I’m trying to request a spark cluster in the SWAN’s configuration at start up. However it says:
“Access to the Analytix cluster is not granted. Please request access.” The given link is not working.
What am I missing?
Thanks a lot!
Thanks for reporting this issue, we will review the link to the snow request. I have enabled access to the cluster for your account, it can take up to 4hrs for everything to be setup, kindly try after that
Thank you very much. I have a question related to this issue please. I was attracted by this spark Analytix cluster because accordingly to https://indico.cern.ch/event/847492/attachments/2181923/3686486/Day%201%20Service%20presentation.pdf (slide 34) one can benefit when performing ML in large datasets.
I’m using SWAN with configuration LCG_99, 4 cores, max mem 16 GB. When running my code the kernel dies always at a certain point and I attribute this issue to the large input data files involved. There is no problem if the input files are smaller.
There is a manual for spark configuration with several options related to memory, but they are so many that I wonder if you can give me some advice.
Thanks again and best wishes,
i’m experiencing the same issue.
I was using k8s, but it seems not to be running with stack 99, only in 97a
Can you enable my access to analytix as well?
I granted access to analytix cluster and you will be able to access in 2-3h. Regarding your other question, you should be able to use ‘k8s’ cluster with 99 stack, before creating spark session please ensure you select ‘Spark3Shuffle’ configuration bundle (checkbox).