Issue No. 10 - October (1994 vol. 5)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/71.313127
<p>We present a classification scheme for array language primitives that quantifies the variation in parallelism and data locality that results from the fusion of any two primitives. We also present an algorithm based on this scheme that efficiently determines when it is beneficial to fuse any two primitives. Experimental results show that five LINPACK routines report 50% performance improvement from the fusion of array operators.</p>
Index Termsparallel programming; program compilers; parallel languages; optimisation; programming theory; array language primitives; classification; loop fusion; parallelization; classification scheme; data locality; LINPACK routines; performance improvement; array operators; compiler optimisation; data parallelism
P. Carini, C. Wu and D. Ju, "The Classification, Fusion, and Parallelization of Array Language Primitives," in IEEE Transactions on Parallel & Distributed Systems, vol. 5, no. , pp. 1113-1120, 1994.