Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> 3.2Tbps bandwidth PER MACHINE

Looking into the Ultraclusters page you linked to in a sibling comment, it seems like the host machines pretty much fill out their PCIe connections with Infiniband networking to reach that figure:

    EFA is also coupled with NVIDIA GPUDirect RDMA (P5, P4d) and
    NeuronLink (Trn1) to enable low-latency accelerator-to-accelerator
    communication between servers with operating system bypass.
https://aws.amazon.com/ec2/ultraclusters/


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: