17-server.options\n\ncommit dc67582585a90628e026a693497793b7d8a298ed\nAuthor: Jonathan Ellis \nDate: Tue Jun 20 17:47:27 2023 -0500\n\n copy jvm11 options to jvm17\n\ncommit b6717410f925486a858c4c9414e417eed50950d7\nAuthor: Jonathan Ellis \nDate: Tue Jun 20 17:30:48 2023 -0500\n\n enable simd\n\ncommit 92b9e6e2f136254fb1f8e50732809ea529617f48\nAuthor: Jonathan Ellis \nDate: Mon Jun 19 15:31:09 2023 -0500\n\n make it run under jdk 20\n\ncommit 3e985953637819ac8504955552ce0feba29bdd40\nAuthor: Jonathan Ellis \nDate: Tue Jun 20 17:13:56 2023 -0500\n\n update to lucene with simd\n\ncommit 7a5e374cea785678670444121ad36be6170dd59b\nMerge: 5e5ae4de82 e9ddd5f0d9\nAuthor: Jeremiah D Jordan \nDate: Tue Jun 20 15:00:24 2023 -0500\n\n Merge remote-tracking branch 'datastax/ds-trunk' into vsearch\n\ncommit 5e5ae4de824a7e838bbdfdbabc92e8b256911e3f\nAuthor: Jonathan Ellis \nDate: Mon Jun 19 10:57:18 2023 -0500\n\n we weren't calculating length correctly (supposed to match ordinals range) but it didn't matter b/c length() is not called on the search path where these classes are used. make this explicit by throwing in length implementation.\n\ncommit c0d4055a0bc6914d0220fbd4912d6355e9f6a1d9\nMerge: f2c9a7cb59 c605f8c9f0\nAuthor: Jonathan Ellis \nDate: Mon Jun 19 10:20:02 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit f2c9a7cb59b113a349992011b739079106274c1e\nAuthor: Jonathan Ellis \nDate: Mon Jun 19 10:19:34 2023 -0500\n\n VECTOR-61 remove deletedOrdinals collection that was not threadsafe wrt add and remove operations; create postingsByOrdinal map that allows Bits operations to look up the postings, instead\n\ncommit 9744a13b405cbeb1b68894bb08cc24796aedd8f7\nAuthor: Jonathan Ellis \nDate: Mon Jun 19 09:13:24 2023 -0500\n\n add LVT.testMultiplePostings (works fine)\n\ncommit c605f8c9f037e724544b21fe2aadcf0d8503d86f\nAuthor: Zhao Yang \nDate: Sat Jun 17 12:44:36 2023 +0800\n\n skip replica-filtering-protection for ANN requests\n\ncommit 71935bd07c9218ba70f1ffde7b6db1e88009e529\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 20:20:06 2023 -0500\n\n forgot that maxBruteForceRows has to stay non-final for Byteman\n\ncommit b264afc07727305ce88ab2f18b8a2643d2510bd3\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 20:10:19 2023 -0500\n\n VECTOR-54\n - validateIndexable also checks for very large vectors (that explode to NaN when we take the cosine)\n - check query vectors with the same criteria as vectors to index\n\ncommit 66c31a6dcfcbf81b7657ac9c4119803e026d150a\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 20:07:41 2023 -0500\n\n if CFS.apply catches an InvalidRequestException, keep the same type when we re-throw so it gets back to the client as intended\n\ncommit e0bdfdbe421a1cca30746604cda9ef5b992002bb\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 17:33:37 2023 -0500\n\n add [failing] emptyIndexTest\n\ncommit 75cba96f90730b32ea4a39375c72aaff4ffd6221\nAuthor: Jonathan Ellis \nDate: Thu Jun 15 16:52:46 2023 -0500\n\n re-use bitsets across searches\n\ncommit 2af2ce8a321f7fea268d81c393da7f8f6d886776\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 17:05:52 2023 -0500\n\n more asserts that graph is in sane state when we write it\n\ncommit a63492686d13e12b799acdf50c6f0c20dd680f21\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 17:13:23 2023 -0500\n\n update lucene\n\ncommit 4c6b43e3606b8c9eb5e4ec4cf0817dc9b061c555\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 16:24:33 2023 -0500\n\n add debugging information when node is not found on level\n\ncommit 81c8e730bd4ca314a81bc13e30b197603a8eb005\nAuthor: Jonathan Ellis \nDate: Fri Jun 16 11:57:24 2023 -0500\n\n add similarityWithAnn test\n\ncommit 87c8da9a5dddd10215a84c07a09dbaaecaa522b8\nAuthor: Jonathan Ellis \nDate: Wed Jun 14 18:26:02 2023 -0500\n\n add tests to confirm that zero-length vectors are rejected\n\ncommit 2510c1b987d846821e7095fbc2c368fa51d9a15d\nMerge: a3d9e33554 e86f91c568\nAuthor: Jonathan Ellis \nDate: Mon Jun 12 17:09:22 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit a3d9e3355470f6400328cd7ac1eb07ff06f8e00b\nAuthor: Jonathan Ellis \nDate: Mon Jun 12 17:08:43 2023 -0500\n\n use ArrayList in CVV instead of HashMap since we know the keys are consecutive\n\ncommit e86f91c568f8a82e83e61bd0dd768478a7584cfc\nMerge: 77c1bc11f3 8a7a6d9c4f\nAuthor: Jeremiah D Jordan \nDate: Mon Jun 12 09:25:06 2023 -0500\n\n Merge remote-tracking branch 'datastax/ds-trunk' into vsearch\n\ncommit 77c1bc11f31e4ff001c7eff1e6fe50812e136a45\nAuthor: Jeremiah D Jordan \nDate: Mon Jun 12 09:24:56 2023 -0500\n\n Revert \"Revert this after CNDB-6974 is fixed.\"\n\n This reverts commit 8e04280312583a18085e7e7b9d31810790b039f4.\n\ncommit cae30e9879c463d80de2a2b7e4a642979803c13a\nAuthor: Jeremiah D Jordan \nDate: Mon Jun 12 09:23:39 2023 -0500\n\n Revert \"Revert this after CNDB-6974 is fixed.\"\n\n This reverts commit 67a2b7eba3c957a993726d37706643979e92a3a3.\n\ncommit 46d1a45de84f40311f193beba3ca5fc3fe7450d7\nAuthor: Jonathan Ellis \nDate: Mon Jun 12 09:03:03 2023 -0500\n\n replace our TODO entries with VSTODO to make them easier to find\n\ncommit 415b3f29e8e7f1526e28655308a1937a1175e575\nMerge: 2fd8b848be 481d29721a\nAuthor: Jonathan Ellis \nDate: Mon Jun 12 09:02:16 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit 481d29721a7993e9babd44f792be55aca925d41a\nAuthor: Zhao Yang \nDate: Mon Jun 12 21:26:43 2023 +0800\n\n Fix VectorMemtableIndex to handle max token and fix Segment#intersects (#670)\n\n * Fix VectorMemtableIndex to handle max token and min/max bound\n\n * Fix Segment#intersects to compare bound instead of token and add tests for range search\n\n * make brute force rows per query for VectorMemtableIndex\n\n * apply feedback on Segment#intersects\n\n * add comments to VectorMemtableIndex#search\n\n * Fix SegmentTest\n\ncommit 2fd8b848be82e62999e64b57cb9800d03de8e953\nAuthor: Jonathan Ellis \nDate: Mon Jun 12 07:51:18 2023 -0500\n\n clean up a couple REVIEWME\n\ncommit 3c066e07be6188c3f48c51c02cfc99b395e2a6bc\nAuthor: Zhao Yang \nDate: Sun Jun 11 14:26:42 2023 +0800\n\n fix flaky VectorDistributedTest\n\ncommit 6c719a752e90e031916dddf6cc7b7db23627bc38\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:48:01 2023 -0500\n\n fix use of bruteForceRows -- should be larger of limit,maxBruteForceRows\n\ncommit 0a2279d7f010bcb68053bb14ea2974295b21d1c4\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:46:34 2023 -0500\n\n use maxBruteForceRows when deciding whether to skip ANN\n\ncommit 1859f7355231ce7972e3d4b7af70e5d2967da516\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:27:03 2023 -0500\n\n simplify partitionKeySearchTest using euclidean distance\n\ncommit 4f40e1c2d9b1c30675eeaee7d800a8113cce97c0\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:09:24 2023 -0500\n\n typo\n\ncommit 7dbe85ac9a248aa6a0985839df1f40ceb2c08e6c\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:08:23 2023 -0500\n\n rename methods that return Bits but had bitset in their names\n\ncommit 2861bfdba8f2638a1e3f9e81df9cad3878d0e544\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 10:06:51 2023 -0500\n\n simplify skipANN\n\ncommit fbfe4bcdd03a66f4dedfa91ce9d06206f89827cd\nAuthor: Zhao Yang \nDate: Sat Jun 10 16:36:56 2023 +0800\n\n optimize VectorIndexSearcher#searchPosting\n - return empty posting if key range is not found in current sstable\n - return empty posting if all row ids are shadowed\n - skip ANN if matching row ids are less than limit\n\ncommit 168911b70888a3dda0ddf4f615af1b7ff54f43e3\nAuthor: Zhao Yang \nDate: Sat Jun 10 20:54:41 2023 +0800\n\n fix data visibility issue during flush: make sure SSTableAddedNotification is sent before MemtableDiscardedNotification\n\ncommit c5dd47e3ce1c7276703fa3f89a9932a3b2cd43b7\nAuthor: Zhao Yang \nDate: Sat Jun 10 12:31:53 2023 +0800\n\n fix primaryKeySearchTest and partitionKeySearchTest to use correct selector and expected results\n\ncommit 2e636529c63d0d726868d6df9b3e11d15d3871d8\nAuthor: Zhao Yang \nDate: Sun Jun 11 08:31:39 2023 +0800\n\n Vector-48: index#update is not triggered by partition/range deletion,… (#665)\n\n * Vector-48: index#update is not triggered by partition/range deletion, we have to fix VectorMemtableIndex to include shadowed primary keys\n\n - during flush, append ordinals that are removed by partition/range deletion into deletedOrdinals set\n\n * add comments to VP\n\n * revert redundant variables\n * simplify\n\n ---------\n\n Co-authored-by: Jonathan Ellis \n\ncommit daa2623b880e4b82d99204fe718db22e52ae3b42\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 13:54:38 2023 -0500\n\n use mmapped builder in OnDiskHnswGraphTest\n\ncommit 553778a22cfa8757daf3b3ec0e2cfa6fc7ba29cf\nAuthor: Jonathan Ellis \nDate: Sat Jun 10 08:34:41 2023 -0500\n\n fix logic in partitionKeySearchTest, test still fails\n\ncommit a7f5a2f824085691da80ef62e37a430e5637c31c\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 17:57:34 2023 -0500\n\n ignore invalid vectors during build against existing data, instead of failing the build\n\ncommit 2af203646a2e57fca0a2521c31ba4269b0d8f326\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 17:43:47 2023 -0500\n\n switch from ignoring zero vectors to throwing IRE\n\ncommit 4c98fbb447a275cbf557ae0adf8f4f8a30fba17f\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 17:15:57 2023 -0500\n\n Revert \"if there are no rows in the range to search, return EmptyPostingsList instead of giving the hnsw search an empty Bits (which is a pathological case for the search)\"\n\n This reverts commit 6e5eab787a3add985dc30eca82c5dee2646f5564.\n\ncommit da99598a5d11dfa4ecac3b1eeeb9d19a4bad21e5\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 17:09:30 2023 -0500\n\n don't attempt to add zero vector to cosine indexes\n\ncommit a77792447d33600f69dd0a75a280a2b6dac51389\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 16:43:34 2023 -0500\n\n add failing partitionKeySearchTest\n\ncommit b061dd1c77f30c67324da4f94ea8b5a22d50fba7\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:52:08 2023 -0500\n\n inline the test ops\n\ncommit 6e5eab787a3add985dc30eca82c5dee2646f5564\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:51:09 2023 -0500\n\n if there are no rows in the range to search, return EmptyPostingsList instead of giving the hnsw search an empty Bits (which is a pathological case for the search)\n\ncommit 518c055dfeabacf0fa0863e40bf814058f2ad981\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:28:57 2023 -0500\n\n cleanup\n\ncommit e922b03ffe5436dc1341dbe20ee7aaca9901058f\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:27:53 2023 -0500\n\n failing tests for primary key search\n\ncommit 428e4b713e25a5820d02ca790c30e9f1477c0f44\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:21:59 2023 -0500\n\n cleanup\n\ncommit 5319bcdf3b694e861509b2a9f549f39bf51cd253\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:20:02 2023 -0500\n\n move testInvalidColumnNameWithAnn to VectorInvalidQueryTest\n\ncommit d8b64845e59c0bd3658a16df21dcc75564417e3b\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:12:40 2023 -0500\n\n upgrade lucene to reduce Integer boxing on build path\n\ncommit a62356e49c0a32f61732b8525d2bb655f46dd767\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 10:11:55 2023 -0500\n\n cleanup\n\ncommit b754eb7f021d10d0ae8de83ebdf2ce2f19ce59d8\nAuthor: Jonathan Ellis \nDate: Fri Jun 9 08:29:34 2023 -0500\n\n replace CHM with NBHMLong to avoid boxing\n\ncommit 53d232ad2bcddefd72b10542f703cf36728c05f5\nAuthor: Piotr Kołaczkowski \nDate: Fri Jun 9 12:17:46 2023 +0200\n\n STAR-550 Handle SAI AbortedOperationException\n\n AbortedOperationException is thrown by SAI when index search\n hits a timeout. Now instead of allowing this exception to\n bubble up to the top and be eventually logged as error, we catch\n and swallow it in the InboundSink after creating the query response.\n Additionally now we also we set a proper error code (TIMEOUT)\n in the response, so the client has a hint on what happened.\n\ncommit 5e0c5eb3c54b58002162ce3e39d0891643e1f663\nAuthor: Matt Fleming \nDate: Fri Jun 9 08:44:10 2023 +0100\n\n CNDB-7037: Check that memtable exists before flushing and avoid IOOBE (#664)\n\n For offline services such as the compactor it's possible to not have\n live memtables. Account for this during flushing after removing indexes to\n avoid triggering an IndexOutOfBoundsException:\n\n Error happened while updating the schema\n java.lang.IndexOutOfBoundsException: Index: -1\n \tat java.base/java.util.Collections$EmptyList.get(Collections.java:4483)\n \tat org.apache.cassandra.db.lifecycle.View.getCurrentMemtable(View.java:106)\n \tat org.apache.cassandra.db.ColumnFamilyStore.forceFlush(ColumnFamilyStore.java:1115)\n \tat org.apache.cassandra.db.SystemKeyspace.forceBlockingFlush(SystemKeyspace.java:552)\n \tat org.apache.cassandra.db.SystemKeyspace.setIndexRemoved(SystemKeyspace.java:576)\n \tat org.apache.cassandra.index.SecondaryIndexManager.markIndexRemoved(SecondaryIndexManager.java:837)\n \tat org.apache.cassandra.index.SecondaryIndexManager.removeIndex(SecondaryIndexManager.java:420)\n\ncommit c241f7d41ee46c28f60d691611f32e071dac6684\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 19:19:09 2023 -0500\n\n read neighbors using .intBuffer since we know that the HNSW searcher always reads all the neighbors\n\ncommit 32f021a4f0623e57d82d70c54350eab6f0f57db7\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 18:08:30 2023 -0500\n\n VECTOR-51 leave vectors as ByteBuffers during compaction to avoid 2x memory usage incurred by keeping them around as float[] as well\n\ncommit 32d8cbbd19301228530d3d84f7a7e4120b4d4422\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 17:37:08 2023 -0500\n\n r/m node cache from query metrics since there is nothing the operator can act on there\n\ncommit 890df8e6c36c5c9e1314dbcc0a681415401120dd\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 14:15:25 2023 -0500\n\n add more information to exception when reading row offsets goes wrong\n\ncommit 5c7fe0cfed1edb10f81b134f8949dd1b5abb2781\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 13:26:17 2023 -0500\n\n revise ordinals cache as follows:\n // cache full levels including neighbors up to neighborsRamBudget, starting with the top level,\n // but always cache all levels above the bottom two levels -- this will be ~1% of the graph.\n // then on L1, cache at least the offsets\n // L0 we do not cache since we only need one extra seek (no bsearch) to read the neighbors offset\n\ncommit e131014c5cba59624e0ed9fac142abe5340e2fc0\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 13:53:10 2023 -0500\n\n add deletes test\n\ncommit 7c5a880f007535d9f03b7c7d9d3e84dcc79ef3d7\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 13:34:42 2023 -0500\n\n must use graph.size for bitset size -- not correct to use max rowid, since deleted ordinals will not have a rowid but will still be in the graph\n\ncommit 0654fbb61aec86deae69edf83fb1cf653ed7df65\nAuthor: Jonathan Ellis \nDate: Thu Jun 8 08:44:28 2023 -0500\n\n info -> debug for validatePerIndexComponents\n\ncommit b289db2362b928f29b6c5c6289e09b33f0be1e88\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 18:08:18 2023 -0500\n\n update lucene\n\ncommit 347dae644b0d395014d8faf2affd48a6b2546e96\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 18:23:41 2023 -0500\n\n undo burntest logging config change\n\ncommit 87a6a84f38d0a0423acd9a0dda67d386876681b5\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 15:20:06 2023 -0500\n\n add debug logging\n\ncommit 20013bc865c7fb2c34111c98362aaf997dc724dc\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 14:50:39 2023 -0500\n\n add a bit more information to exception when we fail in index construction from disk\n\ncommit 86389ceae8188b8e427d8a528fada2f913f1f633\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 12:25:31 2023 -0500\n\n reduce test vector count from \"all of them\" to 200\n\ncommit 20177a1c84ba34babc82d8c1aa4e3436321b5d9f\nMerge: 2e2fce09ee f2c697ac46\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 12:21:20 2023 -0500\n\n Merge branch 'wip' into vsearch\n\ncommit f2c697ac46348301260c5947ade4ebed7dee90ee\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 12:16:28 2023 -0500\n\n multipleSegmentsMultiplePostingsTest\n\ncommit 59e4a1c53e1c19a7ebba4a0834229e61f9e4d3ce\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 12:16:04 2023 -0500\n\n cleanup\n\ncommit 2e2fce09eeb5bef4780cecf7c7b5e492a810f67a\nAuthor: Zhao Yang \nDate: Wed Jun 7 23:33:00 2023 +0800\n\n fix OnDiskOrdinalsMap to seek to segment offset before reading (#661)\n\ncommit 9502f1040229f2be6619f7e4497dc25ca5126b39\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 10:04:38 2023 -0500\n\n fix build\n\ncommit 3535444b57366fe705e0fd8c1ca095e60ed2a706\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 