ELK: Exploring the Efficiency of Inter-core Connected AI Chips with Deep Learning Compiler Techniques

TL;DR

This paper introduces Elk, a deep learning compiler framework that optimizes the performance of inter-core connected AI chips by balancing compute, communication, and I/O, leading to near-ideal efficiency.

Contribution

Elk is the first compiler framework to systematically explore and optimize the trade-offs among compute, communication, and I/O for ICCA chips, enhancing their efficiency.

Findings

Achieves 94% of ideal roofline performance on ICCA chips.

Enables effective architecture design space exploration.

Supports large deep learning models efficiently.

Abstract

To meet the increasing demand of deep learning (DL) models, AI chips are employing both off-chip memory (e.g., HBM) and high-bandwidth low-latency interconnect for direct inter-core data exchange. However, it is not easy to explore the efficiency of these inter-core connected AI (ICCA) chips, due to a fundamental tussle among compute (per-core execution), communication (inter-core data exchange), and I/O (off-chip data access). In this paper, we develop Elk, a DL compiler framework to maximize the efficiency of ICCA chips by jointly trading off all the three performance factors discussed above. Elk structures these performance factors into configurable parameters and forms a global trade-off space in the DL compiler. To systematically explore this space and maximize overall efficiency, Elk employs a new inductive operator scheduling policy and a cost-aware on-chip memory allocation algorithm. It generates globally optimized execution plans that best overlap off-chip data loading and on-chip execution. To examine the efficiency of Elk, we build a full-fledged emulator based on a real ICCA chip IPU-POD4, and an ICCA chip simulator for sensitivity analysis with different interconnect network topologies. Elk achieves 94% of the ideal roofline performance of ICCA chips on average, showing the benefits of supporting large DL models on ICCA chips. We also show Elk's capability of enabling architecture design space exploration for new ICCA chip development.