Skip to content

Instantly share code, notes, and snippets.

@pmclanahan

pmclanahan/lisa-perf.md

Last active August 29, 2015 14:00
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save pmclanahan/11276046 to your computer and use it in GitHub Desktop.
Save pmclanahan/11276046 to your computer and use it in GitHub Desktop.
Download ZIP
Lisa Benchmark Results
Raw
lisa-perf.md

Prod Server

1 instance

Total Processed: 100,000
Total Time:      87.128194809s
IPs per minute:  68,864.0458253

2 instances

Total Processed: 50,390
Total Time:      57.0778458118s
IPs per minute:  52,969.7636096

Total Processed: 49,610
Total Time:      55.6157798767s
IPs per minute:  53,520.7814509

Total for both

Processed:       100,000
Max time:        57.0778458118s
IPs per minute:  105,119.594383143

1 instance using redis.pipeline every IP

Total Processed: 100,000
Total Time:      60.1294310093s
IPs per minute:  99,784.7459936

1 instance using redis.pipeline ever 1000th IP

Total Processed: 100,000
Total Time:      50.1173391342s
IPs per minute:  119,719.045417

Conclusion

Running Lisas in parallel both works correctly, and improves performance dramatically. We should ask jakem for more proc and memory. Also using Redis pipelines is good. Let's do that.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment