Thomas, This post from Ryan Svihla has a few notes in it that may or may not be useful to you:
>If you read the original throttling Jira you can see that there is a hurry up and wait component to unthrottled compaction (CASSANDRA-2156- Compaction Throttling). Ultimately you will saturate your IO in bursts, backing up other processes and making different bottlenecks spike up a long the way, potentially causing something OTHER than compaction to get so far behind that the server becomes unresponsive (such as GC). via https://medium.com/@foundev/how-i-tune-cassandra-compaction-7c16fb0b1d99 On Mon, Jun 11, 2018 at 12:05 AM, Steinmaurer, Thomas < thomas.steinmau...@dynatrace.com> wrote: > Sorry, should have first looked at the source code. In case of 0, it is > set to Double.MAX_VALUE. > > > > Thomas > > > > *From:* Steinmaurer, Thomas [mailto:thomas.steinmau...@dynatrace.com] > *Sent:* Montag, 11. Juni 2018 08:53 > *To:* user@cassandra.apache.org > *Subject:* compaction_throughput: Difference between 0 (unthrottled) and > large value > > > > Hello, > > > > on a 3 node loadtest cluster with very capable machines (32 physical > cores, 512G RAM, 20T storage (26 disk RAID)), I’m trying to max out > compaction, thus currently testing with: > > > > concurrent_compactors: 16 > > compaction_throughput_mb_per_sec: 0 > > > > With our simulated incoming load + compaction etc., the Linux volume shows > ~ 20 Mbyte/s Read IO + 50 Mbyte/s Write IO in AVG, constantly. > > > > > > Setting throughput to 0 should mean unthrottled, right? Is this really > unthrottled from a throughput perspective and then is basically limited by > disk capabilities only? Or should it be better set to a very high value > instead of 0. Is there any semantical difference here? > > > > > > Thanks, > > Thomas > > > > The contents of this e-mail are intended for the named addressee only. It > contains information that may be confidential. Unless you are the named > addressee or an authorized designee, you may not copy or use it, or > disclose it to anyone else. If you received it in error please notify us > immediately and then destroy it. Dynatrace Austria GmbH (registration > number FN 91482h) is a company registered in Linz whose registered office > is at 4040 Linz, Austria, Freistädterstraße 313 > The contents of this e-mail are intended for the named addressee only. It > contains information that may be confidential. Unless you are the named > addressee or an authorized designee, you may not copy or use it, or > disclose it to anyone else. If you received it in error please notify us > immediately and then destroy it. Dynatrace Austria GmbH (registration > number FN 91482h) is a company registered in Linz whose registered office > is at 4040 Linz, Austria, Freistädterstraße 313 > -- *Joshua Galbraith *| Senior Software Engineer | New Relic C: 907-209-1208 | jgalbra...@newrelic.com