Adaptive Runtime Filtering: Reducing Trace Size and Bias in Event-Based Performance Analysis

Stolle, Jonas; Wagner, Michael; Doleschal, Jens; Schmitt, Felix; Brunst, Holger

doi:10.1109/cse.2015.43

Cited by 1 publication

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Score-P can filter function calls based on their occurrence, i.e., the user can specify a value n so that all calls to a function are filtered after this function is called n times [2]. In addition, Stolle et al introduce a set of adaptive filter techniques that operate during runtime [13]. Score-P also supports a rewind feature that allows to statically or dynamically filter complete program phases, e.g., single iteration steps [2].…”

Section: Related Workmentioning

confidence: 99%

Automatic Adaption of the Sampling Frequency for Detailed Performance Analysis

Wagner

Knuepfer²

2017

2017 17th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID)

Self Cite

View full text Add to dashboard Cite

One of the most urgent challenges in event based performance analysis is the enormous amount of collected data. Combining event tracing and periodic sampling has been a successful approach to allow a detailed event-based recording of MPI communication and a coarse recording of the remaining application with periodic sampling. In this paper, we present a novel approach to automatically adapt the sampling frequency during runtime to the given amount of buffer space, releasing users to find an appropriate sampling frequency themselves. This way, the entire measurement can be kept within a single memory buffer, which avoids disruptive intermediate memory buffer flushes, excessive data volumes, and measurement delays due to slow file system interaction. We describe our approach to sort and store samples based on their order of occurrence in an hierarchical array based on powers of two. Furthermore, we evaluate the feasibility as well as the overhead of the approach with the prototype implementation OTFX based on the Open Trace Format 2, a state-of-the-art Open Source event trace library used by the performance analysis tools Vampir, Scalasca, and Tau.This work is supported by the Spanish Ministry of Economy and Competitiveness under contract TIN2015-65316-P.Peer ReviewedPostprint (author's final draft

show abstract

Section: Related Workmentioning

confidence: 99%