ScoRD: A Scoped Race Detector for GPUs

Kamath, AK and George, AA and Basu, A (2020) ScoRD: A Scoped Race Detector for GPUs. In: Proceedings - International Symposium on Computer Architecture, 30 May-3 June 2020, Virtual, Online; Spain, pp. 1036-1049.

PDF pro_int_sym_com_arc_2020.pdf - Published Version Restricted to Registered users only Download (943kB) | Request a copy

Abstract

GPUs have emerged as a key computing platform for an ever-growing range of applications. Unlike traditional bulk-synchronous GPU programs, many emerging GPU-accelerated applications, such as graph processing, have irregular interaction among the concurrent threads. Consequently, they need complex synchronization. To enable both high performance and adequate synchronization, GPU vendors have introduced scoped synchronization operations that allow a programmer to synchronize within a subset of concurrent threads (a.k. a., scope) that she deems adequate. Scoped-synchronization avoids the performance overhead of synchronization across thousands of GPU threads while ensuring correctness when used appropriately. This flexibility, however, could be a new source of incorrect synchronization where a race can occur due to insufficient scope of the synchronization operation, and not due to missing synchronization as in a typical race. We introduce ScoRD, a race detector that enables hardware support for efficiently detecting global memory races in a GPU program, including those that arise due to insufficient scopes of synchronization operations. We show that ScoRD can detect a variety of races with a modest performance overhead (on average, 35). In the process of this study, we also created a benchmark suite consisting of seven applications and three categories of microbenchmarks that use scoped synchronization operations. © 2020 IEEE.

Item Type:	Conference Paper
Publication:	Proceedings - International Symposium on Computer Architecture
Publisher:	Institute of Electrical and Electronics Engineers Inc.
Additional Information:	cited By 0; Conference of 47th ACM/IEEE Annual International Symposium on Computer Architecture, ISCA 2020 ; Conference Date: 30 May 2020 Through 3 June 2020; Conference Code:161833
Keywords:	Application programs; Benchmarking; Computer hardware; Graphics processing unit; Program processors, Benchmark suites; Complex synchronization; Computing platform; Concurrent threads; Hardware supports; Micro-benchmarks; Synchronization operation; Three categories, Synchronization
Department/Centre:	Division of Electrical Sciences > Computer Science & Automation
Date Deposited:	05 Nov 2020 11:42
Last Modified:	05 Nov 2020 11:42
URI:	http://eprints.iisc.ac.in/id/eprint/67002

Actions (login required)

View Item