+1 on the quality/speed of the connection you test from making a big difference. I was testing a new server last week and got vastly different results from office #1 vs office #2 because one had a much lower bandwidth and made the server being tested look pathetic.
On Saturday, April 6, 2013 8:13:44 AM UTC-5, Niphlod wrote: > > wait a sec. you're launching ab from your "home line" to a server hosted > somewhere ? > Launch it from the server itself, or another server in the same facility ! > You're reading back results that are dependant on all the loops of the net > wired from your "home" to the server :D > > These is my benchmark from Italy to a linode 512 in london, web2py + > postgresql + nginx 1.2.3. > Sadly, my home connection goes as far as 300 KB/sec > > Concurrency Level: 20 > Time taken for tests: 25.609 seconds > Complete requests: 500 > Failed requests: 0 > Write errors: 0 > Total transferred: 6635000 bytes > HTML transferred: 6431000 bytes > Requests per second: 19.52 [#/sec] (mean) > Time per request: 1024.363 [ms] (mean) > Time per request: 51.218 [ms] (mean, across all concurrent requests) > Transfer rate: 253.02 [Kbytes/sec] received > > Connection Times (ms) > min mean[+/-sd] median max > Connect: 292 597 460.6 444 7359 > Processing: 226 405 177.8 349 1432 > Waiting: 226 405 177.8 349 1432 > Total: 606 1002 507.7 807 7666 > > Percentage of the requests served within a certain time (ms) > 50% 807 > 66% 904 > 75% 1103 > 80% 1213 > 90% 1721 > 95% 2001 > 98% 2308 > 99% 2403 > 100% 7666 (longest request) > > > > -- --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to web2py+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.