Hi Team,


We are using spark java 3.5.3 and we have a requirement to run a batch of
millions of transactions. i.e. I am running batch of 1M with 640 executors
and each executor is having 8 cores and 16 GB memory running under
kubernetes cluster.



But our conversation is that one executor is done with completed task, it
is not going away. All executors are released at the end of job with driver
pod.



And also, any performance suggestion which can help in reducing the timing.
Currently we have 10M running with 640 executors and taking 60 plus mins to
be finished.



Let me know if need more information.



Thanks,

Shivang.

-- 


*This e-mail contains PRIVILEGED AND CONFIDENTIAL
INFORMATION intended 
solely for the use of the addressee(s). If you are not the
intended 
recipient, please notify the sender by e-mail and delete the original
message. Further, you are not to copy, disclose, or distribute this e-mail 
or
its contents to any other person and any such actions maybe unlawful*. 
This
e-mail may contain viruses. Provenir has taken every reasonable 
precaution to
minimize this risk, but is not liable for any damage you may 
sustain as a
result of any virus in this e-mail. You should carry out your 
own virus checks
before opening the e-mail or attachment. Provenir reserves 
the right to monitor
and review the content of all messages sent to or from 
this e-mail address.
Messages sent to or from this e-mail address may be 
stored on the Provenir
e-mail system.



 



Think before printing 
- 
please consider the environment 

Reply via email to