TYPO3 Core

Patch set 2 for branch master has been pushed to the review server.
It is available at https://review.typo3.org/19249

Hello,

i don't know much about the issue above. But your "ab" result is not representative at all.

10 parallel request and 100 request total on a local system is good luck on result. I/O to get all files, cache of your system.

Example here

ab -c 10 -n 100 http://www.XYZ
This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking www.XYZ (be patient).....done

Server Software:        Apache
Server Hostname:        www.XYZ
Server Port:            80

Document Path:          /
Document Length:        19849 bytes

Concurrency Level:      10
Time taken for tests:  0.297 seconds
Complete requests:      100
Failed requests:        0
Write errors:          0
Total transferred:      2018500 bytes
HTML transferred:      1984900 bytes
Requests per second:    336.25 [#/sec] (mean)
Time per request:      29.740 [ms] (mean)
Time per request:      2.974 [ms] (mean, across all concurrent requests)
Transfer rate:          6628.15 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median  max
Connect:        0    0  0.3      0      1
Processing:    21  29  5.4    28      47
Waiting:      21  28  5.4    28      47
Total:        21  29  5.4    28      48

Percentage of the requests served within a certain time (ms)
  50%    28
  66%    29
  75%    31
  80%    33
  90%    36
  95%    39
  98%    47
  99%    48
100%    48 (longest request)

Next time i get one long request "100% 1032 (longest request)" and loading time is by 1.03 sec.

Maybe you use more requests. PHP Mode Version and cache modules (apc, opcache ...) make often a difference from speed.

But in the end - Nice to see you try to speedup thing in TYPO3 core!

The test was against another notebook during the FAL code sprint with and without APC (But the results didn't show any changes). So the system was to slow to use other parameters for ab testing.

More data is available in the mailing list:

Starting here: http://lists.typo3.org/pipermail/typo3-team-core/2013-March/053641.html
Continuing here: http://lists.typo3.org/pipermail/typo3-team-core/2013-March/053739.html

Patch set 2 for branch master has been pushed to the review server.
It is available at https://review.typo3.org/19283