PGI News
The latest news on The Portland Group and PGI® products.
Latest News from ST Microelectronics
PGI CUDA Fortran Now Available from The Portland Group
Portland, Ore
Novebmer 17, 2009
NVIDIA CUDA architecture now supported directly in the HPC industry-leading PGI Fortran compiler
The Portland Group®, a wholly-owned subsidiary of STMicroelectronics, today announced the general availability of its CUDA Fortran compiler for x64 and x86 processor-based systems running Linux, Mac OS X and Windows. Developed in collaboration with NVIDIA Corporation (Nasdaq: NVDA), the inventor of the graphics processing unit (GPU), PGI Release 2010 includes the first Fortran compiler compatible with the NVIDIA line of CUDA-enabled GPUs. A compiler is a software tool that translates applications from the high-level programming languages in which they are written by software developers into a binary form a computer can execute.
With developers taking advantage of the hundreds of cores and the relatively low cost of NVIDIA GPUs, programming to take advantage of the CUDA C compiler has become a popular means for accelerating the solution of complex computing problems. The PGI CUDA Fortran compiler is expected to accelerate GPU adoption even further in the High-Performance Computing (HPC) industry, where many important applications are written in Fortran. HPC is the field of technical computing engaged in the modeling and simulation of complex processes, such as ocean modeling, weather forecasting, environmental modeling, seismic analysis, bioinformatics and other areas.
"The GPU is ideally suited for the computationally intensive applications that define the HPC industry," said Andy Keane, general manager, Tesla business at NVIDIA. "The industry has been vocal in expressing its need for a CUDA-compatible Fortran option. NVIDIA customers can now build native CUDA Fortran applications using the widely-used and long-proven PGI Fortran compiler."
The CUDA Fortran compiler is compatible with all NVIDIA GPUs that include Compute Capability 1.3 or higher, which includes most NVIDIA Quadro Professional Graphics solutions and all NVIDIA Tesla GPU Computing solutions. Developers are invited to download the PGI CUDA Fortran compiler from The Portland Group website at www.pgroup.com/support/downloads.php. A 15-day trial license is available at no charge. In an effort to simplify adoption, NVIDIA has granted PGI rights to redistribute the relevant components of the CUDA Software Development Kit (SDK) as part of the PGI CUDA Fortran installation package.
PGI® products are used widely in high-performance computing (HPC). PGI compilers are recognized in the HPC community for delivering world-class performance across a wide spectrum of applications and benchmarks, and they are referenced regularly as the industry standard for performance and reliability. More information on the CUDA Fortran compiler is available directly from The Portland Group web site at www.pgroup.com/cudafortran.
GPU computing forums for news, discussion and programming tips are available at forums.nvidia.com and www.pgroup.com/userforum.
All registered trademarks and marks are the property of their respective owners.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959
PGI Release 2010 Enables Scientists and Engineers
to Harness the Computational Power of GPUs
Portland, Ore
Novebmer 12, 2009
Latest compilers from The Portland Group extend support for the PGI Accelerator Programming Model and PGI CUDA Fortran
The Portland Group®, a wholly-owned subsidiary of STMicroelectronics, today announced that release 2010 of the PGI® line of high-performance parallelizing compilers and development tools for Linux, Mac OS X and Windows will be available on Tuesday, November 17. PGI 2010 is the first general release to include full support for the PGI Accelerator Programming model v1.0 standard on x64 processor-based systems incorporating NVIDIA CUDA-enabled Graphical Processing Units (GPUs). In addition to supporting high-level programming of accelerators using the PGI Accelerator programming model, the PGI Release 2010 also includes PGI CUDA Fortran, an explicit GPU programming model and application programming interface (API) that gives expert programmers direct control of all aspects of programming NVIDIA GPUs.
The PGI Accelerator programming model is 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 to enhance performance. Applications optimized using the PGI Accelerator directives remain 100% portable to other compilers and platforms, and execute on systems with or without a GPU accelerator.
PGI 2010 offers full support for the PGI Accelerator programming model including the following new features:
- GPU device-resident data—the ability to define and leave data on the GPU across accelerator regions and subroutine boundaries
- Support for COMPLEX and DOUBLE COMPLEX data types in Fortran
- Support for C structs and Fortran derived types
- Automatic GPU-side loop unrolling for improved performance
- Support for Accelerator regions nested within OpenMP parallel regions
- Support for Linux, Mac OS X (including Snow Leopard) and Windows (including Windows 7).
