Remember to up the ulimit -n, with http keep-alive you can quickly run into the linux default of 1024 open files/sockets.
Laurence 2009/7/16 Lazy <[email protected]>: >>>> assuming that each user loads 40 files in 1 minute we get >>>> 12000*60/40=18 000 users per minute >>>> >>>> Is it possible to get half of that 18k users/per minute in real word >>>> ignoring the amounts of traffic it will generate ? >>> >>> I'd say so, but it depends on how big the data set is. If you can store it >>> in memory, varnish is ridiculously fast. I also wouldn't recommend relying >> i think it will fit in to ram, i will be a single site >> >>> on a single Varnish for more than a few thousand requests per second. If >>> something goes wrong (suddenly getting lots of misses for instance), it >>> will quickly spread. >>> >>> For comparison, I'm looking at a box with roughly 0.4 load serving >>> 2000req/s as we speak, and that's on 2xDual Core Opteron 2212. Going by >>> those numbers, it should theoretically be able to handle almost ten times >>> as many requests if Varnish scaled as a straight line. >>> >>> That'd give you roughly 18000 req/s at peak (give or take a little...) Now >>> you're talking about 8 cores, that should be 36k req/s. That's _not_ >>> unrealistic, from what we've seen in synthetic tests. If each client >>> requires 40 items, that means roughly 900 clients _per second_. Or 54k in a >>> minute. This math is all rough estimates, but the foundation is production >>> sites and real traffic patterns. >>> >>> The problem is that getting your Varnish to deal with 36k req/s is rather >>> difficult, and you quickly run into network issues and similar. And at 36k >>> req/s you can hardly take any amount of backend traffic or delays before it >>> all falls over. > > today i did some ad-hoc tests with 45 byte body, > > when I enable keep-alive I'm getting 39k req/s with 100 concurrent > gets and over 40k with 300 concurrent connections (max cpu load was > under 2 cores for varnish) > > without keep alive i'm stuck with 12k req/s, that might be end of ab's > performance in making new connections or > kernel, i tried the performance tips from the wiki, but id didn't make > a significant difference in this test, > Later i will try to use a benchmark running on another machine. > > 12k req/s is more then enough for me already so I'm happy with that > > -- > Michal Grzedzicki > _______________________________________________ > varnish-misc mailing list > [email protected] > http://projects.linpro.no/mailman/listinfo/varnish-misc > _______________________________________________ varnish-misc mailing list [email protected] http://projects.linpro.no/mailman/listinfo/varnish-misc
