[concurrency-interest] Ops type names

Neal Gafter neal at gafter.com
Tue Jan 8 22:41:05 EST 2008


On Jan 8, 2008 4:22 AM, Doug Lea <dl at cs.oswego.edu> wrote:

> The main reason this isn't supported (yet?) is that it creates
> another combinatorial API and implementation blowup if done
> completely orthogonally, As in:
> a.withBounds(l,h).withFilter(f).withMapping(m).withCombination(b,c).max()
> (which would entail creating internal classes such as
> WithBoundedFilteredMappedCombination to maintain all the bookkeeping
> needed to flatten out the leaf computations, which is needed to
> get good parallel performance.)


One strategy for something like this is to build expression trees for
intermediate results, and then generate the top-level bytecode at runtime
just before evaluating it.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20080108/2f69155c/attachment.html 


More information about the Concurrency-interest mailing list