09:19:41 2023 -0500\n\n add write-only workload\n\ncommit a112b0965cb298f293b6e384bafc92a1d5a1fb8f\nAuthor: Zhao Yang \nDate: Wed Jun 7 21:15:53 2023 +0800\n\n VECTOR-44: improve in-memory partition-restricted query perf (#660)\n\n * VECTOR-44: improve in-memory partition-restricted query perf\n - using post-filter top-k processor instead of ANN: 14x improvement on partition-restricted query in LongVectorTest\n\ncommit cf1ec3be50ddbe40ae99dfc34eea51b9fc5c2648\nAuthor: Zhao Yang \nDate: Wed Jun 7 21:06:37 2023 +0800\n\n Vector-45: fix bitsetForShadowedPrimaryKeys to skip shadowed primary keys outside of current sstable/segment (#659)\n\n * VECTOR-45: fix bitsetForShadowedPrimaryKeys to skip shadowed primary keys outside of current sstable/segment\n\n * return empty iterator if results are empty instead of ReorderingRangeIterator\n\ncommit 87217b4f49935fbbd13c73f4d0aadb84836696a7\nAuthor: Jonathan Ellis \nDate: Wed Jun 7 07:31:46 2023 -0500\n\n don't make areL0ShardsEnabled final, it breaks mocks\n\ncommit 2bb3a299df43561369544b662674872575975b7c\nAuthor: Matt Fleming \nDate: Wed Jun 7 10:56:08 2023 +0100\n\n VECTOR-42: Check columns exist to avoid NPE during ANN expr binding (#657)\n\n Using an non-existent column with an ANN expression triggers an NPE like\n so,\n\n java.lang.NullPointerException: null\n \tat org.apache.cassandra.cql3.ArrayLiteral.forReceiver(ArrayLiteral.java:43)\n \tat org.apache.cassandra.cql3.ArrayLiteral.prepare(ArrayLiteral.java:54)\n \tat org.apache.cassandra.cql3.Ordering$Raw$Ann.bind(Ordering.java:178)\n \tat org.apache.cassandra.cql3.Ordering$Raw.bind(Ordering.java:139)\n\n Use TM.getExistingColumn() which throws InvalidRequestException if the\n column is undefined.\n\ncommit f29d4528fc43f34889e560acafa810eee1b88ba9\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 21:18:47 2023 -0500\n\n restore query timeouts\n\ncommit 6e5734e52b41ab1e355d202ad0433440828fbb75\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 19:08:52 2023 -0500\n\n looking at performance over time in LongVectorTest\n\ncommit f96af8f0b9185163a49156bdab107801d2588bf0\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 16:00:23 2023 -0500\n\n log level = info for burn tests\n\ncommit 1247cf372927cc185be5c9767a06fdd15f102dbe\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 17:48:00 2023 -0500\n\n write the in-memory deleted ordinals to disk at the start of the postings component\n\ncommit e319d71bd6b0d265f616c807dcaf4a9d9fc38aef\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 16:36:27 2023 -0500\n\n don't allocate unnecessary objects on the happy path of no tombstones\n\ncommit 6d146a8460a17d146de756aab2ccbbf2abdb967a\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 16:19:56 2023 -0500\n\n VECTOR-23: force UCS to not shard L0\n\ncommit 79d6e177093312bc1b951d6740ad45ce8a6c5875\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 16:01:05 2023 -0500\n\n mark AutoCloseable\n\ncommit 5ec2f7d7f792bfa2dcf582ea4eed3fa8745fdaff\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 14:37:15 2023 -0500\n\n Support string literals as vectors\n\n Co-authored-by: Andrés de la Peña \n\ncommit 73a53c37536835e462e3cf17c452027c7aa591ed\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 12:25:39 2023 -0500\n\n (take 3, this time with the code changes) fix race conditions across concurrent inserts + searches in memtable\n\ncommit 51f4f419d31916a81088f171b158f1e002b5c800\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 12:25:39 2023 -0500\n\n (take 2) fix race conditions across concurrent inserts + searches in memtable\n\ncommit 94097a1f7cf1831c03e52b1f539bdc2e42fa038b\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 12:35:32 2023 -0500\n\n Revert \"fix race conditions across concurrent inserts + searches in memtable\"\n\n This reverts commit c4d41b492190fb2644f83bf4902acf71d1e4f891.\n\ncommit c4d41b492190fb2644f83bf4902acf71d1e4f891\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 12:25:39 2023 -0500\n\n fix race conditions across concurrent inserts + searches in memtable\n\ncommit 22143a8c5db160ef0fa7687ff3e8cb227de160f0\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 10:28:30 2023 -0500\n\n fix AOOB in bruteForceRows logic\n\ncommit b5624f62c88378e03260399440edf4d335ccc3af\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 10:12:53 2023 -0500\n\n call deserializeFloatArray (which gives float[]) instead of deserialize (which gives List)\n\ncommit 13ad40718737b54652c875d81ec20992b5828777\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 08:36:32 2023 -0500\n\n create CassandraHnswGraphBuilder with concurrent and serial implementations, so that single-threaded compaction doesn't have to pay the concurrency overhead\n\ncommit b83f4e4007e3d597963023b8ae32f4d0934b792d\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 08:00:44 2023 -0500\n\n add ASL header to injections.md to make CI happy\n\ncommit 16074336b1322d4bade40eb71562ca50221399dc\nMerge: 227c6c13a3 1ff6992354\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 07:48:04 2023 -0500\n\n Merge branch 'VECTOR-37' into vsearch\n\ncommit 227c6c13a3cfe42484fc560d629d54a93f8265e0\nAuthor: Jaroslaw Grabowski \nDate: Tue Jun 6 13:02:37 2023 +0200\n\n CNDB-7007 return expired tables level from getLevels\n\ncommit 312d07de8d297cdede39017354b678f2fb1b1006\nAuthor: Jonathan Ellis \nDate: Tue Jun 6 07:31:26 2023 -0500\n\n randomize our brute force threshold, which will get the actual index scans exercised more\n\ncommit 0b897961869965629893c94250b7d4b1fb7c0f86\nAuthor: Zhao Yang \nDate: Tue Jun 6 13:44:01 2023 +0800\n\n Reference ANN sstable indexes in case of ann hybrid search (#653)\n\n * Reference ANN sstable indexes in case of ann hybrid search\n\n * simplify\n\n ---------\n\n Co-authored-by: Jonathan Ellis \n\ncommit 1ff6992354e8411f9f7c48a76f0aadb4a00f2789\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 21:55:14 2023 -0500\n\n per-query hnsw metrics\n\ncommit d3d6ae107737020eb0ca3b00c117d5bc84027dc0\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 21:16:14 2023 -0500\n\n reduce test size to prevent Jenkins OOM\n\ncommit 7765ac73fd29c9664b6f8e946c067d6f1b0a9a03\nMerge: 5a04dbcf67 c5cd09cebf\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 21:15:44 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit c5cd09cebf0c63f7bafcd2f4548395efe68b12cd\nAuthor: Zhao Yang \nDate: Tue Jun 6 09:38:42 2023 +0800\n\n Skip warning log about receiving a range that is not owned by the current replica, see VECTOR-30\n\ncommit 5a04dbcf6738d7681732e28518d6eb37f74357b8\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 11:21:15 2023 -0500\n\n add injections.md from bdp repo\n\ncommit baf9cce301a39f9506bd60c5a75d2011150f2fea\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 10:02:02 2023 -0500\n\n VECTOR-36 fix looking up indexes and SSTableContext by SSTableReader, because the SSTR will be removed from internal structures once it's compacted\n\ncommit e7fdc8454e750e2ee886c4ab27fb09abb46303fa\nAuthor: Jonathan Ellis \nDate: Mon Jun 5 09:53:51 2023 -0500\n\n limitToTopResults should skip rows that are not in the current segment (or more preciesely, have no vectors that were indexed)\n\ncommit 2f64695a90651d9e76eee0a3399032ea241c983c\nAuthor: Zhao Yang \nDate: Mon Jun 5 14:55:20 2023 +0800\n\n Vector 6 take 2: reapply Vector-6 commit and fix NPE in SelectStatement (#648)\n\n * Revert \"Revert VECTOR-6\"\n\n This reverts commit 63a6ff609a02e4f7d9f5499f1af48ca5e58f63a8.