Skip to main content

Locality analysis through static parallel sampling

Dong Chen, Fangzhou Liu, Chen Ding, Sreepathi Pai

PDF DOI 10.1145/3192366.3192402 dblp BibTeX

Abstract

Locality analysis is important since accessing memory is much slower than computing. Compile-time locality analysis can provide detailed program-level feedback for compilers or runtime systems faster than trace-based locality analysis.

Related papers

PLUM: static parallel program locality analysis under uniform multiplexing PPoPP 2020
The garbage collection advantage: improving program locality OOPSLA 2004
Sampling-based program locality approximation ISMM 2008
Using sample-based time series data for automated diagnosis of scalability losses in parallel programs PPoPP 2020
Statically bounded-memory delayed sampling for probabilistic streams OOPSLA 2021
Region-Based Data Layout via Data Reuse Analysis CC 2024
Abstract Interpretation with Confidence: Quantifying the Precision of Dataflow Analysis with Probabilities PLDI 2026
Scalable data race detection for partitioned global address space programs PPoPP 2013