3 ms·
> For many years, RDMA NICs could cache the state for only a few hundred connections; if the number of active connections exceeded the cache size, information h
by GoblinSlayer 2y ago
> For many years, RDMA
NICs could cache the state for only a few hundred connections; if the number of active connections exceeded the cache
size, information had to be shuffled between host memory and
the NIC, with a considerable loss in performance.
A massively parallel task? Sounds like something doable with GPGPU.
- mafuy 2y agoThis has nothing to do with computing, it is about memory access.
- dboreham 2y agoNow the information has to be shuffled between the NIC and host memory and the GPU.
- GoblinSlayer 2y agoJust plug ethernet cable into the graphics card, then you need to shuffle memory between GPGPU and the wire.