• Site Map
• Contact
• Log In
• Search

On this page:

• Webinars
• Videos
• Articles
• PGInsider newsletters
• White papers and specifications
• Technical papers and presentations
• Laboratories and Code Samples

Webinars

• PGI Accelerator For C - Simplified GPU Programming Using Directives
• PGI Accelerator For Fortran - Simplified GPU Programming Using Directives, By Michael Wolfe, Portland Group
• GPU Programming with CUDA Fortran webinar
  by Michael Wolfe

  Sponsored by NVIDIA on 24 March 2010. Webinar slide deck is also available in PDF format.

• PGI x64+GPU Fortran & C99 Compilers IEEE webinar
  by Michael Wolfe

  Hosted by the IEEE Spectrum on 16 June 2009. Webinar slide deck is also available in PDF format.

Videos

• PGI Accelerator Compiler Overview (time: 3:20)
• PGI Accelerator Compiler Examples (time: 10:00)
• OpenMP 3.0 and Task Parallelization (time: 4:18)
• PGPROF 8.0 New Features (time: 7:50)
• PGDBG Parallel Debugger Demonstration. (time: 7:00)
• PGI Visual Fortran and Windows HPC Server 2008 (time: 10:41)

Articles

• "Compilers and More" series
  by Michael Wolfe

  Published by HPCwire

  • The Heterogeneous Programming Jungle, 19 March 2012
  • Arm Yourself for Exascale, 9 November 2011
  • Exascale Programming Requirements, 14 April 2011
  • Expose, Express, Exploit, 28 March 2011
  • Programming at Exascale, 9 March 2011
  • Hardware/Software Co-design, 2 November 2010
  • Knights Ferry Versus Fermi, 5 August 2010
  • A Computing Larrabee, 20 October 2009
  • OpenCL Promises and Potential, 10 September 2009
  • A GPU Accelerator Programming Model, 20 November 2008
  • Optimizing GPU Kernels, 30 October 2008
  • Programming GPUs Today, 9 October 2008
  • GPU Architecture & Applications, 10 September 2008
  • Parallel Programming Made Easy?, 2 September 2008
  • Accelerating High Performance, 7 May 2008
  • The Dangers of COTS Supercomputing, 4 April 2008
  • Gloptimizations, 8 November 2007
  • Are Optimizing Compilers Important?, 19 October 2007
  • Productivity and Compilers, 27 July 2007
  • Precision and Accuracy, 11 May 2007
  • What To Do With All Those Cores, 6 April 2007
  • Industrial Strength Interprocedural Analysis, 16 March 2007
    
  • AMD versus Intel: The Compiler as Referee
    interview with Michael Wolfe, 15 November 2006
  • Compilers/Tools for HPC
    Excerpt of CASES 2005 keynote presentation, 7 October 2005
  
