Go to  Advanced Search

Please note that cIRcle is currently being upgraded to DSpace v5.1. The upgrade means that the cIRcle service will not be accepting new material from 05:00 on September 1/15 until 08:00 on September 4/15. Read only access will be available during this period. Apologies for the inconvenience.

Dynamic warp formation : exploiting thread scheduling for efficient MIMD control flow on SIMD graphics hardware

Show full item record

Files in this item

Files Size Format Description   View
ubc_2008_fall_fung_wilson_wai_lun.pdf 1.157Mb Adobe Portable Document Format   View/Open
Title: Dynamic warp formation : exploiting thread scheduling for efficient MIMD control flow on SIMD graphics hardware
Author: Fung, Wilson Wai Lun
Degree Master of Applied Science - MASc
Program Electrical and Computer Engineering
Copyright Date: 2008
Publicly Available in cIRcle 2008-09-18
Subject Keywords GPU; SIMD; Control flow; Graphics processing unit
Abstract: Recent advances in graphics processing units (GPUs) have resulted in massively parallel hardware that is easily programmable and widely available in commodity desktop computer systems. GPUs typically use single-instruction, multiple-data (SIMD) pipelines to achieve high performance with minimal overhead for control hardware. Scalar threads running the same computing kernel are grouped together into SIMD batches, sometimes referred to as warps. While SIMD is ideally suited for simple programs, recent GPUs include control flow instructions in the GPU instruction set architecture and programs using these instructions may experience reduced performance due to the way branch execution is supported by hardware. One solution is to add a stack to allow different SIMD processing elements to execute distinct program paths after a branch instruction. The occurrence of diverging branch outcomes for different processing elements significantly degrades performance using this approach. In this thesis, we propose dynamic warp formation and scheduling, a mechanism for more efficient SIMD branch execution on GPUs. It dynamically regroups threads into new warps on the fly following the occurrence of diverging branch outcomes. We show that a realistic hardware implementation of this mechanism improves performance by an average of 47% for an estimated area increase of 8%.
URI: http://hdl.handle.net/2429/2268
Scholarly Level: Graduate

This item appears in the following Collection(s)

Show full item record

All items in cIRcle are protected by copyright, with all rights reserved.

UBC Library
1961 East Mall
Vancouver, B.C.
Canada V6T 1Z1
Tel: 604-822-6375
Fax: 604-822-3893