Skip to content

kongsally/Project2-Stream-Compaction

 
 

Repository files navigation

CUDA Stream Compaction

University of Pennsylvania, CIS 565: GPU Programming and Architecture, Project 2

  • Sally Kong
  • Tested on: Windows 8, i7-5500U CPU @ 2.40GHz 2.40 GHz, GEForce 920M (Personal)

Summary: This project is an implementation of a GPU stream compaction in CUDA, from scratch. This is a widely used algorithm that I later plan to use to accelerate my path tracer.

A few different versions of the Scan (Prefix Sum) algorithm were implemented: a CPU version, and a few GPU implementations: "naive" and "work-efficient."

Algorithm overview & details: There are two primary references for details on the implementation of scan and stream compaction.

Performance Analysis

About

CIS 565 Project 2: Stream Compaction

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • CMake 88.4%
  • Cuda 6.3%
  • C++ 4.8%
  • Makefile 0.5%