  
• How We Should Program GPGPU's
  by Michael Wolfe

  Published by Linux Journal magazine November 2008

PGInsider Newsletters

• March 2013
  • PGI C++ with OpenACC
  • CUDA 5 Features in PGI CUDA Fortran 2013
  • Using CULA with PGI Fortran
  • Writing Efficient OpenCL Code for Android Using PGCL
• August 2012
  • OpenACC Kernels and Parallel Constructs
  • Thread- and Instruction-level Parallelism in CUDA Fortran
  • Creating an OpenCL-enabled Android App with PGCL
  • Introduction to Texture Memory Support in CUDA Fortran
• March 2012
  • The PGI Accelerator Compilers with OpenACC
  • The New PGC++ Compiler for Linux
  • 5x in 5 Hours: Porting a 3D Elastic Wave Simulator to GPUs Using PGI Accelerator
  • Using PGDBG to Debug Programs on Remote Systems
• October 2011
  • Multi-GPU Programming Using CUDA Fortran, MPI, and GPUDirect—Part 1
  • Accelerating Scientific Computing Code in FORTRAN: The Quantum Chemistry Project
  • Debugging PGI CUDA Fortran with Allinea DDT
  • A Look at the Impact of AMD Bulldozer FMA4 Instructions on Data Precision
• June 2011
  • Targeting AVX-Enabled Processors Using PGI Compilers and Tools
  • First Look: PGI CUDA C/C++ for x86
  • Object Oriented Programming with Fortran 2003—Part 2
  • An Introduction to Debugging CUDA-x86 Applications
• February 2011
  • Optimizing Data Movement in the PGI Accelerator Programming Model
  • Object Oriented Programming with Fortran 2003—Part 1
  • CUDA Fortran Asynchronous Data Transfers
  • Using GPU-enabled Math Libraries with PGI Fortran
• November 2010
  • PGI CUDA-x86: CUDA Programming for Multi-core CPUs
  • Porting Computational Chemistry Algorithms to GPUs Using PGI Accelerator Directives
  • Tuning Application Performance Using Hardware Event Counters in the PGPROF Profiler
  • Using MAGMA with PGI Fortran
• September 2010
  • CUDA Fortran: The Next Level
  • Porting the SPEC Benchmark BWAVES to GPUs with CUDA Fortran
  • Accelerating Visual C++ Applications on GPUs
  • Using the CULA GPU-enabled LAPACK Library with CUDA Fortran
• May 2010
  • PGI Accelerator Programming Model Support on NVIDIA Fermi GPUs
  • CUDA Fortran Device Kernels
  • Using NVIDIA GPU Accelerators with PGI Visual Fortran
  • Parallel Random Number Generation Using OpenMP, OpenCL and PGI Accelerator Directives
  • OpenMP on Accelerators—A Position Paper
• February 2010
  • The PGI Accelerator Programming Model on NVIDIA GPUs—Part 4: New Features
  • Using PGPROF and the CUDA Visual Profiler to Profile GPU Applications
  • CUDA Fortran Data Management
  • Tuning a Monte Carlo Algorithm on GPUs
  • Understanding the CUDA Data Parallel Threading Model—A Primer
• October 2009
  • The PGI Accelerator Programming Model on NVIDIA GPUs—Part 3 Porting WRF
  • Introduction to CUDA Fortran
  • Using Microsoft MPI with PGI Workstation
  • Porting WRF to Microsoft Windows Using PGI Workstation
• August 2009
  • The PGI Accelerator Programming Model on NVIDIA GPUs—Part 2 Performance Tuning
  • A New Direction for PGI Performance Profiling
  • Using Microsoft MPI with PGI Visual Fortran
  • Building Cactus BenchADM with PGI Accelerator Compilers
  • Speed-Up for Physicists
• June 2009
  • The PGI Accelerator Programming Model on NVIDIA GPUs—Part 1
  • Accessing Compiler Performance Advice
  • Developing Local MPI Programs
  • An Accelerator Programming Model for Multicore

White Papers and Specifications

• PGI Fortran & C Accelerator Programming Model new features (ver. 1.3 Nov. 2010, 690KB)

  This document describes the next generation features and capabilities planned for the PGI Accelerator programming model.

• PGI Fortran & C Accelerator Programming Model current features (ver. 1.2 Mar. 2010, 133KB)

  This document describes the currently supported features and limitations of the PGI Accelerator programming model.

• PGI Fortran & C Accelerator Programming Model full specification (ver. 1.0 Jun. 2009, 188KB)

  This document describes a collection of compiler directives used to specify regions of code in Fortran and C programs that can be offloaded from a host CPU to an attached accelerator. The method outlined provides a model for accelerator programming that is portable across operating systems and various types of host CPUs and accelerators. The directives extend the ISO/ANSI standard C and Fortran base languages in a way that allows a programmer to migrate applications incrementally to accelerator targets using standards-compliant Fortran or C.

• CUDA Fortran Programming Guide and Reference (ver. 1.3, May 2010, 136KB)

  NVIDIA CUDA™ a general purpose parallel programming architecture with compilers and libraries to support the programming of NVIDIA GPUs. This document describes CUDA Fortran, a small set of extensions to Fortran that supports and is built upon the CUDA computing architecture.

