Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.757157
Title: A semantically aware transactional concurrency control for GPGPU computing
Author: Shen, Qi
ISNI:       0000 0004 7429 980X
Awarding Body: Newcastle University
Current Institution: University of Newcastle upon Tyne
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
The increased parallel nature of the GPU affords an opportunity for the exploration of multi-thread computing. With the availability of GPU has recently expanded into the area of general purpose programming, a concurrency control is required to exploit parallelism as well as maintaining the correctness of program. Transactional memory, which is a generalised solution for concurrent conflict, meanwhile allow application programmers to develop concurrent code in a more intuitive manner, is superior to pessimistic concurrency control for general use. The most important component in any transactional memory technique is the policy to solve conflicts on shared data, namely the contention management policy. The work presented in this thesis aims to develop a software transactional memory model which can solve both concurrent conflict and semantic conflict at the same time for the GPU. The technique differs from existing CPU approaches on account of the different essential execution paths and hardware basis, plus much more parallel resources are available on the GPU. We demonstrate that both concurrent conflicts and semantic conflicts can be resolved in a particular contention management policy for the GPU, with a different application of locks and priorities from the CPU. The basic problem and a software transactional memory solution idea is proposed first. An implementation is then presented based on the execution mode of this model. After that, we extend this system to resolve semantic conflict at the same time. Results are provided finally, which compare the performance of our solution with an established GPU software transactional memory and a famous CPU transactional memory, at varying levels of concurrent and semantic conflicts.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.757157  DOI: Not available
Share: