/[pgestraier]/trunk/misc/bench.txt
This is repository of my old source code which isn't updated any more. Go to git.rot13.org for current projects!
ViewVC logotype

Diff of /trunk/misc/bench.txt

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 25 by dpavlin, Fri May 27 21:06:01 2005 UTC revision 26 by dpavlin, Sun May 29 19:32:28 2005 UTC
# Line 8  Benchmark: timing 1000 iterations of 0_o Line 8  Benchmark: timing 1000 iterations of 0_o
8       3_new: 10 wallclock secs ( 1.93 usr +  0.04 sys =  1.97 CPU) @ 507.61/s (n=1000)       3_new: 10 wallclock secs ( 1.93 usr +  0.04 sys =  1.97 CPU) @ 507.61/s (n=1000)
9       4_old: 13 wallclock secs ( 2.66 usr +  0.05 sys =  2.71 CPU) @ 369.00/s (n=1000)       4_old: 13 wallclock secs ( 2.66 usr +  0.05 sys =  2.71 CPU) @ 369.00/s (n=1000)
10       5_new: 12 wallclock secs ( 1.96 usr +  0.03 sys =  1.99 CPU) @ 502.51/s (n=1000)       5_new: 12 wallclock secs ( 1.96 usr +  0.03 sys =  1.99 CPU) @ 502.51/s (n=1000)
11    
12    
13    If you want to re-create this benchmark on your configuration, checkout
14    revision between 19 and 24 because revision 25 obsoleted (and removed) old
15    version of function.
16    
17    Have also in mind, that old API always fetched 6 attributes (4 found output
18    and 2 for debug) while new API fetched just one for this benchmark.
19    
20    Each attribute fetch decrease performance for about 20 transactions in
21    second, so specifying many unneeded attributes isn't free. That's why new API
22    has attribute specification - both for performance and usability.
23    
24    And at end, a well-known secret: since we are using materialized views
25    (which take place in PostgreSQL sort memory), increasing sort_mem parameter in
26    postgresql.conf will dramatically increase performance.
Legend:
Removed from v.25  
changed lines
  Added in v.26
  ViewVC Help
Powered by ViewVC 1.1.26