|
Benchmark questions: msg#00090web.lighttpd
Hi, I was looking at the benchmark section of the website and got a few questions: 1) Is there any benchmark data for PHP performance at concurrency higher than 1 (e.g. 100 or 400)? I noticed from the static file benchmark that while the performance of Apache stays relatively constant as concurrency increases, the performance of lighttpd tends to drop as concurrency increases. I wondered if the same trend exists for PHP performance and that at a high concurrency Apache might actually have a higher performance than lighttpd, especially given the relatively small edge in PHP performance lighttpd has at 1 concurrency (25%) 2) The CPU-tuilization benchmark shows that Apache has a higher throughput than lighttpd (e.g. at 1000 con, 100K files, Apache is 8258.31, lighttpd is 8096.55). On the other hand, the static file throughput benchmark shows lighttpd having much higher throughput than Apache. What causes the discrepancy between the 2 benchmarks? 3) Has anyone looked at Lite Speed web server at litespeedtech.com? Looks like it has similar architecture to lighttpd. Anyone done any benchmark between the 2 servers? Thanks! Andy _______________________________ Do you Yahoo!? Declare Yourself - Register online to vote today! http://vote.yahoo.com |
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| Previous by Date: | Re: PRE-RELEASE: lighttpd-1.3.1-20040922-0900.tar.gz: 00090, Steffen Weber |
|---|---|
| Next by Date: | Re: Benchmark questions: 00090, Jan Kneschke |
| Previous by Thread: | PRE-RELEASE: lighttpd-1.3.1-20040922-0900.tar.gzi: 00090, Jan Kneschke |
| Next by Thread: | Re: Benchmark questions: 00090, Jan Kneschke |
| Indexes: | [Date] [Thread] [Top] [All Lists] |
| News | FAQ | advertise |