AWS is launching its own machine learning chip to train models for what CEO Andy Jassy says will be the “most cost effective training in the cloud.”
The custom machine learning processor, called AWS Trainium, follows what is becoming a common blueprint for its silicon strategy.
Here’s the strategy in a nutshell:
- Offer instances for Intel, AMD as well as Nvidia.
- But develop your in-house silicon that can leverage its Nitro hypervisor and drive costs lower.
- Once customers adopt instances based on AWS’ Graviton and Inferentia processors they’ll be more comfortable with other custom chips.
- Roll out more custom processors over time.
AWS is ultimately targeting enterprises that are just starting to train models and build out their AI strategies.
Trainium will launch in 2021 and follow AWS instances on Intel’s Habana Gaudi processors.
