XLA : Compiling Machine Learning for Peak Performance

XLA (accelerated linear algebra) is a compiler-based linear algebra execution engine. It is the backend that powers machine learning frameworks such as TensorFlow and JAX at Google, on a variety of devices including CPUs, GPUs, and TPUs. This talk will cover how ML growth has fueled accelerator architectures and the way XLA and related technologies help obtain high performance from the accelerators.

• Hardware and Architecture

• Software Systems