Ejabberd resource tests/try 5: Difference between revisions

From OLPC
Jump to navigation Jump to search
 
(One intermediate revision by the same user not shown)
Line 2: Line 2:


It turns out that in the hyperactivity agents were not interacting in
It turns out that in the hyperactivity agents were not interacting in
[[Ejabberd resource tests/try 4]].
[[Ejabberd resource tests/try 4]], due to the shared roster being off by mistake.


For this test several hyperactivity instances were used, each running 50 clients. 4 XO laptops were also connected, so the total number of clients was at most stages 50n + 4.
For this test several hyperactivity instances were used, each running 50 clients. 4 XO laptops were also connected, so the total number of clients was at most stages 50n + 4.
Line 16: Line 16:
=== Memory use vs connections ===
=== Memory use vs connections ===


This shows max-median-min values for stable points along the way. (Where the readings appear to stack up on each other, the number of connections varies by one or so).
This shows max-median-min values for stable points along the way (3 minutes with the same number of connections).


[[Image:try5-resident_mem_min_per_conn-resident_mem_max_per_conn-resident_mem_median_per_conn.png]]
[[Image:try5-resident_mem_min_per_conn-resident_mem_max_per_conn-resident_mem_median_per_conn.png]]

Here's the raw data (duplicates exist because a number of connections can vary slightly as clients disconnect and reconnect).

resident_mem
clients minimum median maximum
54 96 97 98
104 134 139 143
154 164 167 174
204 220 224 253
204 229 232 236
254 270 278 289
304 345 399 421
254 329 348 386
304 337 352 419
354 455 471 515
403 499 577 654
404 502 504 514
454 635 644 696
453 642 644 646
454 608 613 664


=== CPU usage ===
=== CPU usage ===

Latest revision as of 03:22, 4 November 2008

Try 5: a few hundred users, interacting properly

It turns out that in the hyperactivity agents were not interacting in Ejabberd resource tests/try 4, due to the shared roster being off by mistake.

For this test several hyperactivity instances were used, each running 50 clients. 4 XO laptops were also connected, so the total number of clients was at most stages 50n + 4.

Memory use

Here you can see the memory use is slightly over 1 MB per user, and growth is approximately linear or slightly in excess thereof. The server has 1GB of ram.

Try5-users active vs resident mem.png

Try5-psmem-resident mem-virtual mem.png

Memory use vs connections

This shows max-median-min values for stable points along the way (3 minutes with the same number of connections).

Try5-resident mem min per conn-resident mem max per conn-resident mem median per conn.png

Here's the raw data (duplicates exist because a number of connections can vary slightly as clients disconnect and reconnect).

resident_mem
clients              minimum     median    maximum
54                        96         97         98
104                      134        139        143
154                      164        167        174
204                      220        224        253
204                      229        232        236
254                      270        278        289
304                      345        399        421
254                      329        348        386
304                      337        352        419
354                      455        471        515
403                      499        577        654
404                      502        504        514
454                      635        644        696
453                      642        644        646
454                      608        613        664

CPU usage

Cumulative CPU use by the ejabberd process:

Try5-users active vs cputime.png

Load averages, by connections and by time:

Try5-load avg 5 min per conn-load avg 5 max per conn-load avg 5 median per conn.png

Try5-load avg 1-load avg 5-load avg 15.png


The laptops

Note that they all see a different number of clients.

Try5-hyperactivity-3-800.jpg