Featured Articles

AMD Never Settle Forever bundle hits 200-series cards

AMD Never Settle Forever bundle hits 200-series cards

AMD’s Never Settle bundles have been around for a while and the community response has been extremely positive. When AMD launched…

More...
AMD shipping Beema APUs

AMD shipping Beema APUs

According to Lisa Su, SVP & GM, Global Business Units at AMD, Beema notebook parts have started shipping to manufacturers last…

More...
IHS teardown reveals Galaxy S5 BOM

IHS teardown reveals Galaxy S5 BOM

Research firm IHS got hold of Samsung’s new flagship smartphone and took it apart to the last bolt to figure out…

More...
Galaxy S5, HTC One M8 available selling well

Galaxy S5, HTC One M8 available selling well

Samsung’s Galaxy S5 has finally gone on sale and it can be yours for €699, which is quite a lot of…

More...
KFA2 GTX 780 Ti Hall Of Fame reviewed

KFA2 GTX 780 Ti Hall Of Fame reviewed

KFA2 gained a lot of overclocking experience with the GTX 780 Hall of Fame (HOF), which we had a chance to…

More...
Frontpage Slideshow | Copyright © 2006-2010 orks, a business unit of Nuevvo Webware Ltd.
Thursday, 14 November 2013 21:15

Nvidia announces new CUDA 6

Written by Slobodan Simic

nvidia cuda-logo

Unified memory for CUDA

Ahead of the annual International Conference for High Performance Computing, Networking, Storage, and Analysis, better known as the SC, Nvidia announced its new version of CUDA, CUDA 6. While most recent updates to CUDA had quite a few new features, the CUDA 6 will mostly focus on a few elements that might be quite important.

The bigest news is definitely the unified memory support wihtin CUDA. While CUDA 4 already had unified virtual addressing support which allowed x86 and GPU memory to be addressed together in a single space, it only simplified memory management and did not solve some issues that such a solution had and still required memory copying that will allow GPU to work with it.

With CUDA 6, Nvidia moved towards removing issues like memory copying by implementing a unified memory system on top of the existing memory pool structure. Basically, unified memory allows programmers to access any resource or addres within the legal address space, regardless of the place where it is in and actually use it without the need to copy it over.

While it does not resolve technical limitations that require memory copying, like limited bandwidth and latency of the PCIe, it moves memory management to CUDA rather than other toolkit, thus simply making the life of the programmer much easier.

Basically it does not solve any performance issues but rather simplifes CUDA programming. This move is made in order to make CUDA programming more accessible to wider audiences.

Nvidia did not shed any light regarding performance impact at this time, and while manual memory management is still present, CUDA unified memory will definitely have some kind of performance penalty.

Nvidia also revealed that while we are currently seeing software implementation of unified memory with CUDA 6, Maxwell will most likely have some sort of hardware functionality for implementing unified memory thus removing any performance penalties and most likely doing it much better than software solution with CUDA 6. 

Via Anandtech.com.

nvidia CUDA6unifiedmemory-1

nvidia CUDA6unifiedmemory-2

Last modified on Friday, 15 November 2013 08:27
blog comments powered by Disqus

To be able to post comments please log-in with Disqus

 

Facebook activity

Latest Commented Articles

Recent Comments