Hi,

Now I have got to the point where I can assign different sized chunks to 
different locales, though for now the domain is partitioned only along one 
dimension.

I was wondering if someone could help me with writing and running some tests 
for my domain map, both correctness and performance tests? The code is on 
github and I'm willing to contribute it if it gets developed enough to be 
included in Chapel. I guess that many correctness tests for BlockDist could be 
easily modified for this dist. I'd also like to do performance testing versus 
BlockDist.

Also, some questions...

 - Can I have the assignment of my domain map to be done by reference? Or would 
it be best to just generate compiler error in dsiAssign, since I can't figure 
how assign-by-value should be done...

 - What is the role of dsiClone? Could I just return "this"?

 - Is there a requirement to targetLocDom to have same rank as the distribution 
(see BlockDist)? (I _recall_ Brad saying that at least not intentionlly, but 
I'm not sure...) I tried changing that, but couldn't get it compile because of 
some RAD cache functions.

19.03.2015, 22:10, "Vassily Litvinov" <[email protected]>:
> John,
>
> Yes, it is primarily due to us being more comfortable with rectangular
> arrays rather than associative arrays. You may (or not) observe
> performance difference. Also associative arrays do not provide
> multi-dimensional indexing. If you want your target locale grid to be
> multi-dimensional, you will need to store tuples of indices in your
> associative array.
>
> Vassily
>
> On 03/13/15 09:34, John MacFrenz wrote:
>>  Hi,
>>
>>  Just a (hopefully) quick question: I noticed following comment in 
>> BlockDist.chpl:
>>
>>  ..... I still think the locArr/locDoms arrays should be associative over 
>> locale values.
>>
>>  What is the reason this is not implemented yet? I'd like to do that change 
>> in my distribution since it would make things much easier. However I'm 
>> afraid it would cause significant decrease in performance. I've also got an 
>> impression that associative arrays are much less developed than rectangular 
>> arrays, is that right?

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Chapel-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/chapel-users

Reply via email to