[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: CommsTime times?



Peter,

With SEQ in the delta the times are

- 4.1 usec/cycle
- 1.0 usec/context-switch

This test uses two SEQ constructs in the delta as in the occam example.
Sequential C statements will make the code a bit faster.

Gerald.

> -----Original Message-----
> From: P.H.Welch [mailto:P.H.Welch@xxxxxxxxx]
> Sent: Friday 17 November 2000 15:20 PM
> To: g.h.hilderink@xxxxxxxxxxxxx; java-threads@xxxxxxxxx;
> occam-com@xxxxxxxxx
> Cc: P.H.Welch@xxxxxxxxx
> Subject: RE: CommsTime times?
>
>
>
> Gerald,
>
> > > Gerald: you said "with 5 switches per cycle" for commstime.  But there
> > > are 4 communications per cycle - so I can understand that
> causing 4 or 8
> > > switches ... ?
> >
> > Yes, but the PAR in the delta causes one extra context-switch.
>
> Ah!  Do you have times for commstime with a SEQ delta output?  The cycle
> time divided by 4 then gives a focussed (best) measure of channel comms
> overhead.
>
> Thanks,
>
> Peter.