EDA Blog - electronic design automation, embedded systems, ic

Portland Group CUDA C and C++ Compilers for x86

The Portland Group is shipping the PGI CUDA C and C++ compilers for systems based on the industry standard general-purpose 64-bit and 32-bit x86 architectures. CUDA is NVIDIA’s parallel computing architecture. It increases computing performance by harnessing the power of NVIDIA GPUs (graphics processing units). The Portland Group is a wholly-owned subsidiary of STMicroelectronics.

PGI CUDA C/C++ Compilers for x86

• For systems based on 64-bit and 32-bit x86 architectures
• Developers can use the CUDA parallel programming model to optimize the performance of the critical parts of the code base
• Engineers can target servers and clusters with or without NVIDIA GPUs
• Designers can run CUDA apps on any x86 clusters
• PGI CUDA C/C++ x86 applications perform parallel execution by using multiple processor cores
• Supports Streaming SIMD (Single Instruction Multiple Data) Extensions (SSE), including the new AVX instructions available on the latest generation of x86 compatible CPUs from Intel and AMD
• Included as part of the PGI Accelerator product line
• PGI Accelerator compilers are available for C99 and Fortran 2003
• x86 CUDA C/C++ compilers will be released in three phases
  • Phase 1 (available now) demonstrates the capabilities of the technology and enables developers to begin working with the compilers
  • Phase 2 (fourth quarter of 2011) will include performance optimizations intended to extract maximum performance of CUDA programs running on the x86 target platform
  • Phase 3 (mid 2012) will support for PGI Unified Binary technology — the ability to run one executable on both CPUs and GPUs

More info: The Portland Group | NVIDIA CUDA | STMicroelectronics

Custom Search

If you are familiar with RSS feeds, you can also sign up for our free blog feed. Our RSS feed is updated in real-time while our newsletter is updated daily.