|
Graham Dumpleton
grahamd at dscpl.com.au
Sat Mar 3 17:30:57 EST 2007
On 04/03/2007, at 7:54 AM, Jorey Bump wrote: > >> If eager, use a hello world mod_python example and trigger one >> request >> against it and see how size changes. > > top - 15:29:18 up 93 days, 5:52, 1 user, load average: 0.00, > 0.03, 0.07 > Tasks: 78 total, 2 running, 76 sleeping, 0 stopped, 0 zombie > Cpu(s): 4.7% user, 0.3% system, 0.0% nice, 95.0% idle > Mem: 904440k total, 465304k used, 439136k free, 39272k > buffers > Swap: 1951856k total, 32408k used, 1919448k free, 113432k > cached > > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND > 31480 nobody 14 0 13248 12m 8836 S 5.0 1.5 0:00.15 httpd Thanks for the feedback so far guys. First thing this all shows is that need a more conclusive way of determining how memory is used by Apache process as top alone isn't that good. I'll look at pmap as suggested and see what else there might be. Apart from the size of PHP, what is disturbing to me though is how much the amount of memory used jumps up on the first request. In phase 1 of looking at this, wasn't intending to look at that aspect, but just threw in the request to try it as an afterthought. I had actually forgotten that it does jump up quite significantly. The reason I had forgotten is that these days I have been playing with mod_wsgi most of the time and have grown accustomed to a small memory footprint. As an example, on MacOS with both mod_python and mod_wsgi modules loaded I start out with: 408 httpd 0.0% 0:00.00 1 8 39 212K 2.86M 900K 30.4M 407 httpd 0.0% 0:00.00 1 8 39 216K 2.86M 904K 30.4M 406 httpd 0.0% 0:00.00 1 8 39 216K 2.86M 904K 30.4M 405 httpd 0.0% 0:00.00 1 8 39 216K 2.86M 904K 30.4M 404 httpd 0.0% 0:00.00 1 8 39 212K 2.86M 904K 30.4M 403 httpd 0.0% 0:00.06 1 11 39 44K 2.86M 1.78M 30.4M Ie., 212K - 216K memory private to the Apache child process. I then fire off a mod_python and mod_wsgi request. For both modules, I have forced use of the first Python interpreter instance created to avoid the creation of an extra interpreter. Being the first interpreter, it is also used for both mod_python and mod_wsgi requests, whereas normally all other interpreter instances are separate to each package. For both cases use simple hello world program. With mod_python use a basic handler and not mod_python.publisher. The result for mod_python request is (pid 405): 410 httpd 0.0% 0:00.00 1 8 39 212K 2.87M 888K 30.4M 408 httpd 0.0% 0:00.00 1 8 39 212K 2.87M 764K 30.4M 407 httpd 0.0% 0:00.00 1 8 39 216K 2.87M 768K 30.4M 406 httpd 0.0% 0:00.00 1 8 39 216K 2.87M 768K 30.4M 405 httpd 0.0% 0:00.37 1 10 86 2.40M 2.75M 3.94M 32.0M 404 httpd 0.0% 0:00.00 1 8 39 212K 2.87M 768K 30.4M 403 httpd 0.0% 0:00.06 1 11 39 32K 2.87M 1.30M 30.4M Thus, private memory use by Apache child process jumps up to 2.40M. Ie., about an extra 2.2M. For a mod_wsgi request however, the result is (pid 407): 410 httpd 0.0% 0:00.00 1 8 39 212K 2.87M 816K 30.4M 408 httpd 0.0% 0:00.00 1 8 39 212K 2.87M 764K 30.4M 407 httpd 0.0% 0:00.01 1 10 39 568K 2.87M 1.64M + 30.4M 406 httpd 0.0% 0:00.00 1 8 39 216K 2.87M 768K 30.4M 405 httpd 0.0% 0:00.37 1 10 86 2.40M 2.75M 3.95M 32.0M 404 httpd 0.0% 0:00.00 1 8 39 212K 2.87M 768K 30.4M 403 httpd 0.0% 0:00.06 1 11 39 32K 2.87M 1.30M 30.4M That is, private memory use by Apache child process only jumps up to 568K. Ie., about an extra 350K. This is quite a big difference and really makes one wander what mod_python is actually doing. This will also have to be investigated, but probably still want to pursue this shared versus static library thing first. Graham
|