"Within five years, most HPC systems will include both x86 CPUs and accelerators in some form," said Douglas Miles, director, The Portland Group. "The PGI 2010 compilers will play a role in establishing accelerated computing as the mainstream HPC architecture."
PGI CUDA Fortran includes a Fortran 95/03 compiler and tool chain for native programming of NVIDIA GPUs using Fortran. CUDA Fortran subroutines can launch and execute in parallel on the hundreds of cores in an NVIDIA GPU under control of an x64 host CPU. Developed in collaboration with NVIDIA, PGI CUDA Fortran extensions supported in the PGI 2010 Fortran 95/03 compiler enable HPC developers to explicitly control all aspects of data movement, memory utilization and computation on CUDA GPUs.
Additional new features in the PGI 2010 compilers and tools include support for more Fortran 2003 incremental features, the latest EDG 4.1 C++ front-end with enhanced GNU and Microsoft compatibility, OpenMP parallel programming support for up to 256 cores, and AVX code generation. PGI 2010 also includes a major update to the PGPROF performance profiler, which now supports performance profiling of binary executables without re-compiling or any special software privileges, uniform operation and features on Linux, Mac OS X and Windows, support for PGI Accelerator and PGI CUDA Fortran GPU-side performance statistics, and an updated graphical user interface. Finally, PGI 2010 supports the latest operating system releases including Red Hat Fedora 10/11, SuSE 11.1 and Ubuntu 9, Mac OS X Snow Leopard and Windows 7.
Enhancements in the PGI 2010 release of PGI Visual Fortran for Microsoft Visual Studio include full support for the PGI Accelerator Programming model and PGI CUDA Fortran on NVIDIA CUDA-enabled GPUs, and the addition of a new standalone version of the PGPROF performance profiler for x64 and GPUs with support for the Common Compiler Feedback Format (CCFF). CCFF is a draft standard published by PGI that defines what compiler information is stored and how the information is formatted. CCFF enables HPC tools providers to offer more and better information about optimizing performance.
More information about the PGI Accelerator Programming model is available online at www.pgroup.com/accelerate. PGI CUDA Fortran information is available separately at www.pgroup.com/cudafortran. Evaluation copies of the new PGI 2010 compilers are available from The Portland Group web site at www.pgroup.com. Registration is required.
All registered trademarks and marks are the property of their respective owners.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959
PGI Visual Fortran 9.0 Adds Support for MPI Debugging on
Windows HPC Server 2008 Clusters
Portland, Ore
August 8, 2009
Latest version allows users to build, launch and debug from within Microsoft Visual Studio
The Portland Group®, a wholly-owned subsidiary of STMicroelectronics and leading supplier of compilers for high-performance computing (HPC), today announced the general availability of PGI Visual Fortran® (PVF®) Release 9.0 for Windows workstations, servers and clusters. PVF 9.0 is the first general release to include support for the building, launching and debugging of Microsoft MPI (MSMPI) Fortran applications from within the Microsoft Visual Studio integrated development environment.
PVF augments the Visual Studio debugger by adding a Fortran language specific custom debug engine. The PVF debug engine supports debugging of single and multi-thread, OpenMP, multi-thread MSMPI and hybrid MSMPI+OpenMP Fortran applications. It enables debugging of 64-bit or 32-bit applications using source code or assembly code, and provides full access to the registers and hardware state of the processors. Other new multi-process MSMPI capabilities in PVF 9.0 include Visual Studio property pages for configuring compile-time options, launching applications either locally on a workstation or on a distributed-memory Windows HPC Server 2008 cluster system, and debugging of programs running either locally or on a cluster.
"PVF 9.0 is a big step forward in ease-of-use for HPC Fortran programmers porting applications to or developing applications for Windows workstations, servers and clusters," said Douglas Miles, director, The Portland Group. "For Windows Fortran users looking to leverage the power of Windows HPC Server 2008 clusters, the ability to cover all aspects of MPI and parallel Fortran application development from within the Microsoft Visual Studio IDE can simplify their work considerably."
