[07:01:03] <wikibugs>	 10Machine-Learning-Team, 10Analytics: Configure the Hadoop cluster to use the GPUs available on some workers - https://phabricator.wikimedia.org/T276791 (10elukey) Thanks a lot for the list of use cases!   I created a couple of subtasks to deal with the Hadoop config, once done we should be able to restart fro...
[16:04:42] <elukey>	 hello hello
[16:05:19] <elukey>	 I was reviewing cache metrics for Ores and I noticed that the two Redis instances that we use are doing constant evictions
[16:05:22] <elukey>	 so I filed https://gerrit.wikimedia.org/r/c/operations/puppet/+/671186/
[16:06:03] <elukey>	 the dashboard is https://grafana.wikimedia.org/d/RLhtAw6mz/ores-redis?orgId=1&refresh=1m
[16:06:21] <elukey>	 I don't want to spend a lot of time on ORES but it looks a possible easy win just bumping the memory
[16:06:32] <elukey>	 the host has plenty of free memory
[16:06:38] <elukey>	 (not even used by page cache)
[16:07:23] <elukey>	 if it is a horrible idea let me know