NVIDIA Buys PGI Compiler Company To Help OpenACC
NVIDIA announced earlier this week that they have acquired The Portland Group (PGI), a company well-respected for their code compilers.
The Portland Group company has been around for more than two decades to focus on high-performance compilers for the Fortran, C, and C++ programming languages. While not too much is heard about PGI's compilers within enthusiast circles, they have a lot of respect for their HPC compilers and in recent years their GPGPU initiatives. The Portland Group collaborated with NVIDIA over the CUDA Fortran implementation and they have also been involved with OpenACC for GPU programming and OpenCL on ARM.
For advancing their GPGPU computing possibilities, NVIDIA has acquired The Portland Group from STMicroelectronics.
In terms of NVIDIA's intention, their blog post says, "Bringing our teams together further cements our strong, established technical partnership in creating developer tools for the accelerated computing revolution. It also strengthens the OpenACC initiative to create an easy on-ramp to parallel computing by joining the world’s top independent provider of OpenACC compilers with the world’s best GPU designers."
OpenACC is a programming standard for CPU/GPU programming that is similar in nature to OpenMP and supports Fortran/C/C++ languages. OpenACC may merge into OpenMP, but for now there really isn't much OpenACC love found in the open-source compilers.
The Portland Group company has been around for more than two decades to focus on high-performance compilers for the Fortran, C, and C++ programming languages. While not too much is heard about PGI's compilers within enthusiast circles, they have a lot of respect for their HPC compilers and in recent years their GPGPU initiatives. The Portland Group collaborated with NVIDIA over the CUDA Fortran implementation and they have also been involved with OpenACC for GPU programming and OpenCL on ARM.
For advancing their GPGPU computing possibilities, NVIDIA has acquired The Portland Group from STMicroelectronics.
In terms of NVIDIA's intention, their blog post says, "Bringing our teams together further cements our strong, established technical partnership in creating developer tools for the accelerated computing revolution. It also strengthens the OpenACC initiative to create an easy on-ramp to parallel computing by joining the world’s top independent provider of OpenACC compilers with the world’s best GPU designers."
OpenACC is a programming standard for CPU/GPU programming that is similar in nature to OpenMP and supports Fortran/C/C++ languages. OpenACC may merge into OpenMP, but for now there really isn't much OpenACC love found in the open-source compilers.
3 Comments