Search papers, labs, and topics across Lattice.
University of Alabama
1
0
3
Unlock up to 28% faster GPU communication in HPC workloads by cutting the CPU out of the loop with a new MPI-based API.
