Task-Based Parallel Programming for Scalable Algorithms: application to Matrix Multiplication
Year: 2022
Type: report
Abstract: Task-based programming models have succeeded in gaining the interest of the high-performance mathematical software community thanks to how they relieve part of the burden of developing and implementing distributed-memory parallel algorithms in an efficient and portable way. In increasingly larger, more heterogeneous clusters of computers, these models appear as a way to maintain and enhance more complex algorithms. However, task-based programming models lack the flexibility and the features that are necessary to express in an elegant and compact way scalable algorithms that rely on advanced communication patterns. We show that the Sequential Task Flow paradigm can be extended to write a compact yet efficient and scalable General Matrix Multiplication. This extension required few modifications to the StarPU runtime system. The final implementation is shown to be competitive up to 32,768 cores with state-of-the-art libraries and may outperform them on some specific problem configurations.
Source: HAL
Language: English
Subfield: Hardware and Architecture
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4226143699