The CUDA 6 Production Release is now available for download at [url]http://www.nvidia.com/getcuda[/url].
It includes features that further simplify parallel programming like:
• Unified Memory: simplifies programming by enabling applications to access the CPU and GPU memory without the need to manually copy data.
• Drop-in Libraries: automatically accelerate the linear algebra and FFTs in your applications by replacing CPU-only libraries with GPU-accelerated libraries.
• Multi-GPU Scaling: the re-designed BLAS GPU library automatically scales performance across up to 8 GPUs in a single node.
Learn more about these features by signing up to attend a live webinar.
• Webinar: CUDA 6 - Feature Overview
Wednesday, April 23, 2014 at 11 AM - 12 PM PST : [url]https://www2.gotomeeting.com/register/545712674[/url]
• CUDA 6: Unified Memory
Thursday, May 1, 2014 at 10 AM - 11 AM PST : [url]https://www2.gotomeeting.com/register/738246234[/url]
• CUDA 6: Drop-in Performance Optimized Libraries
Wednesday, May 7, 2014 at 10 AM - 11 AM PST : [url]https://www2.gotomeeting.com/register/857096514[/url]
• CUDA 6: Performance Review :
Wednesday, May 14, 2014 at 10 AM - 11 AM PST : [url]https://www2.gotomeeting.com/register/906009186[/url]
For more information on CUDA 6, check out the Parallel Forall blog post : [url]http://devblogs.nvidia.com/parallelforall/powerful-new-features-cuda-6/[/url]