by Anton Shilov
07/23/2010 | 05:04 AM
Nvidia Corp. this week announced the release of Nvidia Parallel Nsight software, the industry's first development environment for graphics processing unit (GPU) -accelerated applications that work with Microsoft Visual Studio. The company has also released CUDA Toolkit 3.1, an update to its CUDA software development kit (SDK).
"With more than six million developers, Visual Studio is one of the world's most popular development environments for Windows-based applications and services. By adding functionality specifically for GPU Computing developers, Parallel Nsight makes the power of the GPU more accessible than ever before," said Sanford Russell, general manager of GPU Computing at Nvidia.
Visual Studio developers can now use Parallel Nsight to debug CUDA C/C++, or DirectCompute applications on the GPU using the same familiar tools and techniques as on the CPU. Parallel Nsight also provides the analysis tools that give developers the information required to achieve the highest levels of GPGPU application performance. Nvidia Parallel Nsight software supports Windows HPC Server 2008, Windows 7 and Windows Vista, which works closely with Visual Studio 2008 SP1 Professional edition.
"By integrating GPU computing into Visual Studio, Nvidia's Parallel Nsight is transforming the way GPU-based parallel computing applications are developed for Windows," said David Rich, director, technical computing at Microsoft Corp.
Parallel Nsight also represents the premier environment for graphics development, delivering the DirectX 11 tools required by graphics developers to quickly and efficiently develop today's top game titles and visual computing applications. Parallel Nsight provides several specific features for graphics developers. The Graphics Debugger allows developers to debug all HLSL graphics shaders directly on the GPU. The Graphics Inspector allows real-time examination of DirectX rendering calls and GPU pipeline state to craft optimal GPU code, and Pixel History shows all operations that affect a given pixel, helping developers to quickly identify rendering errors and to create optimal rendering code. In addition, Parallel Nsight Analyzer allows OpenGL developers to optimize applications by presenting OpenGL API trace information on a timeline correlated with other GPU and CPU events.
High performance computing (HPC) developers in fields such as oil and gas, bioscience and finance, are increasingly tapping the massively parallel processing capabilities of GPUs and Windows HPC Server 2008 to achieve significant performance increases in their work. The Parallel Nsight connects these worlds, providing the professional tools required to develop and deploy HPC cluster applications.