Bi-criteria Pipeline Mappings for Parallel Image Processing

TL;DR

This paper investigates the complex problem of efficiently mapping pipelined JPEG encoding workflows onto parallel platforms, focusing on optimizing throughput and latency through heuristic methods due to the problem's NP-completeness.

Contribution

It introduces polynomial heuristics for bi-criteria pipeline mapping in parallel image processing, specifically for JPEG encoding, addressing an NP-complete problem.

Findings

Heuristics provide effective solutions for complex mapping problems.

Mapping JPEG encoding onto parallel platforms improves processing efficiency.

The bi-criteria approach balances throughput and latency effectively.

Abstract

Mapping workflow applications onto parallel platforms is a challenging problem, even for simple application patterns such as pipeline graphs. Several antagonistic criteria should be optimized, such as throughput and latency (or a combination). Typical applications include digital image processing, where images are processed in steady-state mode. In this paper, we study the mapping of a particular image processing application, the JPEG encoding. Mapping pipelined JPEG encoding onto parallel platforms is useful for instance for encoding Motion JPEG images. As the bi-criteria mapping problem is NP-complete, we concentrate on the evaluation and performance of polynomial heuristics.