ACADEMIA

The Portland Group announced that PGI Accelerator Fortran, C and C++ compilers will soon target the AMD line of accelerated processing units (APUs) as well as the AMD line of discrete GPU (Graphics Processing Unit) accelerators. PGI will work closely with AMD to extend its PGI Accelerator directive-based compilers to generate code directly for AMD GPU accelerators, and to generate heterogeneous x64+GPU executable files that automatically use both the CPU and GPU compute capabilities of AMD APUs.

Introduced in 2009, PGI Accelerator compilers were the industry’s first standard-compliant Fortran and C compilers to automatically offload computations from an x64 host program to a GPU accelerator. The PGI Accelerator programming model enables parallel programmers to offload code from a host CPU to an attached accelerator by using hints, known as directives, to identify areas of code suitable for acceleration. In addition to exposing parallelization opportunities to the compiler, directives can also be used to specify the details of how to efficiently map loops to a particular accelerator and how to optimize data movement. Directives free the developer to focus on algorithms and application functionality while the compiler does the detailed work of offloading computations onto an accelerator. The principle benefit of using compiler directives is significant improvements to application performance through incremental and portable modifications to existing source code.

“The PGI Accelerator compilers will open up programming of AMD APUs and GPUs to the growing number of HPC developers using directives to accelerate science and engineering applications,” said Douglas Miles, director, The Portland Group. “Together with AMD, we are working to make heterogeneous programming easily accessible to mainstream C and Fortran developers, and to unleash the power of these devices.”

“We look forward to working with PGI to ensure that through the use of standard compiler directives the full computational power of AMD platforms with integrated APUs can be easily tapped,” said Terri Hall, Corporate VP, Business Alliances, AMD. “Engagements like this are key to expanding the developer ecosystem and the opportunities for AMD platforms.”

In June 2012, PGI announced support throughout the PGI Accelerator product line for the OpenACC Application Programming Interface (API) version 1.0, a specification for directive-based accelerator programming developed jointly by several HPC industry members including PGI.

In addition to their GPU capabilities, PGI Accelerator compilers include PGI’s complete suite of x86 host-performance optimization technologies including OpenMP extensions, automatic SIMD vectorization, auto-parallelization, inter-procedural analysis, function inlining, memory hierarchy optimizations and more.

Price and Availability

Preliminary support for AMD accelerators from within PGI Accelerator compilers is planned for mid-2013, and a full production release is scheduled for later that year. PGI will be demonstrating a development release of its PGI Accelerator compilers for AMD in booth 1321 at SC12 in Salt Lake City, November 12–15, 2012. Support for AMD accelerators will be included free of charge to PGI Accelerator licensees with a current PGI subscription.