University of Twente Student Theses

Login
This website will be unavailable due to maintenance December 1st between 8:00 and 12:00 CET.

Enabling faster processing of big-data using GPU decompression

Gorgan, Andrei (2022) Enabling faster processing of big-data using GPU decompression.

[img] PDF
905kB
Abstract:Processing big-data has been shown to have a many fold speedup for GPU hardware, however the process of retrieving ready-to-use data from storage devices still requires a process of decompression, currently performed on the CPU. Due to the increasing computational power of GPUs, the decompression step starves the GPU of data, effectively doing nothing until more data is available to process. This research analyses the minimum required speed of decompression on a GPU, such that offloading the decompression step to the GPU, is faster than traditional methods that utilize the CPU. Results show that GPUs cannot outperform CPUs when considering compression ratio, however the improved parallelism of GPUs allows for a 2 times reduction in decompression times.
Item Type:Essay (Bachelor)
Clients:
CERN, Geneva, Switzerland
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science BSc (56964)
Awards:Best Track Paper Award, Best Track Presentation Award
Link to this item:https://purl.utwente.nl/essays/91724
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page