Replies: 2 comments
-
Dask serializes arguments using pickle (or variants thereof). If you send
a string like "myfilename.txt' then it's just serializing that string and
sending it. Dask does not know if a string that you provide as an argument
is a filename or not. It makes no special consideration for strings that
look like filenames.
…On Thu, Nov 3, 2022 at 6:48 AM odo2063 ***@***.***> wrote:
Hi Everybeeing!
Could it be that Dask distributes datafiles to the nodes?
imagine something like:
client.submit(doSomething, someFilename)
with
def doSomething(someFilename): import json data = json.load(someFilename)
Since I run a cluster file system i see low transferrates but the machine
with daskscheduler sends things over the network with it's max rate. If
dask distributes these data files how can I switch it off?
—
Reply to this email directly, view it on GitHub
<#9618>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AACKZTGPCRJHZZWIAE5IXFDWGORA5ANCNFSM6AAAAAARWBRUAQ>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
odo2063
-
Thank You very much...It was a configuration issue on my side with the clusterFS, now everything works as expected. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi Everybeeing!
Could it be that Dask distributes datafiles to the nodes?
imagine something like:
client.submit(doSomething, someFilename)
with
def doSomething(someFilename): import json data = json.load(someFilename)
Since I run a cluster file system i see low transferrates but the machine with daskscheduler sends things over the network with it's max rate. If dask distributes these data files how can I switch it off?
Beta Was this translation helpful? Give feedback.
All reactions