[Gretl-devel] Re: MPI (Re: [Gretl-users] Re: gretl 2023b released)
On Tue, Jul 25, 2023 at 6:24 AM Sven Schreiber
<sven.schreiber(a)fu-berlin.de> wrote:
Am 25.07.2023 um 01:09 schrieb Cottrell, Allin:
> On Mon, Jul 24, 2023 at 4:34 PM Sven Schreiber
> <sven.schreiber(a)fu-berlin.de> wrote:
>> [moving this to gretl-devel]
>>
>> Am 21.07.2023 um 20:49 schrieb Cottrell, Allin:
>>> While I haven't tested very rigorously (and don't really understand
>>> how this would work), it seems this has speeded up MPI performance on
>>> Windows quite substantially.
>> ...
>> Oh, does it fail on Linux with 2023a?
> On Linux, your script runs OK on both 2023a and 2023b.
OK; do you have a script where you saw the speed-up on Windows?
It was the BVAR MPI test script from June. But it wasn't a proper
experiment. I've noticed that there can be tons of processes going on
in the background on Windows, and I guess a difference in the number
or activity-level of those -- something I did not control for -- could
make a big difference to MPI times.
Allin