Wyniki wyszukiwania - Biblioteka Nauki

Ten serwis zostanie wyłączony 2025-02-11.

Nowa wersja platformy, zawierająca wyłącznie zasoby pełnotekstowe, jest już dostępna.
Przejdź na https://bibliotekanauki.pl

Ograniczanie wyników

1 Annals of Computer Science and Information Systems

1 2019

Znaleziono wyników: 1

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: program control structures

Sortuj według:

Ogranicz wyniki do:

Parallel cache-efficient code for computing the McCaskill partition functions

100%

Palkowski M. , Bielecki W.

Annals of Computer Science and Information Systems

2019

tom Vol. 18

207--210

We present parallel tiled optimized McCaskill's partition functions computation code. That CPU and memory intensive dynamic programming task is within computational biology. To optimize code, we use the authorial source-to-source TRACO compiler and compare obtained code performance with that generated with the state-of-the-art PluTo compiler based on the affine transformations framework (ATF). For the considered task, PluTo is able to generate only serial highly cache efficient code without any parallelism. A TRACO tiling and parallelizing strategy uses the transitive closure of a dependence graph to avoid affine function calculation. First, for each loop nest statement, rectangular tiles are formed. Then those tiles are corrected to be valid under lexicographical order if necessary. A correction is carried out by means of applying transitive closure. The validity of tiles guarantees that the inter-tile dependence graph is acyclic. So, a valid schedule for target tiles can be derived and applied to generate parallel tiled code. For this purpose, the ISL scheduler is used. An experimental study carried out on a multi-core computer demonstrates considerable speed-up of generated code for the larger number of threads. Generated parallel tiled code overcomes that generated with the PluTo compiler.