Performance Analysis of Shared-Memory Parallel Applications Using Performance Properties [chapter]

Karl Fürlinger, Michael Gerndt
2005 Lecture Notes in Computer Science  
Tuning parallel code can be a time-consuming and difficult task. We present our approach to automate the performance analysis of shared-memory parallel applications that is based on the notion of performance properties. Properties are formally specified in the ASL (APART specification language) with respect to a certain data model. We describe a data model for summary (profiling) data of OpenMP code and present performance properties that are based on this data model. We evaluate the usability
more » ... f the properties with on several example codes using our OpenMP profiler ompP to derive the necessary profiling data.
doi:10.1007/11557654_70 fatcat:d6xv7tnnand2hkntyjunxvzoky