Re: KROC & occam on Linux

From: A E Lawrence (adrian.lawrence_at_email.domain.hidden)
Date: 1998-07-07 19:29:55


Dave Beckett wrote:
>
> I have just compiled Peter's classic test program in two flavours and
> put the binaries at this URL:
>
> http://www.hensa.ac.uk/parallel/occam/projects/occam-for-all/kroc/linux/
>
> commstime.occ CommsTime source code
> par-ct commstime.occ compiled with a PAR in the PROC Delta
> seq-ct commstime.occ compiled with a SEQ in the PROC Delta
>
>
> CommsTime in occam ...
>
> Last value received = 1000015
> Time = 5508380 microsecs
> Time per loop = 5508 nanosecs
> Time per context switch = 688 nanosecs
> Last value received = 2000015
> Time = 5508560 microsecs
> Time per loop = 5508 nanosecs
> Time per context switch = 688 nanosecs
> foobar 10$
>
> [OK, this is a rather slow PC :-) ]

Hmmm. This is what I got on a P5-133 with 128Mbyte DRAM:-

seq-ct:
=======

CommsTime in occam ...

Last value received = 1000015
Time = 7436961 microsecs
Time per loop = 7436 nanosecs
Time per context switch = 929 nanosecs
Last value received = 2000015
Time = 7432943 microsecs
Time per loop = 7432 nanosecs
Time per context switch = 929 nanosecs

par-ct
======

CommsTime in occam ...

Last value received = 1000015
Time = 8282887 microsecs
Time per loop = 8282 nanosecs
Time per context switch = 1035 nanosecs
Last value received = 2000015
Time = 8329848 microsecs
Time per loop = 8329 nanosecs
Time per context switch = 1041 nanosecs

So this machine is even slower? Do these figures look reasonable?

Adrian

-- 
Dr A E Lawrence (from home)
adrian.lawrence.without_spam_at_oucs.ox.ac.uk

Original text of this message

This archive was generated by hypermail 2.1.7 on 2004-10-31 20:03:56 GMT
© Copyright WoTUG
All rights reserved