Today NVIDIA announced the availability of PGI 2018. PGI compilers & tools are used by scientists and engineers who develop applications for high-performance computing systems. They deliver world-class multicore CPU performance, an easy on-ramp to GPU computing with OpenACC directives, and performance portability across all major HPC platforms.
PGI is the default compiler on many of the world’s fastest computers including the Titan supercomputer at Oak Ridge National Laboratory. PGI production-quality compilers are for scientists and engineers using computing systems ranging from workstations to the fastest GPU-powered supercomputers.
Use PGI Fortran, C and C++ compilers to develop performance-portable parallel applications for multicore x86-64 or OpenPOWER CPUs, and GPUs from NVIDIA. PGI compilers support full Fortran 2003, full C++14 and selected C++17 features. With PGI you can parallelize programs for multicore CPUs and NVIDIA GPUs using OpenACC 2.6, for multicore CPUs using OpenMP 4.5, and for NVIDIA GPUs using CUDA Fortran.
New Features in 2018:
- Support for Intel Skylake, IBM POWER9 and AMD Zen
- AVX-512 code generation on compatible Intel processors
- Full OpenACC 2.6 directives-based parallel programming on both Tesla GPUs and multicore CPUs
- OpenMP 4.5 for x86-64 and OpenPOWER multicore CPUs
- Integrated CUDA 9.1 toolkit and libraries for Tesla GPUs including V100 Volta
- Partial C++17 support and GCC 7.2 interoperability
- New PGI fastmath intrinsics library including AVX-512 support
