In GPGPU remoting, TCP/IP introduces latency and overhead. Our RDMA communicator for GVirtuS eliminates system calls, cutting Matrix Multiplication time by 5.5x and context switches by 98%. The winner? RDMA.
🌟 Follow us on LinkedIn! https://www.linkedin.com/company/clever-project/?viewAsMember=true
🌟 Check the updates from the website: www.cleverproject.eu
🌟 Full paper in: https://zenodo.org/records/14717622