Good news, GPU computing developers. Nvidia has announced that a release candidate of its CUDA 4.0 developer toolkit will be available later today. The new toolkit will include several interesting additions and, in Nvidia's words, "was designed to make parallel programming easier."
Major new features will include:
- NVIDIA GPUDirect™ 2.0 Technology -- Offers support for peer-to-peer communication among GPUs within a single server or workstation. This enables easier and faster multi-GPU programming and application performance.
- Unified Virtual Addressing (UVA) -- Provides a single merged-memory address space for the main system memory and the GPU memories, enabling quicker and easier parallel programming.
- Thrust C++ Template Performance Primitives Libraries -- Provides a collection of powerful open source C++ parallel algorithms and data structures that ease programming for C++ developers. With Thrust, routines such as parallel sorting are 5X to 100X faster than with Standard Template Library (STL) and Threading Building Blocks (TBB).
I'm no GPU computing expert, but it sounds like these changes could bring about some nice performance boosts to CUDA apps. There are more additions in store, too, as you'll see in the official announcement.
The CUDA 4.0 RC toolkit should be available to members of Nvidia's CUDA Registered Developer Program from the Nvidia Developer Zone site. If you haven't joined the developer program yet, you can do so here.