"The majority of HPC applications are still written in Fortran and parallelized using MPI and OpenMP," said Vince Mendillo, Director, Technical Computing Marketing, Microsoft. "By including MSMPI job launch and debugging support within PGI Visual Fortran, PGI has further enhanced the Windows HPC Server 2008 ecosystem and simplified porting of HPC applications to Windows clusters."
Additional new features in PVF 9.0 include support for Intel Core i7 (Nehalem) and six-core AMD Opteron (Istanbul) processors, several incremental Fortran 2003 features, improvements in serial debugging and disassembly speed, and completely updated documentation and online help. PGI Visual Fortran is compatible with both the current version of Visual Studio, Visual Studio 2008, and the previous version, Visual Studio 2005.
PGI Release 9.0 is the first general release to include support for the high-level PGI Accelerator™ programming model on x64 processor-based Linux systems incorporating NVIDIA CUDA-enabled GPUs. Announced last June, the PGI Release 9.0 line of high-performance parallelizing compilers and development tools for Linux, Mac OS X and Windows is now available.
Further information on The Portland Group products, including PGI® Release 9.0, can be found at www.pgroup.com, by calling Sales at (503) 682-2806, or by email to sales@pgroup.com.
All registered trademarks and marks are the property of their respective owners.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959
PGI and NVIDIA Team To Deliver CUDA Fortran Compiler
Portland, Ore
June 23, 2009
Fortran support for NVIDIA CUDA GPUs to be incorporated into a new version of the PGI Fortran compiler
The Portland Group®, a wholly-owned subsidiary of STMicroelectronics and leading supplier of compilers for high-performance computing (HPC), today announced an agreement with NVIDIA under which the two companies plan to develop new Fortran language support for CUDA GPUs.
The NVIDIA® CUDA™ architecture allows developers to offload computationally intensive kernels to the massively parallel GPU. Through function calls and language extensions, CUDA gives developers explicit control over the mapping of general-purpose computational kernels to GPUs as well as placement and movement of data between the x64 processor and the GPU. The NVIDIA CUDA C compiler already provides this capability to C programmers. The CUDA Fortran compiler will provide this same level of control and optimization in a native Fortran environment from PGI.
"Fortran support for CUDA GPUs is a perfect complement to our existing roadmap for the PGI Accelerator Fortran and C compilers," said Douglas Miles, director, The Portland Group. "It enables interoperability of PGI Fortran and CUDA C and gives PGI users a full range of options in porting and optimizing Fortran applications to leverage the power of CUDA-enabled NVIDIA GPUs."
"The GPU computing developer community has made it clear there is a need and demand for a production-quality Fortran solution on the GPU," said Andy Keane, general manager, Tesla GPU Computing Solutions, NVIDIA. "With their large base of Fortran developers for x64 processor-based HPC systems, PGI provides a perfect bridge for migration of production science and engineering codes from existing platforms to NVIDIA Tesla GPUs."
The Portland Group and NVIDIA will release the Fortran language specification for CUDA GPUs at the International Conference on Supercomputing in Hamburg, Germany this week. The CUDA Fortran compiler will be added to a production release of the PGI Fortran compilers scheduled for availability in November 2009. More detailed information about PGI compilers and tools is available online at www.pgroup.com.
For more information on NVIDIA CUDA, please visit www.nvidia.com/cuda.
All registered trademarks and marks are the property of their respective owners.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959
New PGI 9.0 Compilers Simplify x64+GPU Programming
Portland, Ore
June 23, 2009
The Portland Group shares roadmap for PGI Accelerator Fortran and C compiler products for GPU-enabled HPC systems
The Portland Group®, a wholly-owned subsidiary of STMicroelectronics and leading supplier of compilers for high-performance computing (HPC), today announced the general availability of the PGI® Release 9.0 line of high-performance parallelizing compilers and development tools for Linux, Mac OS X and Windows. PGI Release 9.0 is the first general release to include support for the high-level PGI Accelerator™ programming model on x64 processor-based Linux systems incorporating NVIDIA CUDA-enabled GPUs.
