Re: End of optimization...

On Thu, 03 Jul 2003 23:27:09 -0400, maya wrote:


> According to a colleague of mine, this is borne out somewhat by the
> trends at recent conferences - there is not much optimization work being
> published. Instead, papers seem to be about areas such as program
> verification.


People have even been known to argue that this is a good thing, since
optimization research really hasn't really been all that effective.
Todd Proebsting even asserts that advances in compiler optimizations
double computing power only every 18 years. See
http://research.microsoft.com/~toddpro/ for some observations
for/against this. I believe I even heard Todd even claim that
"compiler optimization research ruined PLDI for me", though I can't
remember which talk that was in.


It seems to me that many hardware mechanisms are compensating for
deficiencies in compiler optimization. A current example might be
improving performance on multi-threaded processors. There are compiler
guys doing speculative threading in software, versus processor guys
doing it in hardware. Perhaps we're going to see some akin to the OOO
versus EPIC style split?


niall

Return to the comp.compilers page.
Search the comp.compilers archives again.