HPCClusterScape: Increasing Transparency and Efficiency of Shared High-Performance Computing Clusters for Large-scale AI Models

TL;DR

HPCClusterScape is a visualization tool designed to improve transparency and efficiency in shared HPC clusters used for large-scale AI models, enabling better resource management and troubleshooting.

Contribution

The paper introduces HPCClusterScape, a novel visualization system that enhances monitoring, diagnosis, and management of shared HPC resources for large-scale AI workloads.

Findings

Improved resource utilization in HPC clusters

Enhanced detection of workload imbalances and bottlenecks

Positive user feedback on system effectiveness

Abstract

The emergence of large-scale AI models, like GPT-4, has significantly impacted academia and industry, driving the demand for high-performance computing (HPC) to accelerate workloads. To address this, we present HPCClusterScape, a visualization system that enhances the efficiency and transparency of shared HPC clusters for large-scale AI models. HPCClusterScape provides a comprehensive overview of system-level (e.g., partitions, hosts, and workload status) and application-level (e.g., identification of experiments and researchers) information, allowing HPC operators and machine learning researchers to monitor resource utilization and identify issues through customizable violation rules. The system includes diagnostic tools to investigate workload imbalances and synchronization bottlenecks in large-scale distributed deep learning experiments. Deployed in industrial-scale HPC clusters, HPCClusterScape incorporates user feedback and meets specific requirements. This paper outlines the challenges and prerequisites for efficient HPC operation, introduces the interactive visualization system, and highlights its contributions in addressing pain points and optimizing resource utilization in shared HPC clusters.