Optimizacija prioritetnog raspoređivanja metodama strojnog učenja ostvarena na grafičkom procesoru

master's thesis

• Title (english): Optimization of Priority Scheduling Using Machine Learning on GPU
• Author: Željko Findak
• Mentor: Domagoj Jakobović (mentor)
• Committee member: Domagoj Jakobović (predsjednik povjerenstva)
• Committee member: Goran Delač (član povjerenstva)
• Committee member: Stjepan Groš (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2018-07-11, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu opisan je problem optimizacije prioritetnog raspoređivanja u okolini nesrodnih strojeva. Opisane su metode strojnog učenja koje se koriste u generiranju rasporeda, genetsko programiranje i neuronska mreža. Opisana je arhitektura grafičkog procesora i tehnologija CUDA. Opisane su tehnike paralelizacije evaluacije rješenja navedenih metoda. Implementirana je programska podrška za izvođenje navedenih metoda na grafičkom procesoru u sklopu programskog okvira ECF. Izmjereni su rezultati ubrzanja i komentirani problemi koji se javljaju.

Abstract (english)

This thesis describes the problem of priority scheduling optimization in the environment of unrelated machines. It also describes the machine learning methods, genetic programing and neural networks used for generating schedules. Furthermore, this thesis describes the architecture of a graphics processor and the CUDA technology, as well as the paralelization techniques that are used for the solution evaluation of these methods. Software support for GPU execution is implemented within the ECF framework. Finally, the thesis reports on speedup measures and comments on the problems that arise.

• Keywords: prioritetno raspoređivanje; CUDA; GPU; optimizacija; neuronske mreže; genetsko programiranje; nesrodni strojevi
• Keywords (english): priority scheduling; CUDA; GPU; optimization; neural networks; genetic programming; unrelated machines
• Language: croatian
• URN:NBN: urn:nbn:hr:168:239870
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 20:14:41