What's the best way to monitor and log which processes are responsible for high system load throughout the day? Tools like top and htop only provide immediate values, but I'm looking for a solution that offers historical data to identify the main culprits over time.
Actually, it’s all self-hosted. Granted, I haven’t looked at the code in detail, but building NNs to help efficiency detect and capture stuff is actually a very appropriate use of ML. This project looks kinda cool.
I did a whole stack of servers using SNMP based monitoring years ago and it was amazing. I could see loads, memory stats, NIC utilisation, disk space, and all sorts of other things. I tried Cacti and Icinga and settled on the latter but they are all fairly similar. Once you are generating the data you van do whatever you like with it, so monitoring load attributable to which actual executable is definitely manageable. It is also handy for getting notifications for something being down, losing stability, or just being out if whack.
Look through RHEL stuff. I’m not sure if Tuna has exactly what you’re looking for, but it is the tool for detailed analysis of processes on logical cores, CPU set isolation and monitoring. RHEL has tools for everything in this area, and most are available in any other distro.
www.paessler.com/prtg/downloadWe are using this. Loving it but i think only runs on windows. Free for first 100 sensors which should be enough at home.