Scalable Monitoring Technique for Detecting Races in Parallel Programs

Detecting races is important for debugging shared-memory parallel programs, because the races result in unintended nondeterministic executions of the programs. Previous on-the- y techniques to detect races have a bottleneck caused by the need to check or serialize all accesses to each shared variable in a program that may have nested parallelism with barrier synchronization. The new scalable monitoring technique in this paper reduces the bottleneck signi cantly by checking or serializing at most 2B + 1 non-nested accesses in an iteration for each shared variable, where B is the number of barrier operations in the iteration. This technique, therefore, makes on-the- y race detection more scalable.

• Type: Conference
• Year: 2000
• Where: IPPS
• Authors: Yong-Kee Jun, Charles E. McDowell
• Keywords: Distributed And Parallel Computing, IPPS 2000, Scalable Monitoring Technique, Shared-memory Parallel Programs, Unintended Nondeterministic Executions