A hierarchical model of data locality

Chengliang Zhang, Chen Ding, Mitsunori Ogihara, Yutao Zhong, Youfeng Wu

Research output: Contribution to journalArticle

1 Scopus citations

Abstract

In POPL 2002, Petrank and Rawitz showed a universal result-finding optimal data placement is not only NP-hard but also impossible to approximate within a constant factor if P ≠ N P. Here we study a recently published concept called reference affinity, which characterizes a group of data that are always accessed together in computation. On the theoretical side, we give the complexity for finding reference affinity in program traces, using a novel reduction that converts the notion of distance into satisfiability. We also prove that reference affinity automatically captures the hierarchical locality in divide-and-conquer computations including matrix solvers and N-body simulation. The proof establishes formal links between computation patterns in time and locality relations in space. On the practical side, we show that efficient heuristics exist. In particular, we present a sampling method and show that it is more effective than the previously published technique, especially for data that are often but not always accessed together. We show the effect on generated and real traces. These theoretical and empirical results demonstrate that effective data placement is still attainable in general-purpose programs because common (albeit not all) locality patterns can be precisely modeled and efficiently analyzed.

Original languageEnglish (US)
Pages (from-to)16-29
Number of pages14
JournalACM SIGPLAN Notices
Volume41
Issue number1
DOIs
StatePublished - Jun 26 2006
Externally publishedYes

    Fingerprint

Keywords

  • Hierarchical data placement
  • N-body simulation
  • NP-complete
  • Program locality
  • Reference affinity
  • Volume distance

ASJC Scopus subject areas

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this