The use of Graphics Processing Units (GPUs) as general purpose accelerators has been a growing trend in high-performance computing (HPC). Until now, use of GPUs from Fortran applications has been extremely limited. Developers targeting GPU accelerators have had to program in C at a detailed level using sequences of function calls to manage movement of data between the x64 host and GPU, and to offload computations from the host to the GPU. The PGI Accelerator Fortran and C compilers automatically analyze whole program structure and data, split portions of an application between a multi-core x64 CPU and a GPU as specified by user directives, and define and generate a mapping of loops to automatically use the parallel cores, hardware threading capabilities and SIMD vector capabilities of modern GPUs.
"PGI 9.0 opens up general-purpose programming on NVIDIA CUDA-enabled GPUs to science and engineering domain experts," said Douglas Miles, director, The Portland Group. "Programming models like CUDA and OpenCL enable expert programmers to write efficient GPU code by exposing low-level details of the hardware. The PGI Accelerator model enables any Fortran or C programmer to incrementally add GPU optimizations to a program using compiler directives similar to those used in OpenMP programming for multi-core CPUs."
The PGI Accelerator programming model is available now in the PGI Fortran and C compilers on Linux as an extended free preview for all PGI 9.0 licensees through the end of 2009. Additional PGI 9.0 new features include several Fortran 2003 incremental features, Intel Xeon EX (Nehalem) optimizations including support for SSE 4.1/4.2, Six-Core AMD Opteron (Istanbul) support and optimizations, full OpenMP 3.0 support in C++, an all-new graphical interface on the PGDBG OpenMP/MPI debugger, enhancements to the PGPROF performance profiling environment, and support for Fedora Core 10/11, SuSE 11.1 and Ubuntu 9.
PGI Accelerator compilers product line announced The Portland Group has also unveiled plans for a new x64+GPU compiler and tools product line that will be first available in a release scheduled for Q4 2009. The PGI Workstation Accelerator product line features currently planned include:
- PGI Accelerator Fortran and C compilers on Linux, MacOS and Windows
- Support for GPU device-resident data in the PGI Accelerator programming model
- Profiling of x64+GPU applications using the PGPROF® profiler
- Expanded compiler feedback (CCFF) generation to guide application porting and optimization for x64+GPU targets
An introduction to PGI Accelerator Fortran and C programming is available online at www.pgroup.com/lit/articles/insider/v1n1a1.htm. The PGI Accelerator v1.0 specification is online at www.pgroup.com/lit/whitepapers/pgi_accelerator.pdf. Evaluation copies of the new PGI 9.0 compilers are available from The Portland Group web site at www.pgroup.com. Registration is required.
All registered trademarks and marks are the property of their respective owners.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959
The Portland Group and AMD Ink Agreement to Collaborate on
Accelerator Compiler Technology
Austin, TX
November 18, 2008
Cooperation based around PGI's x64+GPU compiler technology and AMD FireStream™ compute accelerators with an eye toward AMD Fusion
The Portland Group®, a wholly-owned subsidiary of STMicroelectronics and leading supplier of compilers for high-performance computing (HPC), today announced it has entered into an agreement with AMD to cooperate on the development of compiler technology for AMD FireStream™ compute accelerators. As part of the agreement, PGI and AMD will investigate and develop technology to enable PGI® Fortran and C compilers to generate code directly for AMD FireStream boards and to generate heterogeneous x64+GPU (Graphics Processing Unit) executables that enable automatic utilization of AMD FireStream products if present in a system.
The Portland Group recently delivered a technology preview of the industry's first standard Fortran and C compilers that automatically offload computations from an x64 host program to a GPU. Until now, C and C++ developers targeting GPU accelerators have had to rely on language extensions or library-based solutions, and use of GPUs from Fortran applications has been extremely limited. The PGI x64+GPU programming model enables programmers to accelerate applications simply by adding portable compiler directives, treated as comments by other compilers, to existing standard-compliant Fortran and C programs and recompiling with appropriate compiler options. This gives individual developers the power to incrementally use accelerator technology where appropriate in large existing applications, eliminating the need for expensive and difficult corporate-level platform migration or acquisition decisions.
"Until now, industry CIO's and government and university lab directors have encountered difficult decisions when evaluating accelerator technologies," said Douglas Miles, director, The Portland Group. "Which accelerators do I choose? How do I retrain developers to use accelerators? How much will I have to invest in new software technologies? Working together with AMD, we will work to make FireStream compute accelerators easily accessible to mainstream C and Fortran developers and unlock the power of these devices using PGI compilers that are already in use at most of the significant HPC centers and sites."
