Premium
Cache‐efficient renumbering for vectorization
Author(s) -
Löhner Rainald
Publication year - 2010
Publication title -
international journal for numerical methods in biomedical engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.741
H-Index - 63
eISSN - 2040-7947
pISSN - 2040-7939
DOI - 10.1002/cnm.1160
Subject(s) - parallel computing , computer science , vectorization (mathematics) , cache , field (mathematics) , multi core processor , enhanced data rates for gsm evolution , point (geometry) , mathematics , telecommunications , geometry , pure mathematics
A renumbering strategy for field solvers based on unstructured grids that avoids memory contention and minimizes cache‐misses is described. Compared with usual colouring techniques, the new renumbering strategy reduces the spread in point‐data access for edge‐based solvers by more than an order of magnitude. The technique is particularly suited for multicore, cache‐based machines that allow for vectorization or pipelining. Copyright © 2008 John Wiley & Sons, Ltd.