Meetup ElasticFR #90 - Introduction to Universal Profiling

[@my_username_was_taken]

no, sorry, it's not handy. how do you look at your application code? we know kubernetes and various other system processes use a lot of cpu and run many functions. that's great. no one is going to really be able to do much about that. but programmers trying to optimize their application code... it's nice that their functions are represented here as well, but you need to sift through a pile of junk to get to them. no one cares about libc or envoy or vmlinux. filtering a view to just show stack traces for code users can actually do something about would be really nice. do you have any videos on that? having a hard time finding documentation on filtering. i also can't imagine how incredibly wasteful for storage (and cost) keeping all this unusable profiling data is. filtering out ingestion would be nice too. really really don't care about containerd...there's literally nothing i'm going to be able to do about it other than upgrade to a different version and pray someone fixed something that may be going wrong in there, but for the most part there isn't. it's just not where MOST people should be looking.