Multi-Objective Hardware-Mapping Co-Optimisation for Multi-DNN Workloads on Chiplet-based Accelerators

TL;DR

This paper introduces MOHaM, a multi-objective evolutionary framework for optimizing chiplet-based accelerators to efficiently run multiple DNNs, balancing execution time, energy, and cost.

Contribution

It presents a novel multi-objective co-optimization framework using evolutionary algorithms tailored for multi-DNN workloads on chiplet accelerators.

Findings

MOHaM achieves Pareto optimal solutions compared to state-of-the-art.

Reduces latency by up to 96%.

Reduces energy consumption by up to 96.12%.

Abstract

The need to efficiently execute different Deep Neural Networks (DNNs) on the same computing platform, coupled with the requirement for easy scalability, makes Multi-Chip Module (MCM)-based accelerators a preferred design choice. Such an accelerator brings together heterogeneous sub-accelerators in the form of chiplets, interconnected by a Network-on-Package (NoP). This paper addresses the challenge of selecting the most suitable sub-accelerators, configuring them, determining their optimal placement in the NoP, and mapping the layers of a predetermined set of DNNs spatially and temporally. The objective is to minimise execution time and energy consumption during parallel execution while also minimising the overall cost, specifically the silicon area, of the accelerator. This paper presents MOHaM, a framework for multi-objective hardware-mapping co-optimisation for multi-DNN workloads on chiplet-based accelerators. MOHaM exploits a multi-objective evolutionary algorithm that has been specialised for the given problem by incorporating several customised genetic operators. MOHaM is evaluated against state-of-the-art Design Space Exploration (DSE) frameworks on different multi-DNN workload scenarios. The solutions discovered by MOHaM are Pareto optimal compared to those by the state-of-the-art. Specifically, MOHaM-generated accelerator designs can reduce latency by up to $96\%$ and energy by up to $96.12\%$.