Re: testing ProcArrayLock patches

Robert Haas <robertmhaas@gmail.com> wrote:
>>  I think so.  My take was that it was showing 32 of 64 *threads*
>> active -- the hyperthreading funkiness.  Is there something in
>> particular you'd like me to check?
> 
> Not really, just don't understand the number.
I'm having trouble resolving the vmstat numbers I got during the
32-client pgbench runs which modified data.
-M simple:
procs --------------memory------------- ---swap-- -----io-----r  b   swpd   free      buff    cache   si   so    bi
bo----system---------cpu------    in      cs us sy id wa st
 
30  1   4464 513492 205564572 54472124    0    0     0  78170621724 1246300 30  8 61  1  0
27  1   4464 509288 205564572 54474600    0    0     0 125620599403 1192046 29  8 63  1  0
35  1   4464 508368 205564572 54476996    0    0     0  89801595939 1186496 29  8 63  0  0
25  0   4464 506088 205564572 54478668    0    0     0  90121594800 1189649 28  8 63  0  0
-M prepared:
procs --------------memory-------------- ---swap-- -----io-----r  b   swpd    free      buff    cache   si   so    bi
 bo----system---- -----cpu------    in      cs us sy id wa st
 
28  0   5612 1204404 205107344 54230536    0    0     0  93212527284 1456417 22  9 69  0  08  1   5612 1202044
20510734454233336    0    0     0  93217512819 1417457 21  9 70  1  0
 
17  1   5612 1201892 205107344 54236048    0    0     0 132699502333 1412878 21  9 70  0  0
19  1   5612 1199208 205107344 54238936    0    0     0  93612519113 1484386 21  9 69  0  0
So 60% or 70% idle without any I/O wait time.  I don't know how to
explain that.
-Kevin