-
Notifications
You must be signed in to change notification settings - Fork 9
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
MPICH 64-bit? #67
Comments
I actually have a self-answer here, while it's viable it's likely not worth the trouble with the current architecture. I ended up using the pre-built binaries from the multibugs-2.0 Ubuntu release and I can confirm running up to 24 processors in a container (Singularity chosen here). I'll be happy to share or contribute the recipe if there's interest. I could then run the seeds model afterwards. |
Great! Please do share the recipe - we've also had success with running MultiBUGS within Singularity so glad you found the same solution worked for you. I will try to find our recipe in case comparing notes is helpful. |
How do you get mpich:i386 on Ubuntu 20.04? It is not even in the repository any more. It is 2020 why is this software still uses 32 bit libs? |
I think, that this should work for MultiBUGS also. |
Hi, Thanks for the reply. It turns out that the Linux binary tarballs already contains mpich32, I also tried your method with the .deb by extracting the contents and ran it locally, it was all fine provided you change the path in the startup script. However, there seems to be a problem with R2MultiBUGS, it just seqfaults and crashes. Now since this is Linux the debug mode doesn't work I have no idea what happens. But my guess is that in the R environment the environmental variables are not set so it was loading the 64 bit libraries and in this case openmpi. I don't know the internal working of MultiBUGS or R2MultiBUGS, but is my guess correct and if so is there a way to address that without having to use these environmental variables globally (i.e in R itself, and R may not even work if you try that)? |
@kmanalo struggling to adapt your code to work on CentOS 7 distribution. Was wondering if 64-bit MPI was the way forward. |
Any experiences with using non 32-bit MPIs?
I'm on a compute cluster that uses MVAPICH2 (MPICH ABI compatible)
https://www.mpich.org/abi/
The text was updated successfully, but these errors were encountered: