[Hardware] Problem with Client on an IBM iSeries in PASE Mode

Stan Vernaillen Sod75 at pandora.be
Wed Mar 8 12:08:50 EST 2006


hi there, 
the benchmark values don't seem too bad to me, they are just a bit slower then 
what I get. 
This is on an AIX system with 1.2 Ghz Power4's

dnetc v2.9009-494-CTR-04081422 for AIX (AIX 5.2).
<snip>
[Mar 08 10:35:36 UTC] Automatic processor type detection found
                      a PowerPC POWER_4 processor.
<snip>
[Mar 08 10:35:55 UTC] RC5-72: using core #1 (KKS 2-pipe).
[Mar 08 10:36:13 UTC] RC5-72: Benchmark for core #1 (KKS 2-pipe)
                      0.00:00:16.14 [2,964,119 keys/sec]
<snip>
[Mar 08 10:37:08 UTC] OGR-P2: using core #0 (KOGE 2.0 Scalar).
[Mar 08 10:37:28 UTC] OGR-P2: Benchmark for core #0 (KOGE 2.0 Scalar)
                      0.00:00:16.35 [14,106,398 nodes/sec]

if that's too slow, at least it says it's not a problem on your side...

I've never tried PASE before so I'm not familiar with how it really works, but 
I'm surprised to see the client even works :)
according to wikipedia you'lld have to recompile the binary...

http://en.wikipedia.org/wiki/ISeries
AIX programs ARE binary compatible with OS/400 when using OS/400's PASE 
(Portable Applications System Environment). PASE is essentially "an operating 
system within an operating system", supporting the most recent stable version 
of AIX. Binaries need to be re-compiled on the AIX system, with 16-bit 
(quadword) pointer alignment enabled. Once the program is compiled with this 
option, the executable can be FTP-ed to the AS/400, and executed under the 
PASE Korn Shell.

no idea why the client would bench fine, but not run normally, Benches do run 
at a higher priority then normal operations and os/400 is a bit of a strange 
one on these kind of things. 
Does it not run at all ? I saw you submitted a few blocks nevertheless...

Regards, 
Stan

not sure if this belongs on the Hardware list though...a lot more people on 
the rc5 general list I'd guess

On Tuesday 07 March 2006 23:02, Carsten Schulz wrote:
> Hi together,
>
> I have an IBM iSeries with three PowerPC Power_4 Processors.
> I have tried to run the client and the following happend:
>
> Using email address (distributed.net ID) 'is400 at web.de'
>
> [Mar 07 21:45:30 UTC] Automatic processor type detection found
>                       a PowerPC POWER_4 processor.
> [Mar 07 21:45:30 UTC] RC5-72: Running micro-bench to select fastest core...
> [Mar 07 21:45:52 UTC] RC5-72: using core #1 (KKS 2-pipe).
> [Mar 07 21:46:10 UTC] RC5-72: Benchmark for core #1 (KKS 2-pipe)
>
>                       0.00:00:16.32 [2,715,745 keys/sec]
> [Mar 07 21:46:10 UTC] OGR-P2: using core #0 (KOGE 2.0 Scalar).
> [Mar 07 21:46:29 UTC] OGR-P2: Benchmark for core #0 (KOGE 2.0 Scalar)
>
>                       0.00:00:16.59 [12,914,019 nodes/sec]
>
>
> This is the output when I am doing        dnetc --benchmark
> But when I am starting the client with data ti crunch it is doing nothing.
> An I think 2,715,745 keys/sec  is very poor for the processor.
> So I have no idea how to get the client working.
> The log file is the folling:
>
> dnetc v2.9009-494-CTR-04081422 for AIX (OS400 5.3).
> Using email address (distributed.net ID) 'is400 at web.de'
>
> [Mar 07 21:36:06 UTC] Automatic processor detection found 3 processors.
> [Mar 07 21:36:06 UTC] Loading crunchers with work...
> [Mar 07 21:36:06 UTC] Automatic processor type detection found
>                       a PowerPC POWER_4 processor.
> [Mar 07 21:36:28 UTC] RC5-72: using core #1 (KKS 2-pipe).
> [Mar 07 21:36:28 UTC] RC5-72 #a: Loaded CB:3BFE2E6E:00000000:1*2^32
> [Mar 07 21:36:28 UTC] RC5-72 #b: Loaded CB:3BFE2E6F:00000000:1*2^32
> [Mar 07 21:36:28 UTC] RC5-72 #c: Loaded CB:3BFE2E70:00000000:1*2^32
> [Mar 07 21:36:28 UTC] RC5-72: 7 packets (7.00 stats units) remain in
>                       buff-in.r72
>                       Projected ideal time to completion: 0.01:01:29.00
> [Mar 07 21:36:28 UTC] RC5-72: 0 packets are in buff-out.r72
> [Mar 07 21:36:28 UTC] 3 crunchers ('a'-'c') have been started.
> [Mar 07 21:36:59 UTC] *Break* Shutting down...
> [Mar 07 21:36:59 UTC] RC5-72 #c: Saved CB:3BFE2E70:00000000:1*2^32 (0 keys
> done) 0.00:00:31.60 - [0 keys/s]
> [Mar 07 21:36:59 UTC] RC5-72 #b: Saved CB:3BFE2E6F:00000000:1*2^32 (0 keys
> done) 0.00:00:31.60 - [0 keys/s]
> [Mar 07 21:36:59 UTC] RC5-72 #a: Saved CB:3BFE2E6E:00000000:1*2^32 (0 keys
> done) 0.00:00:31.61 - [0 keys/s]
> [Mar 07 21:36:59 UTC] RC5-72: 10 packets (10.00 stats units) are in
>                       buff-in.r72
> [Mar 07 21:36:59 UTC] RC5-72: 0 packets are in buff-out.r72
> [Mar 07 21:36:59 UTC] Shutdown complete.
>
> dnetc v2.9009-494-CTR-04081422 for AIX (OS400 5.3).
> Using email address (distributed.net ID) 'is400 at web.de'
>
> [Mar 07 21:45:30 UTC] Automatic processor type detection found
>                       a PowerPC POWER_4 processor.
> [Mar 07 21:45:52 UTC] RC5-72: using core #1 (KKS 2-pipe).
> [Mar 07 21:46:10 UTC] RC5-72: Benchmark for core #1 (KKS 2-pipe)
>                       0.00:00:16.32 [2,715,745 keys/sec]
> [Mar 07 21:46:10 UTC] OGR-P2: using core #0 (KOGE 2.0 Scalar).
> [Mar 07 21:46:29 UTC] OGR-P2: Benchmark for core #0 (KOGE 2.0 Scalar)
>                       0.00:00:16.59 [12,914,019 nodes/sec]
>
>
>
> best regards,
>
> Carsten
>
>
> ______________________________________________________________
> Verschicken Sie romantische, coole und witzige Bilder per SMS!
> Jetzt bei WEB.DE FreeMail: http://f.web.de/?mc=021193
>
> _______________________________________________
> Hardware mailing list
> Hardware at lists.distributed.net
> http://lists.distributed.net/mailman/listinfo/hardware


More information about the Hardware mailing list