20.500.12556/RUL-91250 Paralelizacija Gram–Schmidtovega algoritma na sistemu Intel Xeon Phi Parallelization of Gram-Schmidt algorithm on Intel Xeon Phi system Gram-Schmidtov postopek je klasičen postopek za ortonormiranje množice vektorjev v vektorskem prostoru s skalarnim produktom. Obstaja več imple- mentacij Gram-Schmidtovega postopka. Klasičen, modificiran in iterativno klasičen Gram-Schmidtov postopek. Izbrali smo klasičen Gram-Schmidtov postopek. Paralelizacijo klasičnega Gram-Schmidtovega postopka smo re- alizirali na koprocesorju Intel Xeon Phi. V programskem jeziku C smo s pomočjo knjižnice OpenMP realizirali več funkcij klasičnega Gram-Schmidtovega postopka. Izkazalo se je, da je najboljša funkcija tista, ki enakomerno poraz- deli delo med vse niti. Vsaka nit najprej vzporedno izračuna enako število projekcij, nato pa se preostale projekcije izvedejo zaporedno, kjer vse niti vzporedno izvedejo eno projekcijo. Implementacija na Intel Xeon Phi je do trikrat hitrejša kot na dveh Intel Xeon procesorjih. Faktor pohitritve se zvišuje z velikostjo vhodne množice vektorjev. The Gram-Schmidt process is a classical process for orthonormalising a set of vectors in inner product space. There are many implementations of the Gram-Schmidt process - classical, modified and iterated classical Gram- Schmidt process. In the thesis we have chosen the classical Gram-Schmidt process. Parallelization of classical Gram-Schmidt process was implemented on Intel Xeon Phi coprocessor. We have implemented many different func- tions of classical Gram-Schmidt process in the C programming language with a help of OpenMP library. The results show that the best function is the one that evenly distributes work among all threads. First every thread calculates the same amount of projections, then the rest of projections are calculated sequentially, where the threads compute each projection in parallel. The implementation of the same function on the Intel Xeon Phi is up to three times faster than the same implementation on two Intel Xeon processors. The speedup factor is increasing with the size of input vectors. paralelizacija Gram-Schmidt Intel Xeon Phi parallelization Gram-Schmidt Intel Xeon Phi true false false Slovenski jezik Angleški jezik Magistrsko delo/naloga 2017-03-27 13:08:26 2017-03-27 13:08:27 2022-08-09 21:22:46 0000-00-00 00:00:00 2016 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 1970-01-01 17441 Mladovan_Domen_-_Paralelizacija_Gram–Schmidtovega_algoritma_na_sistemu_Intel_Xeon_Phi.pdf Mladovan_Domen_-_Paralelizacija_Gram–Schmidtovega_algoritma_na_sistemu_Intel_Xeon_Phi.pdf 1 533FB081F5C2C56302DEFA0490101D99 fa4c78c36f542c95da175186c8da95fd4cfd45401329c0e848947f42d022af06 6195a611-a1b2-11eb-a523-00155dcfd717 20.500.12556/rul/391e91c5-25b7-41cc-a4c8-163a01f268d2 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=95436 Fakulteta za računalništvo in informatiko 0 0 0