S5R3
22 Sep 2007 16:06:15 UTC
Topic 13596
(moderation:
We generated the first Workunits of S5R3, some of you probably already got them.
The code of the Apps hasn't changed very much compared to the 4.4x Apps of S5R2. If you would run the same S5R2 workunit with both of them you'd find the S5R3 App being slightly faster.
As some paramaters of the Workunit split-up have changed, we needed to adjust our "credit normalization factor" (roughly credit per template). We may have to tune this again in the next few days. Please watch your credit per hour ratio and report here (and also take into account a normal fluctuation of +-5%).
BM
BM
S5R3
)
Thanks -- very helpful! Other users, please post your observations.
Bruce
Update: We have found a
)
Update:
We have found a model to predict the run-time variation of tasks with equal number of templates (i.e. tasks that currently get the same credit). Note that the credit assigned is already correct on average over all the workunits of a base frequency.
We are working on integrating this model into our Workunit generator to apply a correction to the credit assigned to the WU such that the credit will roughly match the expected run-time even for a single task. However the amplitude of the variation in run-time differs between platforms, CPU types and even the speed of the memory (interface) of the machines, it's rater impossible to make this completely accurate.
BM
BM
RE: RE: The new
)
And yes, it's this coincidence step that wasn't present in S5R1 and before that causes the run-time variation, the memory bandwidth dependency and some other unexpected behavior.
BM
BM