"We applaud The Portland Group for giving developers advanced, yet simplified tools for fast, easy creation of GPU accelerated applications," said Patricia Harrell, director of Stream Computing at AMD. "The Portland Group clearly understands the needs of today's HPC developers, and their efforts are instrumental in driving GPU compute acceleration into the mainstream. We look forward to working with PGI to help ensure their Fortran and C compilers can take full advantage of the over 1.2 teraFLOPS of single-precision, and over 240 gigaFLOPS of double-precision performance capabilities of the AMD FireStream 9270 compute accelerators."
PGI develops and markets high-performance C/C++ and Fortran compilers and development tools that are widely used by engineers and scientists in high-performance computing (HPC), the field of technical computing engaged in the modeling and simulation of complex processes, such as ocean modeling, weather forecasting, seismic analysis, bioinformatics and other areas. PGI compilers and tools are designed to extract maximum performance from the latest multi-core x64 processors, including the recently announced 45 nanometer Quad-Core AMD Opteron™ processors from AMD. PGI compilers are recognized in the HPC community for delivering world-class performance across a wide spectrum of applications and benchmarks, and they are referenced regularly as the industry standard for performance and reliability.
Key advantages of PGI compilers and tools can be found on The Portland Group web site at www.pgroup.com/about/why_PGI.htm . A free 15-day evaluation of PGI Workstation compilers and tools is available for download at www.pgroup.com/support/downloads.php. Registration is required.
All registered trademarks and marks are the property of their respective owners.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959
The Portland Group Announces PGI 8.0 Optimizing Compilers & Tools
for Multi-core x64 Processors
Portland, OR
November 17, 2008
Latest PGI release supports OpenMP 3.0 on Linux, Windows and MacOS; includes heterogeneous x64+GPU compiler technology preview
The Portland Group®, a wholly-owned subsidiary of STMicroelectronics, today announced the general availability of the PGI® Release 8.0 line of high-performance compilers and development tools for Linux, Mac OS X and Windows. PGI Release 8.0 includes full support for the recently announced OpenMP 3.0 multi-core parallel programming standard in Fortran and C across all supported platforms. The new release also adds support for building and debugging of OpenMPI applications on both Linux and MacOS, complementing existing MPI capabilities on Linux and Windows clusters. PGI 8.0 users can now develop and deploy multi-core and parallel applications on any of the major desktop or cluster operating systems using identical PGI compilers, the latest OpenMP features, MPI implementation of choice and bundled OpenMP/MPI-capable debugging and profiling tools. In a significant new development, PGI Release 8.0 also marks The Portland Group's entry into the field of accelerated computing with provisional support for automatic offloading of parallel computations from x64 host processors to CUDA-enabled GPUs from NVIDIA.
"Together with PGI Unified Binary technology, which enables developers to leverage the latest CPU innovations from both AMD and Intel while treating x64 processors as a single platform, the new features in PGI 8.0 maximize flexibility and independence for HPC users and large multi-platform supercomputing centers," said Douglas Miles, director, The Portland Group.
New Performance Analysis Tools
In addition to building on a compiler and tools product line that now includes all best practices HPC and multi-core programming technologies, the PGI 8.0 compilers include an all-new capability to automatically analyze source code, produce an extensive database describing performance optimizations that are possible or inhibited, and provide advice for modifying the source code to take advantage of the possible optimizations. With Release 8.0, PGI has standardized the organization and interface to this data through the Common Compiler Feedback Format (CCFF). PGI is publishing the CCFF specification and making access to it freely available in an effort to improve the utility and interoperability of PGI, third-party and research-community software tuning tools. More information on CCFF is available from the PGI website at http://www.pgroup.com/ccff.
