<--- Back to Details
First PageDocument Content
Graphics hardware / Video cards / Parallel computing / Nvidia / OpenCL / CUDA / Graphics processing unit / Memory barrier / Comparison of Nvidia graphics processing units / Computer hardware / Computing / GPGPU
Date: 2013-07-28 00:09:19
Graphics hardware
Video cards
Parallel computing
Nvidia
OpenCL
CUDA
Graphics processing unit
Memory barrier
Comparison of Nvidia graphics processing units
Computer hardware
Computing
GPGPU

Optimizing OpenCL for NVIDIA GPUs

Add to Reading List

Source URL: www.hotchips.org

Download Document from Source Website

File Size: 1,09 MB

Share Document on Facebook

Similar Documents

Computing / Computer architecture / Computer memory / Concurrency / Parallel computing / Concurrent computing / Memory barrier / Synchronization / Data dependency / Microarchitecture / Register renaming / Memory model

Understanding POWER Multiprocessors Susmit Sarkar1 1 Peter Sewell1

DocID: 1rpAG - View Document

Software engineering / Software / Computing / Memory management / Subroutines / Automatic memory management / Garbage collection / Jikes RVM / Computing platforms / Reference counting / Garbage / Write barrier

University of London Imperial College London of Science, Technology and Medicine Department of Computing Soft Real-time Garbage Collection for Dynamic Dispatch Languages

DocID: 1rlOP - View Document

Computing / Computer architecture / Software engineering / X86 architecture / Computer memory / Low-level programming language / Spinlock / X86 / Assembly language / Memory barrier / Memory ordering / Exit

The Semantics of x86-CC Multiprocessor Machine Code Susmit Sarkar1 Scott Owens1 Tom Ridge1

DocID: 1r4yz - View Document

Computing / Computer architecture / Computer memory / Transaction processing / Concurrency / Cache coherency / MESI protocol / Memory ordering / Consistency model / Cache coherence / Linearizability / Memory barrier

Review of last lecture  Architecture case studies  Memory performance is often the bottleneck  Parallelism grows with compute performance

DocID: 1qSE1 - View Document

Computing / Computer memory / Memory barrier / Static single assignment form / Symbolic execution / Computer engineering

Partial Orders for Efficient Bounded Model Checking of Concurrent Software? Jade Alglave1 , Daniel Kroening2 , and Michael Tautschnig3 1 3

DocID: 1qKim - View Document