AIPerf: Automated machine learning as an AI-HPC benchmark
This addresses the problem of benchmarking AI-HPC systems for researchers and practitioners, offering a scalable alternative to existing benchmarks like LINPACK and MLPerf, though it is incremental in improving benchmark design.
The paper tackles the lack of scalable and representative benchmarks for AI-HPC systems by proposing AIPerf, an end-to-end benchmark suite using automated machine learning (AutoML) that adapts to various machine scales, achieving near-linear weak scalability from 56.1 Tera-OPS on 4 nodes to 194.53 Peta-OPS on 512 nodes.
The plethora of complex artificial intelligence (AI) algorithms and available high performance computing (HPC) power stimulates the expeditious development of AI components with heterogeneous designs. Consequently, the need for cross-stack performance benchmarking of AI-HPC systems emerges rapidly. The de facto HPC benchmark LINPACK can not reflect AI computing power and I/O performance without representative workload. The current popular AI benchmarks like MLPerf have fixed problem size therefore limited scalability. To address these issues, we propose an end-to-end benchmark suite utilizing automated machine learning (AutoML), which not only represents real AI scenarios, but also is auto-adaptively scalable to various scales of machines. We implement the algorithms in a highly parallel and flexible way to ensure the efficiency and optimization potential on diverse systems with customizable configurations. We utilize operations per second (OPS), which is measured in an analytical and systematic approach, as the major metric to quantify the AI performance. We perform evaluations on various systems to ensure the benchmark's stability and scalability, from 4 nodes with 32 NVIDIA Tesla T4 (56.1 Tera-OPS measured), up to 512 nodes with 4096 Huawei Ascend 910 (194.53 Peta-OPS measured), and the results show near-linear weak scalability. With flexible workload and single metric, our benchmark can scale and rank AI-HPC easily.