Platforma za heterogenu obradu velikih skupova podataka

master's thesis

• Title (english): Platform for the Heterogeneous Processing of Large Datasets
• Author: Luka Bušić
• Mentor: Vlado Sruk (mentor)
• Committee member: Vlado Sruk (predsjednik povjerenstva)
• Committee member: Leonardo Jelenković (član povjerenstva)
• Committee member: Dejan Škvorc (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-15, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Rad razrađuje potencijalnu potrebu korištenja grafičkih procesorskih jedinica pri obradi i vizualizaciji velikih skupova podataka. Prvo su navedene razlike grafičkih i centralnih procesorskih jedinica pa se opisao način na koji su se potonji upotrijebili u opisu praktičnog dijela rada. Poanta je bila obraditi podatke na efikasan način te ih prikazati. To se učinilo uporabom okruženja OpenCL koje omogućuje pristup korištenju grafičkih procesorskih jedinica. OpenCL okruženje je detaljno obrađeno kao i priprema podataka na CPU strani, baza podataka koja se koristila, implementacija obrade podataka i vizualizacija. U posljednjem poglavlju kod opisa rezultata su se napravile analize i usporedbe koje su se činile ključnima. Neke od obrađenih stvari u tom poglavlju su usporedba vremena izvršavanja promjenom parametara broja radnika po grupama, prednosti korištenja lokalne memorije, analize vremena izvršavanja sa različitim grafičkim uređajima, vizualizacija podataka i dr.

Abstract (english)

The paper elaborates the potential need for the use of graphics processing units in the processing and visualization of large data sets. The differences between the graphics and central processing units are first mentioned, and the way in which the latter were used. The point was to process the data efficiently and display it. This was done using an OpenCL environment that provides access to the use of GPUs. The OpenCL environment is covered in detail as well as the preparation of data on the CPU side, the database used, the implementation of data processing and visualization. In the last chapter, analyzes and comparisons were made that seemed crucial. Some of the things covered in this chapter are the comparison of execution time by changing the parameters of the number of workers by groups, the advantages of using local memory, the analysis of execution time with different graphics devices, data visualization, etc.

• Keywords: GPGPU; OpenCL; paralelizacija; kernel
• Keywords (english): GPGPU; OpenCL; paralelization; kernel
• Language: croatian
• URN:NBN: urn:nbn:hr:168:386658
• 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: 2023-01-13 09:05:11