oxidecomputer / oxidecomputer/propolis
Sysbench memory performance substantially worse on older Linux distros
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 270
- Forks
- 42
- Avg merge
- 4d 5h
- Merged PRs (30d)
- 6
Description
The following sysbench command with different parameters was executed on a 4 vcpu / 16 GiB VM on AWS and Oxide rack2 instances on various linux distros:
sysbench memory --memory-oper=$oper --memory-block-size=$blksize --memory-scope=$scope --memory-total-size=100G --threads=4 --time=30 run
The results are documented here: https://docs.google.com/spreadsheets/d/18lUU7RIqjxklMOo5YmvwwFYXto2CGOXBz5b5bjuZxRU/edit#gid=2087813628
As highlighted in column P of the sysbench tab, the events per second are much lower on ubuntu 20.04 (focal) whereas ubuntu 22.04 (jammy) is on par or better on Oxide. For comparison, I also ran the same tests on debian 11 on Oxide and the results are consistent with those on jammy.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reproducing the listed sysbench memory command on Ubuntu 20.04, Ubuntu 22.04, and Debian 11 using the AWS or Oxide environments described. Compare events per second with the sysbench spreadsheet, then determine whether the distro-dependent performance difference has an actionable cause and a measurable resolution.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- aws, linux
- Domain
- cloud, operating-systems, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100