Adaptive Task Aggregation for High-Performance Sparse Solvers on GPUs

• Published in: Proceedings / International Conference on Parallel Architectures and Compilation Techniques pp. 324 - 336
• Main Authors: Helal, Ahmed E., Aji, Ashwin M., Chu, Michael L., Beckmann, Bradford M., Feng, Wu-chun
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.09.2019
• Subjects: Computational modeling; Computer architecture; data dependency; fine-grained parallelism; GPUs; Graphics processing units; Kernel; Processor scheduling; runtime adaptation; scheduling; sparse linear algebra; Task analysis; task parallel execution
• ISSN: 2641-7936
• DOI: 10.1109/PACT.2019.00033

Sparse solvers are heavily used in computational fluid dynamics (CFD), computer-aided design (CAD), and other important application domains. These solvers remain challenging to execute on massively parallel architectures, due to the sequential dependencies between the fine-grained application tasks. In particular, parallel sparse solvers typically suffer from substantial scheduling and dependency-management overheads relative to the compute operations. We propose adaptive task aggregation (ATA) to efficiently execute such irregular computations on GPU architectures via hierarchical dependency management and low-latency task scheduling. On a gamut of representative problems with different data-dependency structures, ATA significantly outperforms existing GPU task-execution approaches, achieving a geometric mean speedup of 2.2X to 3.7X across different sparse kernels (with speedups of up to two orders of magnitude).