Concurrent CPU-GPU Task Programming using Modern C++

TL;DR

Heteroflow is a modern C++ library that simplifies and accelerates the development of CPU-GPU parallel programs using task dependency graphs, demonstrating significant performance improvements in real applications.

Contribution

This paper introduces Heteroflow, a novel C++ library for efficient heterogeneous CPU-GPU programming with task dependency graphs, enhancing performance and productivity.

Findings

Achieved 7.7x speed-up in VLSI timing analysis

Demonstrated scalability across CPU-GPU configurations

Outperformed existing libraries in efficiency and generality

Abstract

In this paper, we introduce Heteroflow, a new C++ library to help developers quickly write parallel CPU-GPU programs using task dependency graphs. Heteroflow leverages the power of modern C++ and task-based approaches to enable efficient implementations of heterogeneous decomposition strategies. Our new CPU-GPU programming model allows users to express a problem in a way that adapts to effective separation of concerns and expertise encapsulation. Compared with existing libraries, Heteroflow is more cost-efficient in performance scaling, programming productivity, and solution generality. We have evaluated Heteroflow on two real applications in VLSI design automation and demonstrated the performance scalability across different CPU-GPU numbers and problem sizes. At a particular example of VLSI timing analysis with million-scale tasking, Heteroflow achieved 7.7x runtime speed-up (99 vs 13 minutes) over a baseline on a machine of 40 CPU cores and 4 GPUs.