[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: mpi 1.1.1 released
From: |
c. |
Subject: |
Re: mpi 1.1.1 released |
Date: |
Sun, 5 Jan 2014 11:52:35 +0100 |
On 3 Jan 2014, at 14:44, c. <address@hidden> wrote:
>
> On 3 Jan 2014, at 14:04, Sukanta Basu <address@hidden> wrote:
>
>> Hi Carlo,
>>
>> Thanks in advance for posting the code in bug-tracker. My code deals
>> with many arrays. Reshaping them for every communication will
>> definitely slow down the computation.
>
> If I understand correctly how reshape is implemented, this will not be a
> large
> slowdown because you will just be redefining the dim_vectors, not moving data
> around ...
>
>> Today, I am going to check if
>> one-dimensional array solves the problem.
>
> that is a useful test.
>
>> Cheers,
>> Sukanta
> c.
Sukanta,
Could you please try to run the attached test (without MPI) and see if displays
the same "leacky" behaviour?
c.
memory_test.m
Description: Binary data
prova.cc
Description: Binary data
- Re: mpi 1.1.1 released, Michael Creel, 2014/01/02
- Re: mpi 1.1.1 released, c., 2014/01/02
- Re: mpi 1.1.1 released, Michael Creel, 2014/01/02
- Re: mpi 1.1.1 released, c., 2014/01/02
- Message not available
- Re: mpi 1.1.1 released, Michael Creel, 2014/01/03
- Message not available
- Re: mpi 1.1.1 released, c., 2014/01/03
- Re: mpi 1.1.1 released, c., 2014/01/03
- Message not available
- Re: mpi 1.1.1 released, c., 2014/01/03
- Re: mpi 1.1.1 released,
c. <=
- Message not available
- Re: mpi 1.1.1 released, c., 2014/01/06
- Re: mpi 1.1.1 released, c., 2014/01/06
- Message not available
- Re: mpi 1.1.1 released, Michael Creel, 2014/01/06
- Re: mpi 1.1.1 released, c., 2014/01/06
- Message not available
- Re: mpi 1.1.1 released, c., 2014/01/07
- Re: mpi 1.1.1 released, c., 2014/01/07
- Re: mpi 1.1.1 released, Doug Stewart, 2014/01/07
- Re: mpi 1.1.1 released, c., 2014/01/07
- Re: mpi 1.1.1 released, Doug Stewart, 2014/01/07
- Re: mpi 1.1.1 released, Doug Stewart, 2014/01/07