\n\n * Vector-6 take 2:\n - fix NPE in SelectStatement by skipping reversed() for null ColumnComparator\n - fix getOrderingColumns() to return LinkedHashMap to preverse ordering columns' order\n - fix VectorLocalTest compilation\n - remove debug log in CassandraOnHeapHnsw\n\ncommit 9da51b315d207417de4f3e005616e275c62c74cb\nAuthor: Zhao Yang \nDate: Mon Jun 5 12:38:21 2023 +0800\n\n fix SAI test failures in vsearch branch (#649)\n\n - BatchlogEndpointFilterTest\n - rangeRestrictedTest\n - SegmentFlushTest\n - SegmentMergerTest\n - OperationTest\n - RangeIntersectionIteratorTest\n\ncommit dc53d7db5f0f8fe1c780f6d2b7580794f7b6b5fb\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 13:35:11 2023 -0500\n\n create views for ordinals map so we don't have to open a new Reader for each method call\n\ncommit c6385022ff2949c23e0dd0161a043cf4061f050c\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 08:19:32 2023 -0500\n\n add assert sstableContext != null\n\ncommit f340e7590db578b71201326e82006f72dd491047\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 07:37:36 2023 -0500\n\n on-disk Searcher bits should not need to be growable\n\ncommit fd5cc7f989a47d03689dd29495b974697ad1338b\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 07:37:18 2023 -0500\n\n add asserts\n\ncommit 6436e5f4ad49881a62c1442692edfedfe31968dc\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 07:21:50 2023 -0500\n\n cleanup\n\ncommit 32c8e005740d3c7ec0b1008091728e12260aecdd\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 13:57:18 2023 -0500\n\n comment\n\ncommit 091d96832ff9a68f6ea7857eafd2ec7ee0c09a0b\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 13:32:24 2023 -0500\n\n r/m unused search method returning iterators over PrimaryKey\n\ncommit 4e78a4b429b0ba6690871df9a91c8ffd8d2e53fa\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 11:25:30 2023 -0500\n\n we can use the slightly more lightweight RangeConcatIterator instead of RangeUnionIterator when combining results from different Segments\n\ncommit e7a9186bf3426ad025e809481e47db85a7bf7190\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 11:22:35 2023 -0500\n\n r/m obsolete FIXME\n\ncommit c76ee9ea0cab5b70c535de3f50b81cf333bf94a2\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 07:12:43 2023 -0500\n\n add testAppendedGraphs\n\ncommit d00bdbfb6acf9f08d2bdfa2ef06e833e567cf9eb\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 06:57:37 2023 -0500\n\n move write-to-File to test class\n\ncommit b10fd1100daf4cde0291b3b4ef06f5ceb3ca650c\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 06:46:01 2023 -0500\n\n fix Ref leaks in test\n\ncommit 0656e558bb51172df4e0cde1ebe92946d4dcec8b\nAuthor: Jonathan Ellis \nDate: Sat Jun 3 06:26:05 2023 -0500\n\n fix tests to use View\n\ncommit 92797ced03ed26f9a5cce398504d6f0de519b899\nAuthor: Zhao Yang \nDate: Sat Jun 3 19:19:38 2023 +0800\n\n VECTOR-35: fix vector on-disk writer to append segments (#647)\n\ncommit 22f5c01860367586d0cc39def7d975a6d6bae784\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 18:54:41 2023 -0500\n\n add \"Vector indexes only support ANN queries\" check\n\ncommit bfc0b056dbacd0a1eca86020e901f545dca7670e\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 17:34:24 2023 -0500\n\n split invalid requests into separate test class\n\ncommit f66768c35920101382c1975c9d0ba0e3301b7c61\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 17:28:56 2023 -0500\n\n add specific error message when trying to do ANN without an index\n\ncommit 63a6ff609a02e4f7d9f5499f1af48ca5e58f63a8\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 12:31:22 2023 -0500\n\n Revert VECTOR-6\n\n This reverts commits:\n a37009c187edeba68389d239dc1b9f40519b1187\n 5565690fbe1056d4c159ddbe233fa22c7695320a\n e7733bb8f858a16b082b8a5c64d0322db6f6271a\n\ncommit 3d496662b259470b505d88212074c436660c27ad\nMerge: fab67bb134 4bdae7e362\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 11:05:53 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit 4bdae7e36213f5112efce085696dd24aaa0adfad\nAuthor: Mike Adamson \nDate: Fri Jun 2 14:36:01 2023 +0100\n\n Stabilise random tests using word2vec model vectors\n\ncommit 2ddc3ec6b11f6810f6c4cbaf6b62bfe46d2d1b6c\nMerge: 8e04280312 1f9179002a\nAuthor: Jeremiah D Jordan \nDate: Fri Jun 2 10:05:15 2023 -0500\n\n Merge remote-tracking branch 'datastax/ds-trunk' into vsearch\n\ncommit 8e04280312583a18085e7e7b9d31810790b039f4\nAuthor: Jeremiah D Jordan \nDate: Fri Jun 2 09:51:06 2023 -0500\n\n Revert this after CNDB-6974 is fixed.\n\ncommit fab67bb13499691f813c893cd39a3bbd406653d2\nAuthor: Jonathan Ellis \nDate: Fri Jun 2 08:59:12 2023 -0500\n\n vector cache defaulting to 1MB per segment\n\ncommit 67a2b7eba3c957a993726d37706643979e92a3a3\nAuthor: Jeremiah D Jordan \nDate: Thu Jun 1 17:49:35 2023 -0500\n\n Revert this after CNDB-6974 is fixed.\n\ncommit 4538956b31d8c40d2e9b603b3b0a3392343e1853\nMerge: 7518680a4a af4c83aef4\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 16:17:14 2023 -0500\n\n Merge branch 'ds-trunk' into vsearch\n\ncommit 7518680a4a17cb2589cf06a9175befc10c9eab1a\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 16:14:33 2023 -0500\n\n re-use float[] across calls to vectorValue to avoid allocation overhead (credit to Jake for the idea)\n\ncommit d20bd8aedeb6232358dd16f901e2708f217b8108\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 15:35:17 2023 -0500\n\n optimize vectorValue() with direct access to the mmap-ed region\n\ncommit 24572025af5b7893279b5b03c58555b682f3abdb\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 14:33:55 2023 -0500\n\n decomposeVector does not modify the underlying buffer, so no need to duplicate() here\n\ncommit efef35d47a24222fae300d1cbd5a801e9ed79442\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 14:29:30 2023 -0500\n\n specialize deserializeFloatArray for ByteBuffer and FloatBuffer. this saves about 3% total CPU on search workloads\n\ncommit 63c27f5550d6c3e22d960b9a42b9097029e3b760\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 11:04:23 2023 -0500\n\n default target size of 5GB\n\ncommit 69c55c8d07c6e1d58b4829c153f5ebf8a4343669\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 07:20:38 2023 -0700\n\n make the on-disk hnsw code threadsafe using FileHandle.createReader (#641)\n\ncommit b194cae2e56133b946d0231768254064a47f01b6\nMerge: 7ee6422084 23c2891e7a\nAuthor: Jonathan Ellis \nDate: Thu Jun 1 06:22:17 2023 -0500\n\n Merge branch 'ds-trunk' into vsearch\n\ncommit 7ee642208485c9e2ae62aa662dbbddc7b295bae1\nMerge: c286ec0ee2 a37009c187\nAuthor: Jonathan Ellis \nDate: Wed May 31 20:41:48 2023 -0500\n\n Merge branch 'VECTOR-6' into vsearch\n\ncommit c286ec0ee231ededded18e81357edebe72064c59\nAuthor: Jonathan Ellis \nDate: Wed May 31 20:40:21 2023 -0500\n\n add testLargeGraph\n\ncommit a37009c187edeba68389d239dc1b9f40519b1187\nAuthor: Zhao Yang \nDate: Thu Jun 1 08:50:07 2023 +0800\n\n cleanup unused code\n\ncommit 6246dbe3a970a0f4672701632003cdf576705c24\nAuthor: Jonathan Ellis \nDate: Wed May 31 16:43:42 2023 -0500\n\n comment\n\ncommit c7420440c7d3c81eac96e716148d570ae3cceb1d\nAuthor: Jonathan Ellis \nDate: Wed May 31 16:43:37 2023 -0500\n\n encapsulate shadowedPrimaryKey better\n\ncommit a1a25c33b8bec9d78354dd2c1d992444db2bf76f\nAuthor: Jonathan Ellis \nDate: Wed May 31 15:57:52 2023 -0500\n\n make hnsw cache size configurable, and make the default 128KB (roughly the size of the bloom filter for a 1GB sstable)\n\ncommit 55747d2351e915645697fa8dec404c749540456f\nAuthor: Andrés de la Peña \nDate: Wed May 31 19:28:47 2023 +0100\n\n Replace FunctionParameter.sameAsFirst by FunctionParameter.sameAs, allowing type inferrence in both directions for vector functions\n\ncommit 5565690fbe1056d4c159ddbe233fa22c7695320a\nAuthor: Jonathan Ellis \nDate: Wed May 31 12:59:42 2023 -0500\n\n cleanup and fix\n\ncommit e7733bb8f858a16b082b8a5c64d0322db6f6271a\nAuthor: Zhao Yang \nDate: Wed May 24 16:01:26 2023 +0800\n\n VECTOR-6: return vector results to client in ANN order instead of token order\n\ncommit 771067d1475c94484da178236928d2bad78e00b1\nAuthor: Mike Adamson \nDate: Wed May 31 18:33:55 2023 +0100\n\n Fix annOrderingMustHaveLimit test\n\ncommit 8dd76a73541e70da0da022dd463e6711a315f971\nAuthor: Mike Adamson \nDate: Wed May 31 18:30:49 2023 +0100\n\n Improve the validation of ORDER BY ANN OF (#639)\n\n * Improve the validation of ORDER BY ANN OF\n\n * Change to hasNonClusteredOrdering and improve limit message\n\ncommit 1ef6f2e3418db19af23f0eae24cd344d72290455\nAuthor: Jonathan Ellis \nDate: Fri May 26 18:34:25 2023 -0500\n\n add vector similarity functions\n\ncommit f4e3a39a8ece623647ba7891981e7c5ee10ed96b\nAuthor: Jonathan Ellis \nDate: Wed May 31 09:39:45 2023 -0500\n\n pull in the smallest set of changes possible from 93e0ae9a to get FunctionParameter.sameAsFirst\n\ncommit 0ef2614346c9426a966004df220221f74353de70\nAuthor: Jonathan Ellis \nDate: Wed May 31 06:22:41 2023 -0700\n\n Add support for updates + deletes (#636)\n\ncommit 5da9fefe1e18f733bb8ffd0a35ef2e1ac3cbd975\nAuthor: Jonathan Ellis \nDate: Tue May 30 16:32:02 2023 -0500\n\n rename SegmentOrdering.reorderOneComponent to limitToTopResults to align with MemtableOrdering\n\ncommit caf6eb2154ddf57c88d69c6246404b65e183057d\nAuthor: Mike Adamson \nDate: Tue May 30 22:41:29 2023 +0100\n\n Fix V1SearchableIndex.reorderOneComponent to use segments (#635)\n\ncommit 72cfa94e9389d0078d64aee60d26b8b01c6ca22f\nAuthor: Piotr Kołaczkowski \nDate: Mon May 29 20:17:44 2023 +0200\n\n VECTOR-14: Move ANN OF expressions from WHERE to ORDER BY\n\n Before:\n SELECT * FROM table WHERE ANN OF ;\n\n After:\n SELECT * FROM table ORDER BY ANN OF ;\n\ncommit 226266ef124fc220819d328ed8547c5d86626c4b\nAuthor: Mike Adamson \nDate: Tue May 30 18:58:16 2023 +0100\n\n Implement getInstance(TypeParser) for VectorType. Fix losing data at startup.