Someone just gave me an idea... I'll just do 50 small simulations on 50
different computers and then I'll import the 50 clouds in the same scene.
Not the most elegant way to do it but it should work.
Of course, I'm still curious about that networking solution as other
problems might eventually come my way requiring that kind of solution. But
at least, it becomes much less urgent and stressful.
Mathieu
-----Original Message-----
From: owner-xsi(at)Softimage.COM [mailto:owner-xsi(at)Softimage.COM]On Behalf
Of Mathieu Leclaire
Sent: Tuesday, April 25, 2006 4:38 PM
To: XSI(at)Softimage.COM
Subject: RE: network distributed particle simulations?
Hum... ok, I'll take a look at that. Thanks for pointing it out. I just hope
it won't take too long to digest.
Mathieu
-----Original Message-----
From: owner-xsi(at)Softimage.COM [mailto:owner-xsi(at)Softimage.COM]On Behalf
Of Halfdan Ingvarsson
Sent: Tuesday, April 25, 2006 3:11 PM
To: XSI(at)Softimage.COM
Subject: RE: network distributed particle simulations?
There's always Open-MPI, which does exactly (and more) what you're looking
for. http://www.open-mpi.org/
It implements the interface defined here: http://www.mpi-forum.org/docs/
Like most distributed systems' internals, it's rather quite scary at first,
although it is a hell of a lot easier than trying to develop one yourself.
- ½
> -----Original Message-----
> From: owner-xsi(at)Softimage.COM
> [mailto:owner-xsi(at)Softimage.COM]On Behalf
> Of Mathieu Leclaire
> Sent: Tuesday, 25 April, 2006 14:54
> To: XSI(at)Softimage.COM
> Subject: RE: network distributed particle simulations?
>
>
> well, like I said, it's not the simulation that needs to be
> distributed...
> it's the calculations done in my custom command. I need to ba
> able in my
> custom command to call some evaluations on different
> computers. For example,
> the custom commands asks computer 1 to evaluate particles 0
> to 99, computer
> 2 to evaluated particles 100 to 199... and then wait for the
> results before
> writting them to the UserDataBlob and then going to the next
> frame. So it's
> not the particle simulation per say but the custom command
> that the event
> calls that needs to be distributed through the network.
>
>
> Mathieu
>
> -----Original Message-----
> From: owner-xsi(at)Softimage.COM
> [mailto:owner-xsi(at)Softimage.COM]On Behalf
> Of Kris Rivel
> Sent: Tuesday, April 25, 2006 2:24 PM
> To: XSI(at)Softimage.COM
> Subject: Re: network distributed particle simulations?
>
>
> An hour per frame....wow, you don't mess around do you :-) I didn't
> think distributing particles was possible. I know that the Next Limit
> guys are working on this I think but the obvious problem is that you
> need the previous frame to simulate the next frame, etc.
>
> Kris
>
> Mathieu Leclaire wrote:
> > OK, I'm still working on my particle simulations here... A
> particle cloud
> > with a scripted event that calls a custom command on every frame to
> > calculate a special collision event, where it passes
> information from one
> > cloud to another through a UserDataBlob when certain
> particles collide.
> > Anyway, the thing is, for the number of particles that I
> need, and even
> > though I believe I've optimized the simulation as best I
> can, it takes
> > roughly an hour a frame to calculate my simulations. At 200
> frames, well,
> it
> > takes more then a week of simulation using one machine. We
> have a pretty
> big
> > render farm here so if I could find a way to use it to split the
> simulations
> > over multiple machines, then we can make the simulations in a more
> > reasonable time frame. So my question is this: how can I
> distribute my
> > simulations over a network? What would be the best way to
> approach the
> > issue? I need to have xsi evaluate the cloud normally, make
> the particles
> > move depending on forces and once the particles are
> positioned correctly,
> my
> > scripted event is called that calls the custom command. How
> can I separate
> > the process of that custom command over multiple machines
> so that one
> > machine calculates lets say the first 100 particles, and
> the next one
> > calculates another group of 100 particles and so on, until all the
> particles
> > have been evaluated and all the resulting data can be passed to the
> > UserDataBlob... I just want to make sure you guys
> understand that what
> needs
> > to be distributed is the evaluation in my custom command
> that happens
> > between each frames and that, depending on that result, the
> particles will
> > change and affect the following frame. So before the next
> frame can be
> > evaluated, the current network process must be done before
> the next frame
> > can start. Any suggestions on how to do something like
> that? Anyone had to
> > deal with something similar? Any help or suggestions would
> be greatly
> > appreciated.
> >
> >
> > Mathieu Leclaire
> > R&D Programmer
> > Hybride Technologies
> >
> >
> >
> > ---
> > Unsubscribe? Mail Majordomo(at)Softimage.COM with the
> following text in body:
> > unsubscribe xsi
> >
> >
> >
>
> ---
> Unsubscribe? Mail Majordomo(at)Softimage.COM with the following
> text in body:
> unsubscribe xsi
>
>
>
> ---
> Unsubscribe? Mail Majordomo(at)Softimage.COM with the following
> text in body:
> unsubscribe xsi
>
---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi
---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi
---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi
