-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Robin,
On 5/18/2009 11:35 AM, Robin Wilson wrote: > I'm curious by your comment that Coyote/APR is performing on par with > httpd, from the results in your first message I saw it was a pretty > big difference. Or are you saying that wasn't using APR? Right. Sorry if I wasn't clear: the first test (already published in a previous message) was using the standard, non-APR Coyote connector. I'm currently running a test which shows that APR brings Tomcat back into contention. Here is the full set of numbers as far as the test has run: File Size Apache httpd TC5.5 Coyote TC 5.5 APR 4kiB 6215.20 5467.43 7700.63 8kiB 11630.17 9732.60 14054.87 16kiB 21271.92 16266.66 23027.55 32kiB 38257.77 27530.75 36269.42 64kiB 63729.12 41946.45 59215.64 128kiB 91768.23 56000.19 90059.61 256kiB 126919.12 72062.84 124735.84 512kiB 160506.81 66654.86 148727.29 1MiB 180850.63 72533.92 167928.56 2MiB 185157.29 75957.71 174630.72 4MiB 185262.91 77969.20 (data not yet available) 8MiB 185152.66 79075.54 (data not yet available) 16MiB 256399.66 80364.26 (data not yet available) 32MiB 189933.50 80243.41 (data not yet available) > Also, I'd be curious about the big disparity between the 16MiB files > and the other 1MiB-32MiB files... It looks like all of them are > relatively consistent at the KiB/sec rates you show - but suddenly > there's a huge burst of speed on the 16MiB file (for httpd). So I'd > be really curious to understand why the large disparity is evident > there. I agree. I'll be re-running the tests against httpd, anyway, so maybe we'll see it happen again. If anything, I would have expected to see a sharp /decrease/ in transfer rate at some point (like a batch job ran in the middle of the night or something). Observing the opposite is baffling to me. > I'd like to see your results using TC6.0.18 and APR as well. Patience! I'll publish everything once I've got it. > Also, just idle curiosity - does HTTPS affect the performance > difference between the two at all? Even though these are static > files, it would probably show if there are any SSL handling > differences between tomcat and apache. I'm sure it would. Once I have these stats, I could publish another set based upon HTTPS. The APR documentation suggests that if you want decent HTTPS performance in Tomcat, you should use APR instead of the Java SSL libraries. So, I suspect non-APR SSL in Tomcat to perform poorly in comparison to Apache httpd. - -chris -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAkoRiFIACgkQ9CaO5/Lv0PBBlACgtFiXzCUZo07QwaIiKTRV5UE1 59QAoLZxKky9P0OqEFgkvB59qvRBpfL2 =LNpJ -----END PGP SIGNATURE----- --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org