• Common Compiler Feedback Format (CCFF) Draft Specification

  CCFF is the Common Compiler Feedback Format, initially defined and implemented by PGI. PGI compilers add CCFF information to object and executable files that can be extracted into a file or read directly from the section of the object or executable file. The CCFF information is stored as an XML file, whose structure we describe here. See also CCFF XML schema and CCFF Repository XML schema

• Migrating HPC Applications from UNIX to Windows

  A review of options, issues and techniques for porting to Windows with particular attention on porting HPC programs to Windows HPC Server 2008. Includes review of different Windows development environment options. Discussion and examples for addressing migration issues including inter-language calling between Visual C++ and Fortran. Separate in-depth case studies look at using different build environments, porting MPI programs to MSMPI and porting shared objects. (Version 2.0, published September 2008.)

• Tools for the Classic HPC Developer

  Fortran was one of the first high-level computer programming languages, and has been in use for over 50 years. During that time, a huge reservoir of HPC applications has been developed in Fortran. The Fortran language has evolved to support both performance-oriented programming and modern best practices features in programming languages and parallel computing, and many HPC applications have incorporated components in C and C++ as well. As a result, there is an incredible range and diversity of HPC applications in use today, most of which were developed on UNIX-heritage HPC and cluster platforms. This paper presents the compiler and tools resources available and shows practical examples of porting such codes for use on Windows HPC Server 2008 clusters. (Version 2.0, published September 2008.)

Technical Papers and Presentations

• Michael Wolfe's SC12 GPU Programming with OpenACC Tutorials (Nov-2012)

  – Introduction slide deck 783KB PDF
  – Advanced slide deck 602KB PDF
  – Tutorial examples 17KB TAR

• Michael Wolfe's SC11 Advanced GPU Programming Tutorial (Nov-2011)

  – Slide deck 967KB PDF
  – Tutorial examples 555KB TAR

• Michael Wolfe's GPU Programming Tutorial (updated Mar-2011)

  Course materials from Michael Wolfe's day long tutorial on programming GPUs using the directive-based PGI Accelerator compilers and CUDA Fortran.

  1. Introduction 341KB PDF
  2. CUDA Fortran 905KB PDF
  3. PGI Accelerator Compilers 909KB PDF
  4. Conclusion 112KB PDF

  – Tutorial examples and labs 25.5MB TAR

• SC|08 PGI Vendor Presentation: PGI Compilers, GPUs and You!
  by Michael Wolfe

  Presented at SC|08 in Austin on 18 November 2008.

• Optimizing Application Performance on x64 Processor-based Systems with PGI Compilers and Tools
  by Douglas Miles, Brent Leback and David Norton

  PGI Fortran, C and C++ compilers and tools are supported on most x64 processor-based systems. Optimizing performance of the x64 processors in these systems often depends on maximizing SSE vectorization, ensuring alignment of vectors, and minimizing the number of cycles the processors are stalled waiting on data from main memory. The PGI compilers support a number of directives and options that allow the programmer to control and guide optimizations including vectorization, parallelization, function inlining, memory prefetching, interprocedural optimization, and others. In this paper we provide detailed examples of the use of several of these features as a means for extracting maximum single-node performance from x64 processor-based systems using PGI compilers and tools.

• Tuning C++ Applications for the Latest Generation x64 Processors with PGI Compilers and Tools
  by Douglas Doerfler and David Hensinger
  Sandia National Laboratories
  Brent Leback and Douglas Miles
  The Portland Group (PGI)

  Tuning numerically intensive C++ applications for maximum performance can be a challenge. This paper illustrates the importance of SSE vectorization on modern processors, and uses the ALEGRA shock physics code as an example of how a C++ application can be re-structured to enable vectorization and other optimizations that lead to dramatic performance improvements.

• SC|07 PGI Vendor Presentation
  by Michael Wolfe

  Presented at SC|07 in Reno on 14 November 2007.

• AMD versus Intel: The Compiler as Referee
  by Michael Wolfe

  Presented at SC|06 in Tampa on 21 November 2006.

• How compilers and tools differ for embedded systems
  by Michael Wolfe

  Keynote address at CASES 2005 San Franciso, 25 September 2005

Laboratories and Code Samples

• Hands-on: CUDA Fortran examples from GTC'13
• Hands-on: OpenACC Optimization examples from GTC'13

Quick Links

• PGI Accelerator Files
• FAQ
• Benchmarks
• Trial Download
• Documentation
• Resellers
• User Forums