ASTRA-sim2.0: Modeling Hierarchical Networks and Disaggregated Systems for Large-model Training at Scale

TL;DR

This paper extends the ASTRA-sim simulation infrastructure to model hierarchical networks and disaggregated systems, enabling scalable analysis of large-model distributed training architectures.

Contribution

It introduces new modeling capabilities for arbitrary model parallelization, multi-dimensional topologies, and advanced memory systems within ASTRA-sim.

Findings

Supports arbitrary model parallelization strategies.

Enables simulation of large-scale, heterogeneous distributed systems.

Provides accurate performance estimates for emerging training platforms.

Abstract

As deep learning models and input data are scaling at an unprecedented rate, it is inevitable to move towards distributed training platforms to fit the model and increase training throughput. State-of-the-art approaches and techniques, such as wafer-scale nodes, multi-dimensional network topologies, disaggregated memory systems, and parallelization strategies, have been actively adopted by emerging distributed training systems. This results in a complex SW/HW co-design stack of distributed training, necessitating a modeling/simulation infrastructure for design-space exploration. In this paper, we extend the open-source ASTRA-sim infrastructure and endow it with the capabilities to model state-of-the-art and emerging distributed training models and platforms. More specifically, (i) we enable ASTRA-sim to support arbitrary model parallelization strategies via a graph-based training-loop implementation, (ii) we implement a parameterizable multi-dimensional heterogeneous topology generation infrastructure with analytical performance estimates enabling simulating target systems at scale, and (iii) we enhance the memory system modeling to support accurate modeling of in-network collective communication and disaggregated memory systems. With such capabilities, we run comprehensive case studies targeting emerging distributed models and platforms. This infrastructure lets system designers swiftly traverse the complex co-design stack and give meaningful insights when designing and deploying distributed training platforms at scale.