> It could be that the two threads are contending over the Array object
> reference (Java arrays aren't pointers). Is there a nice way to
> create "subvector" objects that only reference the underlying memory
> and not the parent array object?
You might be interested in the *Buffer classes (e.g.
Is the mmap interface a possibility?
You could bulkget an array for each thread. I guess that would
probably not gain much against just copying an array for each thread.
I couldn't find a way to create a subvector of a java array.
Chris
On Jan 21, 12:20 pm, "Mark H." wrote:
> On Jan 21, 8:56
On Jan 21, 8:56 am, Perry Trolard wrote:
> If I understand what you did correctly, the reason it worked for you
> is that Rich committed a fix to SVN in the meantime!
Awesome, real-time fix!
> > The last (count result-p) call takes a few seconds (probably
> > because there's a one-to-one mappin
Thanks for testing, Mark.
If I understand what you did correctly, the reason it worked for you
is that Rich committed a fix to SVN in the meantime!
> The last (count result-p) call takes a few seconds (probably
> because there's a one-to-one mapping of tasks to sequence
> elements) but it finish
On Jan 20, 8:36 am, Perry Trolard wrote:
> Yesterday I had a strange case of pmap (parallel map) eating up all of
> the heap space I'd make available -- from 256M up to a Gig. I track
> SVN, & I know for sure this wasn't happening last Wednesday (the
> 14th). Calling map in place of pmap caused t
On Jan 20, 10:38 am, Perry Trolard wrote:
> But I don't think lazyness is the problem:
>
> user=> (def r (doall (map inc (range 100
> #'user/r
> user=> (count r)
> 100
> user=> (pmap inc [0])
> java.lang.OutOfMemoryError: Java heap space (NO_SOURCE_FILE:0)
Hm, that's funny, the latte
On Jan 20, 1:38 pm, Perry Trolard wrote:
> > Doesn't pmap have to construct the whole sequence explicitly in order
> > to map its execution across more than one processor? or does it take
> > in a lazy fashion?
>
> Semi-lazy, according to the doc:
>
> Like map, except f is applied in parallel
On Jan 20, 10:38 am, Perry Trolard wrote:
> > Doesn't pmap have to construct the whole sequence explicitly in order
> > to map its execution across more than one processor? or does it take
> > in a lazy fashion?
>
> Semi-lazy, according to the doc:
>
> Like map, except f is applied in parallel.
> Doesn't pmap have to construct the whole sequence explicitly in order
> to map its execution across more than one processor? or does it take
> in a lazy fashion?
Semi-lazy, according to the doc:
Like map, except f is applied in parallel. Semi-lazy in that the
parallel computation stays ahe
On Jan 20, 8:36 am, Perry Trolard wrote:
> Yesterday I had a strange case of pmap (parallel map) eating up all of
> the heap space I'd make available -- from 256M up to a Gig. I track
> SVN, & I know for sure this wasn't happening last Wednesday (the
> 14th). Calling map in place of pmap caused t
10 matches
Mail list logo
