Draft page

This page is an outline. It describes what will be covered and is not yet complete technical documentation.

When a model or its workload exceeds one accelerator, the work must be partitioned. The available partitioning schemes cut along different axes, impose different communication patterns, and are routinely combined.

This page explains each scheme and how hybrid strategies are assembled.

What you will learn

  • What each parallelism scheme partitions, and along which axis.
  • The communication pattern each one requires.
  • Why expert parallelism suits mixture-of-experts models.
  • How hybrid strategies are composed in practice.

This page is an outline. The subsections below are the planned structure; they are filled in as the handbook is written.

Why partition at all

Draft

Data parallelism

Draft

Tensor parallelism

Draft

Pipeline parallelism

Draft

Expert parallelism

Draft

Hybrid strategies

Draft

Interconnect requirements

Draft