I started on getting GPU resident MPI working on the Pitt CRC cluster. So far it compiles but doesn’t run. I’m working with the CRC and other group members to get that fixed. I got access to Summit and other OLCF systems this week so after getting setup there and familarizing myself with the systems I worked on getting MPI GPU to work there. After several hurdles and with help I got it working with both the IBM XL and GCC compilers then sent that test info to the Pitt CRC personell.
I started on the OLCF CUDA Training Series and so far I’ve completed the first three parts.
- Get Cholla to run with MPI GPU on H2P
- Figure out how I want to do future tests in Cholla
- Check if Cholla’s regression test is working and port it to whatever the new testing system is
- Learn how Cholla works
- Learn CUDA