Region Templates: Data Representation and Management for Large-Scale Image Analysis

TL;DR

This paper introduces a region template abstraction for efficient data management in large-scale high-resolution image analysis on hybrid CPU-GPU clusters, enabling scalable and optimized processing with minimal overhead.

Contribution

The paper presents a novel region template abstraction and runtime system that simplify data management and optimize execution for large-scale image analysis on hybrid computing clusters.

Findings

Negligible overhead of about 3% introduced by the abstraction

Good scalability demonstrated on a hybrid cluster

Effective performance-aware scheduling and data transfer reduction techniques

Abstract

Distributed memory machines equipped with CPUs and GPUs (hybrid computing nodes) are hard to program because of the multiple layers of memory and heterogeneous computing configurations. In this paper, we introduce a region template abstraction for the efficient management of common data types used in analysis of large datasets of high resolution images on clusters of hybrid computing nodes. The region template provides a generic container template for common data structures, such as points, arrays, regions, and object sets, within a spatial and temporal bounding box. The region template abstraction enables different data management strategies and data I/O implementations, while providing a homogeneous, unified interface to the application for data storage and retrieval. The execution of region templates applications is coordinated by a runtime system that supports efficient execution in hybrid machines. Region templates applications are represented as hierarchical dataflow in which each computing stage may be represented as another dataflow of finer-grain tasks. A number of optimizations for hybrid machines are available in our runtime system, including performance-aware scheduling for maximizing utilization of computing devices and techniques to reduce impact of data transfers between CPUs and GPUs. An experimental evaluation on a state-of-the-art hybrid cluster using a microscopy imaging study shows that this abstraction adds negligible overhead (about 3%) and achieves good scalability.