\n\ncommit 2bbe88ad88e2243f12dc8df789ea9bdf208ff74b\nMerge: dd0ffb64f3 a3b8661746\nAuthor: Jonathan Ellis \nDate: Tue May 30 12:37:58 2023 -0500\n\n merge ds-trunk\n\ncommit dd0ffb64f3e54e2ce1af75da52fe3ece66a49f3f\nAuthor: Jonathan Ellis \nDate: Tue May 30 11:40:20 2023 -0500\n\n reduce startup log noise at info level\n\ncommit 1ef17101def1aacbb84f3434633a9b706291aa4b\nAuthor: Jonathan Ellis \nDate: Tue May 30 11:34:26 2023 -0500\n\n add partialUpdateTest\n\ncommit f816046a17809b06ad754a16c0f089c0d26909f9\nAuthor: Jonathan Ellis \nDate: Mon May 29 12:57:12 2023 -0500\n\n fix recall computation in test\n\ncommit 1c1442c36baafdf4e1641a9c0288dbd520cf7c0e\nAuthor: Jonathan Ellis \nDate: Mon May 29 12:34:02 2023 -0500\n\n also apply tolerance for inexact results to searchWithKey, but only for size > 10\n\ncommit 39df5112c6a551b53649ad98164f2bde93cbdc10\nMerge: d598206329 8ca4bba861\nAuthor: Jonathan Ellis \nDate: Mon May 29 12:27:24 2023 -0500\n\n merge ds-trunk\n\ncommit 8ca4bba8617e8d0fa4f48de1c479de66e3a77dd3\nAuthor: Jonathan Ellis \nDate: Mon May 29 12:19:32 2023 -0500\n\n rename MemtableIndex -> TrieMemtableIndex to make merge to vsearch easier\n\ncommit d5982063292b6a91b156da32454abacd36fa6980\nMerge: ae96e0f423 1488a5f0b9\nAuthor: Jonathan Ellis \nDate: Mon May 29 09:03:24 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit ae96e0f42393eab3547e2314f1acdb1ff766f2a6\nAuthor: Jonathan Ellis \nDate: Mon May 29 09:02:12 2023 -0500\n\n check for results within a percentage (I went with 5%) of the expected; the A in ANN means we shouldn't expect to find 100% of matches unless the graph is tiny\n\ncommit 1488a5f0b9c65473bb6a69c9e18c045d931a1fe3\nAuthor: Zhao Yang \nDate: Mon May 29 09:20:28 2023 +0800\n\n Vector-19: fix PrimaryKey min/max prefix ByteComparable and add reversed lookup for row id (#627)\n\n * Vector-19: fix PrimaryKey min/max prefix ByteComparable and add reversed lookup for row id\n\ncommit 14c41f5f913dc7f7f4cfc64b932bb6b9df475f19\nAuthor: Jonathan Ellis \nDate: Wed May 17 17:35:16 2023 -0500\n\n use incremental bytes used estimate from hnsw to avoid recomputing full ramBytesUsed on every call to add\n\ncommit f8d2aaad6e96693f532d745f71e0eaabb1ddf934\nAuthor: Jeremiah D Jordan \nDate: Fri May 26 12:02:54 2023 -0500\n\n Update cqlsh for new syntax\n\ncommit f0a882ae1171d41bc9c6ccf224ce3be80f7128e8\nAuthor: Mike Adamson \nDate: Fri May 26 16:24:38 2023 +0100\n\n Only allow VectorType to accept float\n\ncommit a29712b1365cdfac9a709e35325e85f7fa6c61c8\nAuthor: Mike Adamson \nDate: Fri May 26 16:17:51 2023 +0100\n\n Fix max term size for vectors at 16k is SSTableIndexWriter\n\ncommit 363d3f4e35c1a4df0f257557309bb042fd3f3ec7\nAuthor: Mike Adamson \nDate: Fri May 26 15:23:02 2023 +0100\n\n Merge CASSANDRA-18504 vector grammar (#628)\n\n - Now uses vector to describe vector\n - This commit does not bring in the whole 18504 patch\n only the essential grammar and type parts of the\n patch\n\ncommit 18c4e35d4ce14cda7a3c03398e16edf368ad9e6a\nMerge: 16158a5b48 b0f71ee2c3\nAuthor: Jonathan Ellis \nDate: Thu May 25 16:24:41 2023 -0500\n\n Merge branch 'VECTOR-3' into vsearch\n\ncommit b0f71ee2c371d1003cf241c3aedd7437385bcecb\nAuthor: Jonathan Ellis \nDate: Thu May 25 15:51:27 2023 -0500\n\n cleanup\n\ncommit 2d840eae74c079229729767731d3719d12ca1931\nMerge: c01cacf300 197a3207b1\nAuthor: Jonathan Ellis \nDate: Thu May 25 15:47:39 2023 -0500\n\n Merge branch 'vsearch' (early part) into VECTOR-3\n\ncommit 16158a5b48665f57bde649e87a09d078298932f5\nAuthor: Jonathan Ellis \nDate: Thu May 25 15:45:51 2023 -0500\n\n optimize ramBytesUsed\n\ncommit 197a3207b16ec97bae4924883247e2cd6a2923bd\nAuthor: Jonathan Ellis \nDate: Thu May 25 13:34:19 2023 -0500\n\n update lucene\n\ncommit a7bfcc7a6f31090c3c8443d299dabb7acd437ccd\nAuthor: Jonathan Ellis \nDate: Thu May 25 08:05:41 2023 -0500\n\n optimize ConcurrentVectorValues.write\n\ncommit 456fb08af0b922922ea0e4f24def0184cf4b32c6\nAuthor: Jonathan Ellis \nDate: Wed May 24 21:25:52 2023 -0500\n\n fix NPE better\n\ncommit c01cacf30025d181a2d86c4632042449396333bc\nAuthor: Zhao Yang \nDate: Thu May 25 08:53:38 2023 +0800\n\n Return negative ordinal if row id is not found; add failing test for null vector\n\ncommit 009fadf2ed889bcc6748f500b7fa4b1d43e910df\nAuthor: Jonathan Ellis \nDate: Wed May 24 18:00:23 2023 -0500\n\n search() errors out when an empty graph is passed to it, so special-case that\n\ncommit 8c9972d7453fad46f55e6fe5e7538442235e7fd6\nAuthor: Jonathan Ellis \nDate: Wed May 24 17:46:50 2023 -0500\n\n use serializer dimensions instead of trying to cache it from the first vector added, because we might need it before vectors are added (if someone tries to search an empty graph)\n\ncommit 8423541c4997462b65fe8f1231b0baf15dab3297\nAuthor: Jonathan Ellis \nDate: Wed May 24 17:42:30 2023 -0500\n\n fix NPEs when nulls are inserted\n\ncommit d4510e0f1934f03f66d89b7c585b3d7aac17dda2\nAuthor: Mike Adamson \nDate: Wed May 24 15:54:38 2023 +0100\n\n Rebuild HNSW graph on flush since we don't know ahead of time where UCS will want to split the range boundaries\n\ncommit 6b925531011798285a6b25d70961643092e2febe\nAuthor: Mike Adamson \nDate: Mon May 22 14:58:22 2023 +0100\n\n disable index segment compaction by default\n\ncommit 3480a80e3303c13b8c04f718f83f87489315e381\nAuthor: Jonathan Ellis \nDate: Wed May 24 14:46:03 2023 -0500\n\n move chatty logs to debug level\n\ncommit 05c5758244f6f6d4fe30fbcefbb13932f68f8891\nAuthor: Jonathan Ellis \nDate: Wed May 24 13:59:01 2023 -0500\n\n r/m fixme obsoleted by #624\n\ncommit b09db4e0425c92491e3e9c73ebdbdb84508075fd\nAuthor: Jonathan Ellis \nDate: Wed May 24 13:56:46 2023 -0500\n\n pre-size the intersection builder\n\ncommit 6d62d8c2546cda1f487c14dcde1303e14a18c804\nAuthor: Jonathan Ellis \nDate: Wed May 24 13:45:14 2023 -0500\n\n fix nested boolean expressions\n\ncommit b8301a86d37513b67545883548c5670131281dc1\nAuthor: Zhao Yang \nDate: Wed May 24 14:19:41 2023 +0800\n\n Vector-3: support partition/range restricted query\n\ncommit 8ba56fad62d21da228e3a61ed7d5971441572cf9\nAuthor: Jonathan Ellis \nDate: Wed May 24 10:35:16 2023 -0500\n\n update lucene version hash\n\ncommit a4768c583eb1a029f629e5ebda274ab2fcce3130\nAuthor: Jeremiah D Jordan \nDate: Tue May 23 10:45:02 2023 -0500\n\n Revert the revert of \"Use published lucene fork jar.\"\"\n\n This reverts commit 7a8ff12860a97d451045c9dd1978a3599a0f4679.\n\ncommit 063a176caeec0f1e298d044490aa29ecdd468811\nAuthor: Zhao Yang \nDate: Wed May 24 22:03:32 2023 +0800\n\n Union the results from multiple segments per index (#624)\n\n - Union the results from multiple segments per index\n - Revert intersection behavior to pick 2 most selective indexes if there is no ANN\n - Fix SSTableRowIdPostingList to return END_OF_STREAM if next row is END_OF_STREAM\n - Skip TermTree for vector index\n - Fixes:\n - VectorMemtableIndexTest#randomQueryTest\n - SegmentMergerTest\n - SingleNodeQueryFailureTest#testFailedRangeIteratorOnMultiIndexesQuery\n - SelectiveIntersectionTest\n - QueryTimeoutTest\n\ncommit 643f34c2b654b80573ba92381e386b4286f7fb63\nAuthor: Jonathan Ellis \nDate: Wed May 24 08:25:35 2023 -0500\n\n switch all the hnsw internals to use float[] so we don't need to keep both representations around\n\ncommit 7a8ff12860a97d451045c9dd1978a3599a0f4679\nAuthor: Jonathan Ellis \nDate: Wed May 24 08:06:28 2023 -0500\n\n Revert \"Use published lucene fork jar.\"\n\n This reverts commit 53c6c90920dd2a48fc3334a6cddb215d3f4ebfcd.\n\ncommit 8731eefe4122a5af0b3723a3b07884a201616e7a\nMerge: e50d5e955b 53c6c90920\nAuthor: Jonathan Ellis \nDate: Wed May 24 08:05:18 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit e50d5e955bfde6357a5bb892b2ea70d46e8cc77d\nAuthor: Jonathan Ellis \nDate: Wed May 24 08:02:52 2023 -0500\n\n cache float[] from ByteBuffer\n\ncommit 53c6c90920dd2a48fc3334a6cddb215d3f4ebfcd\nAuthor: Jeremiah D Jordan \nDate: Tue May 23 18:29:00 2023 -0500\n\n Use published lucene fork jar.