I have finally understood ...
I will make you understand the USL methodology...
If you read this book about USL methodology from Baron Schwartz:
This Baron Schwartz is saying in this book that USL methodology
using nonlinear regression can predict scalability, but i have
finally understood that this Baron Schwartz is making a mistake,
this is what have made me make a mistake, simply because, in a parallel program, a locked region using locks can escape contention at fewer
cores and fewer threads and can than hit the contention at more cores
and more threads.. so you can not use nonlinear regression to predict scalability as is saying Baron Schwartz in his book..
so what is the purpose of the USL methodology that uses nonlinear
regression? its purpose is to permit us to test and see
how much there is contention and how much there is crosstalk(cache
misses etc..), that's all, it's not a tool to predict scalability.
Amine Moulay Ramdane.
|Location:||Huddersfield, West Yorkshire, UK|
|Nodes:||8 (1 / 7)|