MTIA 300 is the first of Meta’s family of in-house training and inference accelerators optimized for training ranking and recommendation models. We’re sharing how MTIA 300’s built-in NIC chiplets allow it to meet the communication needs associated with training recommendation models with superior performance over general-purpose GPUs. By co-designing MTIA’s communication library, HCCL, alongside the [...]
The post MTIA 300: Meta’s First Training Chip with Built-in NICs and Communication-Off Engines appeared first on Engineering at Meta.
source & further reading
engineering.fb.com — original article