\n\ncommit f2567cd2fb2a683e7c0fee23945baac3ceca6ad4\nAuthor: Jonathan Ellis \nDate: Tue May 23 14:56:25 2023 -0500\n\n attempting to add back IN support but with limited success, even \"basicOrTest\" gets parsed as child nodes\n\ncommit 9e1f018f5c0d5832446fc6a0338aaf95c8323303\nAuthor: Jonathan Ellis \nDate: Tue May 23 14:21:23 2023 -0500\n\n reduce log level of some of the chattiest locations\n\ncommit efaf501fe3158da42dcd874b06014acb127f5cd9\nAuthor: Jonathan Ellis \nDate: Tue May 23 13:55:40 2023 -0500\n\n clean up RangeIntersection Builder overloads\n\ncommit e7127d9d32392347f7010de6267521742f831217\nAuthor: Jonathan Ellis \nDate: Tue May 23 13:46:27 2023 -0500\n\n complete generification of RangeIterator in src/ (but tests are still incomplete)\n\ncommit c07ded6420aa25b9dfa2b3c0785178b08d286b26\nAuthor: Jonathan Ellis \nDate: Tue May 23 13:04:16 2023 -0500\n\n TermIterator is more or less replaced by CheckpointingIterator\n\ncommit 14d03ae99c69c2419dc0cca3291b4f5eaefbc054\nAuthor: Jonathan Ellis \nDate: Tue May 23 14:45:18 2023 -0500\n\n update lucene jar\n\ncommit a15b964d56dcba285c6871d354755d7aa489af1d\nAuthor: Jonathan Ellis \nDate: Tue May 23 12:48:41 2023 -0500\n\n disk and memory recalls should be the same (since disk graph should be exactly equivalent to in-memory)\n\ncommit f03680bdf8581fd13a2d772e8265e3c8a713d247\nMerge: 868a6b8f11 fd1293dbaf\nAuthor: Jonathan Ellis \nDate: Tue May 23 12:08:46 2023 -0500\n\n Merge branch 'vsearch' of github.com:datastax/cassandra into vsearch\n\ncommit 868a6b8f11a25c20b7e59a93d65cc7e2b021a432\nAuthor: Jonathan Ellis \nDate: Tue May 23 12:04:11 2023 -0500\n\n use brute force if we expect to perform fewer comparisons that way than with a graph search\n\ncommit 2ed32bf53acaf8180130bef1aa5cebd9b6d32d13\nAuthor: Jonathan Ellis \nDate: Tue May 23 11:59:11 2023 -0500\n\n refactor AnnKeyRangeIterator -> ReorderingRangeIterator\n\ncommit d28bad9b2df3fc84b117f9ed0a04314ecb9c0a02\nAuthor: Jonathan Ellis \nDate: Tue May 23 11:55:15 2023 -0500\n\n rename reorderOneComponent to limitToTopResults\n\ncommit fd1293dbaf82f054593b3150249ac00110b286cf\nAuthor: Jeremiah D Jordan \nDate: Tue May 23 10:45:02 2023 -0500\n\n Update python driver to one with supports the Vector type\n\ncommit 6603f7601671357928948dfb0921b0c731bb96d3\nAuthor: Jonathan Ellis \nDate: Tue May 23 11:06:12 2023 -0500\n\n extract ReorderingPostingList\n\ncommit 94b4ff20ccc3a51e966b4798f5364a2d9ed4b966\nAuthor: Jonathan Ellis \nDate: Tue May 23 10:29:33 2023 -0500\n\n deserialize doesn't modify buffer so duplicate is unnecessary\n\ncommit 840053d212985a109cb5242c19e8a5876fdec7b3\nAuthor: Jonathan Ellis \nDate: Tue May 23 10:26:57 2023 -0500\n\n use LongHeap instead of generic PQ\n\ncommit aac90dc3e0fdb745e465f044f2624228e705be10\nAuthor: Jonathan Ellis \nDate: Tue May 23 09:47:22 2023 -0500\n\n r/m unused defer flag from toPrimaryKeyIterator / toSSTableRowIdsIterator\n\ncommit 1fd71c1cb7f2e785237b36c6e42842fb326d2380\nAuthor: Jonathan Ellis \nDate: Tue May 23 09:26:03 2023 -0500\n\n remove unused method marked \"to remove\"\n\ncommit b96c5a9efccf3226acf2f2a08ca149d0acc0b21d\nMerge: 54f41e32d3 ba8aa07dd0\nAuthor: Jonathan Ellis \nDate: Tue May 23 09:24:44 2023 -0500\n\n Merge remote-tracking branch 'zhao/vsearch-row-id-iterator-for-reordering' into vsearch\n\ncommit 54f41e32d3359941ddf70d34fc8e68a842b9091d\nAuthor: Jonathan Ellis \nDate: Tue May 23 08:14:10 2023 -0500\n\n update lucene to fix NPE in ram usage\n\ncommit ba8aa07dd09f52a72e5dffccb950a21ec37a1c11\nAuthor: Zhao Yang \nDate: Tue May 23 16:51:36 2023 +0800\n\n intersect on row-ids before fetching primary key for multiple non-ann indexes or single index\n\ncommit 2a69547167f8dc1b40e8a26b71cdf43a45779280\nAuthor: Zhao Yang \nDate: Tue May 23 09:44:00 2023 +0800\n\n Use sstable-row-id iterator to re-order ann index\n\n - add searchSSTableRowIds to searcher\n\ncommit 04b9a606d9a1aaeeeab3ec0981a5f1300cc2a423\nAuthor: Zhao Yang \nDate: Tue May 23 00:30:12 2023 +0800\n\n Make RangeIterator generic\n\ncommit 4ff716a7411d3ad80009d4fb97bf53af51dd5eba\nAuthor: Jonathan Ellis \nDate: Mon May 22 14:57:34 2023 -0500\n\n r/m obsolete FIXMEs\n\ncommit 175ad0edce9cef3fb29f230fb64fb1913328d101\nAuthor: Zhao Yang \nDate: Tue May 23 00:30:12 2023 +0800\n\n fix vector distributed test: use 1 token and ignore multi-ann-index test (#622)\n\ncommit 78adf77d9e6b7f20ece71fe665a5d1aec51f79b6\nAuthor: Mike Adamson \nDate: Mon May 22 14:26:48 2023 +0100\n\n Copy lucene snapshot to build folder in build-resolver.xml\n\ncommit 5172315dcb51e05a44af224ddfe944bca29005d0\nAuthor: Jonathan Ellis \nDate: Mon May 22 08:17:41 2023 -0500\n\n declare lucene dependency to come from lib/\n\ncommit d4ee45b8675f79b4a1e06f97d0ce618ab08f030a\nAuthor: Jonathan Ellis \nDate: Sun May 21 05:40:30 2023 -0500\n\n add caching and test code\n\ncommit 5102d96e2fb3460f1e541cc1b481e224d112eac1\nMerge: 6afcfabe88 87d39a7b25\nAuthor: Jonathan Ellis \nDate: Sat May 20 11:15:14 2023 -0500\n\n merge cep-vsearch\n\ncommit 6afcfabe883c5627c96eda8eb488c51c06e17c91\nMerge: a993dbe49a fa85a191c5\nAuthor: Jonathan Ellis \nDate: Sat May 20 09:15:13 2023 -0500\n\n Merge commit 'fa85a191' into vsearch\n\ncommit a993dbe49a75dde4b9d5528e0816b81ae427a1dd\nAuthor: Jonathan Ellis \nDate: Fri May 19 22:00:31 2023 -0700\n\n FIXME not sure why version is messed up post merge, this hacks around it\n\ncommit aadf168e47a4271cc3faeb62ec7044cbfe193893\nMerge: c217120935 5bc4d4b42e\nAuthor: Jonathan Ellis \nDate: Sat May 20 08:52:48 2023 -0500\n\n merge 5bc4d4b42e\n\ncommit c2171209350cc307f3b5fcc73a6ec09423d203df\nAuthor: Jonathan Ellis \nDate: Fri May 19 22:00:31 2023 -0700\n\n FIXME not sure why version is messed up post merge, this hacks around it\n\ncommit 2e44491d3db6a881324126490e6e1511d6082f39\nMerge: 68f85f7009 0ff4566080\nAuthor: Jonathan Ellis \nDate: Sat May 20 08:15:16 2023 -0500\n\n merge 0ff4566080\n\ncommit 68f85f700995b274ec0d64654ed2692c7fc8bf61\nMerge: c8843779be 2ae60e7411\nAuthor: Jonathan Ellis \nDate: Sat May 20 07:50:16 2023 -0500\n\n Merge commit '2ae60e74' into vsearch\n\ncommit c8843779be9fe0c3edd047754bb08f633cff5646\nMerge: 3cb57548bb 93616d080c\nAuthor: Jonathan Ellis \nDate: Sat May 20 07:43:18 2023 -0500\n\n Merge commit '93616d080c539e880e37cd22fe5f27396a7a2594' into vsearch\n\ncommit 87d39a7b252b3f3cc1f4aae0f149085f6bc83abc\nAuthor: Jonathan Ellis \nDate: Fri May 19 17:24:55 2023 -0700\n\n fix intersection count, and add comments to KRI api\n\ncommit 4fdc58fced9ca350333675cb19a42169bebeaccf\nAuthor: Jonathan Ellis \nDate: Fri May 19 15:27:34 2023 -0700\n\n test recall\n\ncommit 26fb1006430513a37f267a4dfb81ef473516ac7b\nAuthor: Jonathan Ellis \nDate: Fri May 19 14:19:08 2023 -0700\n\n add tests\n\ncommit 1cbf78bbf71fff2d597ddd199fd3a758313c7bf4\nAuthor: Jonathan Ellis \nDate: Fri May 19 14:02:36 2023 -0700\n\n comments\n\ncommit 65e41490f041535f1a48a726032e28f6c8febcff\nAuthor: Jonathan Ellis \nDate: Fri May 19 13:57:04 2023 -0700\n\n rm AnnResult\n\ncommit 06b10a28c3f6f19c76b056526cdabb4c9aff1528\nAuthor: Jonathan Ellis \nDate: Fri May 19 13:54:08 2023 -0700\n\n cleanup commented-out code\n\ncommit e33236ab8b9da7a025f4c59406c52ecbc3b5ae4f\nAuthor: Jonathan Ellis \nDate: Fri May 19 09:52:50 2023 -0700\n\n convert suspicious one-off binary search in OnDiskHnswGraph to use DiskBinarySearch\n\ncommit fbd30bce13e87c63b129db26c5c123ae9bf8f9ed\nAuthor: Jonathan Ellis \nDate: Fri May 19 09:31:16 2023 -0700\n\n perform disk-based binary search for ordinals\n\ncommit 77bb1b5cfcc743224bdb01d21563365caa184c21\nAuthor: Jonathan Ellis \nDate: Fri May 19 08:22:36 2023 -0700\n\n move OnDiskOrdinalsMap to its own class, and move write() code into it\n\ncommit 8a1fd9f0a38f4c16b72d5bc7b2aeb3ee07bacc7a\nAuthor: Jonathan Ellis \nDate: Thu May 18 21:35:28 2023 -0700\n\n add ordinal to row mapping\n\ncommit db32d9c7b077133a213ac3a0da161f163c4310a1\nAuthor: Jonathan Ellis \nDate: Wed May 17 20:43:49 2023 -0500\n\n mostly-working reordering\n\ncommit 62ba004d79aa7511f519ebb56fa8af2cb424e21e\nAuthor: Jonathan Ellis \nDate: Wed May 17 20:13:10 2023 -0500\n\n turn QueryView inside out; perform intersections first for each sstable, then union the results\n\ncommit d65039a6ccd8364e8eb8ef31933cdf8dca298437\nAuthor: Jonathan Ellis \nDate: Wed May 17 18:37:07 2023 -0500\n\n update MIM to give individual iterators back\n\ncommit a190f1e917b93b6cabd15347dec55b1683807b1f\nAuthor: Jonathan Ellis \nDate: Wed May 17 18:59:13 2023 -0500\n\n refactor CheckpointingIterator to just take an iterator to wrap, and index references to close if something goes wrong\n\ncommit 307fe1c20b400e17048a356f73822ef51f66f341\nAuthor: Jonathan Ellis \nDate: Wed May 17 18:47:20 2023 -0500\n\n rename IndexSearchResultIterator -> CheckpointingIterator\n\ncommit 690751b216afbe3e77ed0f89ce04d16e5c6e985d\nMerge: 4b6d0f2fb5 fa85a191c5\nAuthor: Jonathan Ellis \nDate: Wed May 17 10:17:49 2023 -0500\n\n Merge branch 'cep-7-sai' into cep-vsearch\n\ncommit 4b6d0f2fb531f6832d2b24465b0982683e741dfe\nAuthor: Zhao Yang \nDate: Wed May 17 16:45:42 2023 +0800\n\n Add vector top-k filter at replica side before returning to coordinator (#618)\n\n * added QueryPlan#postIndexQueryProcessor to filter top-k at replica side before sending response to coordinator\n\ncommit 5bc4d4b42e77180df050bdbd9f5fd18fbb37332c\nAuthor: Jonathan Ellis \nDate: Tue May 16 07:38:45 2023 -0500\n\n failing test was failing b/c test was wrong\n\ncommit f37e40f5c5d99aeaafb33f41be0bd10bff6b5756\nMerge: ae1bffa49b 1c41a095e1\nAuthor: Jonathan Ellis \nDate: Mon May 15 21:24:47 2023 -0500\n\n merge\n\ncommit ae1bffa49b4c8757a3d9110ae065b8cb70d8e416\nMerge: bbc4f72b8d 381e04aaa1\nAuthor: Jonathan Ellis \nDate: Mon May 15 21:22:16 2023 -0500\n\n Merge commit '381e04aaa132cf7a255efad3f792852d9ec1729d' into cep-vsearch\n\ncommit 1c41a095e130bce4c7c215650f9ef3f817ae635e\nAuthor: Jonathan Ellis \nDate: Mon May 15 21:11:47 2023 -0500\n\n fix computeNext to return endOfData when done\n\ncommit b845395ad7e5f4670597cd171f28b98c941d03a7\nAuthor: Jonathan Ellis \nDate: Mon May 15 21:11:15 2023 -0500\n\n cleanup\n\ncommit 58df11dad894aa5fb36bfbdaaf4c7dc1987fcc94\nAuthor: Jonathan Ellis \nDate: Mon May 15 20:59:12 2023 -0500\n\n failing tests\n\ncommit ea5977620ebb07d220ad51800c9b90f470494079\nAuthor: Jonathan Ellis \nDate: Mon May 15 20:58:52 2023 -0500\n\n FIXME temporarily remove checkstyle\n\ncommit 715790922379ad7b5dbdea5d3358f24542762889\nAuthor: Jonathan Ellis \nDate: Mon May 15 20:49:10 2023 -0500\n\n cleanup\n\ncommit 19e47d363c856f81bda3323be02bfe18f60b51c0\nAuthor: Jonathan Ellis \nDate: Mon May 15 17:55:52 2023 -0500\n\n update NeighborSet usage\n\ncommit 920abafce00c94bbc7d6b0c86ba78a2a9d3f9540\nAuthor: Jonathan Ellis \nDate: Mon May 15 17:53:43 2023 -0500\n\n cleanup\n\ncommit d58db09004427506efe03e81364cadf44fc914c4\nAuthor: Jonathan Ellis \nDate: Mon May 15 15:46:06 2023 -0500\n\n pull in latest hnsw optimizations\n\ncommit bbc4f72b8d7d99db003103103c90d78db4ee04d1\nAuthor: Jonathan Ellis \nDate: Mon May 15 10:02:40 2023 -0500\n\n use IndexOutputWriter in other components\n\ncommit 381e04aaa132cf7a255efad3f792852d9ec1729d\nAuthor: Jonathan Ellis \nDate: Mon May 15 09:32:25 2023 -0500\n\n switch other components to also use IndexOutputWriter\n\ncommit 5951690f5472502bcb093179e8905212c1fcf815\nAuthor: Mike Adamson \nDate: Mon May 15 14:48:16 2023 +0100\n\n Fix CassandraHnswGraphWriter to use correct output writer (#616)\n\ncommit bc5b956607fa77d53fcab9ad338776af69610feb\nAuthor: Zhao Yang \nDate: Mon May 15 17:35:42 2023 +0800\n\n Distributed vector search (#613)\n\n * distributed vector search\n\n - query all replicas selected by consistency level at once with full request range\n\n - filter top-K results at coordinator in QueryPlan#postProcessor\n\n - skip short-read-protection, read repair and replica filtering protection because replica response will be top-k\n\n - fail ANN query without limit or limit exceeding MAX_TOP_K\n\n - make vector search max_top_k configurable and default to 1k\n\ncommit 6d8c94535354f162f60099deecb0523f9f71bb99\nAuthor: Jonathan Ellis \nDate: Sun May 14 10:01:08 2023 -0500\n\n fix imports for checkstyle\n\ncommit 68c20dc9ceeaf0c2169718aa4a1dce9c0f92f003\nAuthor: Jonathan Ellis \nDate: Sat May 13 19:23:58 2023 -0500\n\n add the index and ordinal mapping to the set of components that the system knows about\n\ncommit ba7efaac142c8b3181881f86020f659837c90463\nAuthor: Jonathan Ellis \nDate: Sat May 13 17:57:22 2023 -0500\n\n cleanup\n\ncommit 5ae06fe31840e5ba1899086f3b8eb8194f825db7\nAuthor: Jonathan Ellis \nDate: Sat May 13 17:23:30 2023 -0500\n\n update to latest lucene snapshot that addresses all known concurrency bugs\n\ncommit f996edd63f66cf7b688f8e014b31f15450dd1d3d\nAuthor: Jonathan Ellis \nDate: Fri May 12 16:08:40 2023 -0500\n\n fix for cqlsh by Bret McGuire\n\ncommit fa85a191c5e0bd508da584412648308888769cb9\nAuthor: Andrés de la Peña \nDate: Mon Apr 24 16:39:26 2023 +0100\n\n Allow CQL queries on multiple indexes without ALLOW FILTERING\n\n patch by Andrés de la Peña; reviewed by Berenguer Blasi for CASSANDRA-18217\n\ncommit e2f3d2150ab2bc5e038f2661b8e2337d3b5cd4bf\nAuthor: Jonathan Ellis \nDate: Thu May 11 14:34:22 2023 -0500\n\n imports\n\ncommit 68e59bc715de6d1f69d69530310655a598a54200\nAuthor: Mike Adamson \nDate: Thu May 11 17:20:52 2023 +0100\n\n Use correct bind types for vector\n\ncommit c57119254b7e065433804e36ed6ead71d53984ea\nAuthor: Jonathan Ellis \nDate: Wed May 10 23:06:19 2023 -0500\n\n avoid calling neighborSet.size() in write path\n\ncommit f91e9ea40fa4bfc07c0bb280b7c479d473956da3\nAuthor: Jonathan Ellis \nDate: Wed May 10 23:05:11 2023 -0500\n\n add more asserts. countNeighbors is failing\n\ncommit 1bd6789304a60762f0ef497c2137dd96f806a262\nAuthor: Jonathan Ellis \nDate: Wed May 10 22:40:01 2023 -0500\n\n lock out updates to the graph while we're writing it to disk\n\ncommit da7cb2984f96c06a5890ed2da21c42d946d7c754\nAuthor: Jonathan Ellis \nDate: Wed May 10 22:06:58 2023 -0500\n\n imports\n\ncommit 3676a86b821c6c05d9c5df33f4e4fbc9eeb1f591\nAuthor: Jonathan Ellis \nDate: Wed May 10 22:03:14 2023 -0500\n\n compact by building graph in memory, like it did before\n\ncommit fa1ebd013bc03def3d211932fe2a98d1bb4442d9\nAuthor: Jonathan Ellis \nDate: Wed May 10 16:15:58 2023 -0500\n\n rewrite save-to-disk without Lucene code, to support multiple rows having the same vector value\n\n bonus: we don't need to rebuild an index that we already have in memory\n\ncommit 3b328324d7c3afcda34a95c4b9fcbb9fe017384c\nAuthor: Jonathan Ellis \nDate: Wed May 10 17:00:02 2023 -0500\n\n upgrade to latest lucene snapshot\n\ncommit 1f4e768e9ec3327a56d143fa17c86…","shortMessageHtmlLink":"Squashed commit of the following:"}},{"before":"02b32015d7c2ca8d75fd966e6f62347483e0df39","after":"59f774d6e76d281050100005dc159281390fc911","ref":"refs/heads/todo-cleanup","pushedAt":"2024-05-16T14:27:42.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"jbellis","name":"Jonathan Ellis","path":"/jbellis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/42158?s=80&v=4"},"commit":{"message":"fix NPE","shortMessageHtmlLink":"fix NPE"}},{"before":"692b7a10abfb00781a24f5599345fee3eaa9ab0f","after":"9de62fcc76979eef6cee0a58b9f21ce4f6019cfa","ref":"refs/heads/cndb-9487","pushedAt":"2024-05-16T10:27:38.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"sbtourist","name":"Sergio Bossa","path":"/sbtourist","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/103671?s=80&v=4"},"commit":{"message":"Minor review changes.","shortMessageHtmlLink":"Minor review changes."}},{"before":"1bba5f701321fef42bde8f6e8c86ff1b3b4172e9","after":"0d5789722faf36eaa689ea714a03613c26ed750d","ref":"refs/heads/cc-ucu-intenode-sb","pushedAt":"2024-05-16T10:12:53.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"sbtourist","name":"Sergio Bossa","path":"/sbtourist","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/103671?s=80&v=4"},"commit":{"message":"Added SensorsInternodeTest + bug fixes.","shortMessageHtmlLink":"Added SensorsInternodeTest + bug fixes."}},{"before":null,"after":"02b32015d7c2ca8d75fd966e6f62347483e0df39","ref":"refs/heads/todo-cleanup","pushedAt":"2024-05-16T00:33:58.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jbellis","name":"Jonathan Ellis","path":"/jbellis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/42158?s=80&v=4"},"commit":{"message":"write postings in a separate thread while we cleanup()","shortMessageHtmlLink":"write postings in a separate thread while we cleanup()"}},{"before":"730701acc9ff08531fedf58635364bba746cbb19","after":null,"ref":"refs/heads/fix-flaky-softlimitutiltest","pushedAt":"2024-05-15T20:38:10.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"jkni","name":"Joel Knighton","path":"/jkni","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5061375?s=80&v=4"}},{"before":"1457b91dbd76ea1642ef151bcd6174f125148d41","after":"478ed37f6322128b5837ec826096bd35d56be25e","ref":"refs/heads/vsearch","pushedAt":"2024-05-15T20:38:09.