[concurrency-interest] help show speed-up on a trivial but manual array-map operation?

Michael Hicks mwh at cs.umd.edu
Fri Mar 9 08:24:50 EST 2012


Mitkowicz et al also present a very interesting cautionary tale in this regard:

http://www.multicoreinfo.com/research/papers/2009/asplos09-producing-data.pdf

>From the paper: changing the order in which you link your .o files can change performance by up to 10%  Given that many compiler optimizations make "improvements" in the range of 5% makes this result a bit disturbing.

-Mike

On Mar 9, 2012, at 8:20 AM, Doug Lea wrote:

> On 03/09/12 08:11, Doug Lea wrote:
> 
>> 1. Microbenchmarking artifacts: The results of the
>> computations are never used so JVMs can kill some of the
>> code. The attached edited version includes a "checkSum" method
>> that combats this.
> 
> Plus, as I should have checked before, the use of a linear
> stream of initial values is also subject to microbenchmarking
> artifacts. Filling with random values instead combats this.
> 
> The main moral is that reliably measuring anything on modern
> systems is harder than you'd think it should be.
> 
> -Doug
> <VecAdd.java>_______________________________________________
> Concurrency-interest mailing list
> Concurrency-interest at cs.oswego.edu
> http://cs.oswego.edu/mailman/listinfo/concurrency-interest




More information about the Concurrency-interest mailing list