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"jkni","name":"Joel Knighton","path":"/jkni","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5061375?s=80&v=4"},"commit":{"message":"Fix flaky test SoftLimitUtilTest.randomized\n\nFix SoftLimitUtilTest.randomized when softLimit is capped at Integer.MAX_VALUE. Fix assert messages for upper bound. (#1120)","shortMessageHtmlLink":"Fix flaky test SoftLimitUtilTest.randomized"}},{"before":"8aa88ca2734c355700213e0805d96e607f35d6ab","after":"1bba5f701321fef42bde8f6e8c86ff1b3b4172e9","ref":"refs/heads/cc-ucu-intenode-sb","pushedAt":"2024-05-15T18:33:06.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"sbtourist","name":"Sergio Bossa","path":"/sbtourist","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/103671?s=80&v=4"},"commit":{"message":"Opinionated changes to internode tracking:\n* Removed message count as bytes should really be the most important/accurate.\n* Tracked both inbound and outbound bytes, as writes take up inbound bytes, while reads take up outbound bytes.\n* Added internode sensors to custom params in the same place where we add the other sensors.\n* Added internode sensors for request.","shortMessageHtmlLink":"Opinionated changes to internode tracking:"}},{"before":null,"after":"8aa88ca2734c355700213e0805d96e607f35d6ab","ref":"refs/heads/cc-ucu-intenode-sb","pushedAt":"2024-05-15T18:28:37.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"sbtourist","name":"Sergio Bossa","path":"/sbtourist","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/103671?s=80&v=4"},"commit":{"message":"Opinionated changes to internode tracking:\n* Removed message count as bytes should really be the most important/accurate.\n* Tracked both inbound and outbound bytes, as writes take up inbound\nbytes, while reads take up outbound bytes.\n* Added internode sensors to custom params in the same place where we add the other sensors.\n* Added internode sensors for request.","shortMessageHtmlLink":"Opinionated changes to internode tracking:"}},{"before":null,"after":"730701acc9ff08531fedf58635364bba746cbb19","ref":"refs/heads/fix-flaky-softlimitutiltest","pushedAt":"2024-05-15T18:11:32.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jkni","name":"Joel Knighton","path":"/jkni","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5061375?s=80&v=4"},"commit":{"message":"Fix SoftLimitUtilTest.randomized when softLimit is capped at Integer.MAX_VALUE. Fix assert messages for upper bound.","shortMessageHtmlLink":"Fix SoftLimitUtilTest.randomized when softLimit is capped at Integer.…"}},{"before":"a472efd6a20b040250f67f98b93e4602007e72fc","after":null,"ref":"refs/heads/DSP-24218-fix-group-completion-marker","pushedAt":"2024-05-15T18:02:08.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"jkni","name":"Joel Knighton","path":"/jkni","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5061375?s=80&v=4"}},{"before":"02ba08b06cd469882ba2666afbbf669724c6e003","after":"1457b91dbd76ea1642ef151bcd6174f125148d41","ref":"refs/heads/vsearch","pushedAt":"2024-05-15T18:02:08.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"jkni","name":"Joel Knighton","path":"/jkni","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5061375?s=80&v=4"},"commit":{"message":"DSP-24218: Close writers before creating the completion marker (#1118)\n\nClosing writers after creating the completion marker opened a tiny race\r\nwindow where other things might consider the index ready, when it was still being\r\nwritten.\r\n\r\nThis bug was causing CI failures in MultiMapIntTest and other SAI map tests\r\nin DSE 6.9.","shortMessageHtmlLink":"DSP-24218: Close writers before creating the completion marker (#1118)"}},{"before":"bdf559a51f8ace1b98f0ad82ad01868263a66bfa","after":"a472efd6a20b040250f67f98b93e4602007e72fc","ref":"refs/heads/DSP-24218-fix-group-completion-marker","pushedAt":"2024-05-15T16:05:14.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pkolaczk","name":"Piotr Kołaczkowski","path":"/pkolaczk","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1352795?s=80&v=4"},"commit":{"message":"DSP-24218: Close writers before creating the completion marker\n\nClosing writers after creating the completion marker opened a tiny race\nwindow where other things might consider the index ready, when it was still being\nwritten.\n\nThis bug was causing CI failures in MultiMapIntTest and other SAI map tests\nin DSE 6.9.","shortMessageHtmlLink":"DSP-24218: Close writers before creating the completion marker"}},{"before":"8cbe513f4968df84c6b59ab250630d3d5d28ac6c","after":"692b7a10abfb00781a24f5599345fee3eaa9ab0f","ref":"refs/heads/cndb-9487","pushedAt":"2024-05-15T15:41:23.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"sbtourist","name":"Sergio Bossa","path":"/sbtourist","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/103671?s=80&v=4"},"commit":{"message":"CNDB-9487 RequestSensors#syncAllSensors() is now idempotent and properly increments sensors with their delta values.","shortMessageHtmlLink":"CNDB-9487 RequestSensors#syncAllSensors() is now idempotent and prope…"}},{"before":null,"after":"8cbe513f4968df84c6b59ab250630d3d5d28ac6c","ref":"refs/heads/cndb-9487","pushedAt":"2024-05-15T15:38:45.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"sbtourist","name":"Sergio Bossa","path":"/sbtourist","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/103671?s=80&v=4"},"commit":{"message":"CNDB-9487 RequestSensors#syncAllSensors() is now idempotent and properly increments sensors with their delta values.","shortMessageHtmlLink":"CNDB-9487 RequestSensors#syncAllSensors() is now idempotent and prope…"}},{"before":"e3eff6e2851722daee3b1d74a1187ae0f4544145","after":"a5da4befc32e97fcade092e9b05260e25e9fa105","ref":"refs/heads/immutable-sai-components","pushedAt":"2024-05-15T15:15:30.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"pcmanus","name":"Sylvain Lebresne","path":"/pcmanus","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/479957?s=80&v=4"},"commit":{"message":"Add test for immmutable components","shortMessageHtmlLink":"Add test for immmutable components"}},{"before":"616b6dac92b954ac56655fb4bbd2fa9598720736","after":"55259305a61c1e6fe09a05884bb270cd291300dc","ref":"refs/heads/CNDB-9099-STAR-1907","pushedAt":"2024-05-15T13:53:13.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jacek-lewandowski","name":"Jacek Lewandowski","path":"/jacek-lewandowski","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6516951?s=80&v=4"},"commit":{"message":"Fix the remaining issues with ZCS","shortMessageHtmlLink":"Fix the remaining issues with ZCS"}},{"before":"111430e444849dcf852e81f6a10c45ba24af9275","after":"e3eff6e2851722daee3b1d74a1187ae0f4544145","ref":"refs/heads/immutable-sai-components","pushedAt":"2024-05-15T12:32:24.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pcmanus","name":"Sylvain Lebresne","path":"/pcmanus","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/479957?s=80&v=4"},"commit":{"message":"Disable immutable components by default","shortMessageHtmlLink":"Disable immutable components by default"}},{"before":null,"after":"bdf559a51f8ace1b98f0ad82ad01868263a66bfa","ref":"refs/heads/DSP-24218-fix-group-completion-marker","pushedAt":"2024-05-15T08:30:30.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"pkolaczk","name":"Piotr Kołaczkowski","path":"/pkolaczk","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1352795?s=80&v=4"},"commit":{"message":"DSP-24218: Close writers before creating the completion marker\n\nClosing writers after creating the completion marker opened a tiny race\nwindow where other things might consider the index ready, when it was still being\nwritten.\n\nThis bug was causing CI failures in MultiMapIntTest and other SAI map tests\nin DSE 6.9.","shortMessageHtmlLink":"DSP-24218: Close writers before creating the completion marker"}},{"before":"62594dead1b870fb79dd543cf9c29799d5cb6e7b","after":null,"ref":"refs/heads/jvector3-enhance","pushedAt":"2024-05-15T03:04:34.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"jbellis","name":"Jonathan Ellis","path":"/jbellis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/42158?s=80&v=4"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAETVaYswA","startCursor":null,"endCursor":null}},"title":"Activity · datastax/cassandra"}