PGI's PGPROF® 8.0 performance profiler displays CCFF data coupled with user source code in a logical, compact and intuitive graphical user interface (GUI). A command-line interface is also supported. Programmers can quickly and easily identify code segments that are already well-structured, as well as those that can be restructured to improve performance. In addition to identifying sections of an application that consume most of the compute time or system resources, PGPROF provides developers with specific actionable performance optimization feedback about their source code. The data, presented on a per-thread and/or per-process basis, simplifies performance tuning by identifying:
- Streaming SIMD Extensions (SSE) vector loops, and why vectorization is inhibited on non-vector loops
- Loops auto-parallelized for multi-core, and why parallelization is inhibited on serial loops
- Loops that are candidates for OpenMP parallelization
- Compute intensity of loops, and candidates for offloading to a GPU (Graphics Processing Unit) or accelerator
- Loops with very small or very large iteration counts, and how they can be modified to maximize performance for SSE and the cache-based memory hierarch
- Data prefetching, and opportunities for prefetch tuning using directives and pragmas
In addition to these detailed analyses, PGPROF also includes overall program level analyses including information about in-lined functions and subroutines and information about how each file was compiled, comprehensive system configuration information and many other performance-critical characteristics of Fortran, C and C++ source code. Unlike traditional performance tuning tools which only report on and help tune performance for a specific type of processor or system, or focus solely on parallelization, the PGI 8.0 compilers and tools provide developers with feedback and insight on how to restructure loops and algorithms to enhance performance on any modern multi-core x64 CPU or GPU accelerator.
"Parallelism does not equate to performance," said Michael Wolfe, compiler engineer, The Portland Group. "The focus needs to be not on parallelism, but on performance, where parallelism is one of the tools to get it."
Provisional GPU Support
PGI Release 8.0 also includes a technology preview of the industry's first Fortran and C compilers that automatically offload computations from an x64 host program to a GPU. Until now, C and C++ developers targeting GPU accelerators have had to rely on language extensions to their programs. Use of GPUs from Fortran applications has been extremely limited. x64+GPU programmers have been required to program at a detailed level including a need to understand and specify data usage information and manually construct sequences of calls to manage all movement of data between the x64 host and GPU. Using the provisional support in PGI Release 8.0, programmers can accelerate Linux applications on x64+GPU platforms by adding OpenMP-like compiler directives to existing high-level standard-compliant Fortran and C programs and then recompiling with appropriate compiler options.
"PGI is joining the increasing number of software publishers offering innovative approaches to harnessing the power of NVIDIA GPUs by leveraging the CUDA development environment," said Andy Keane, general manager, Tesla computing solutions, NVIDIA. "With their 20 year history and track record of success, we expect PGI's offering will open the door for members of the HPC community to begin incrementally porting large legacy production science and engineering codes to take full advantage of NVIDIA Tesla accelerators."
The PGI 8.0 x64+GPU compilers automatically analyze whole program structure and data, split portions of the application between the x64 CPU and GPU as specified by user directives, and define and generate an optimized mapping of loops to automatically use the parallel cores, hardware threading capabilities and SIMD vector capabilities of modern GPUs. In addition to directives and pragmas that specify regions of code or functions to be accelerated, the PGI Fortran and C compilers support user directives that give the programmer fine-grained control over the mapping of loops, allocation of memory, and optimization for the GPU memory hierarchy. The PGI compilers generate unified x64+GPU object files and executables that manage all movement of data to/from the GPU device while leveraging all existing host-side utilities—linker, librarians, makefiles—and require no changes to the existing standard HPC Linux/x64 programming environment.
Other significant new features included in PGI Release 8.0 are support for OpenMP parallel and local OpenMPI parallel debugging in Mac OS X, new simplified licensing setup on Microsoft Windows, support for Microsoft HPC Server 2008 clusters and support for the latest processors from AMD and Intel.
Evaluation copies of the new PGI compilers are available from The Portland Group web site at www.pgroup.com. Registration is required.
All registered trademarks and marks are the property of their respective owners.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959
The Portland Group Adds IMSL Fortran Numerical Library for Windows
Portland, OR
March 25, 2008
Popular Visual Numerics library now available as an option with PGI multi-core optimizing parallel Fortran compilers
The Portland Group®,a wholly-owned subsidiary of STMicroelectronics (NYSE: STM) and a leading supplier of compilers and development tools for High-Performance Computing, today announced that the IMSL® Fortran Numerical Library from Visual Numerics®, Inc. is available from The Portland Group for PGI's line of multi-core optimizing parallel Fortran compilers running under the Microsoft Windows operating systems.
The IMSL Fortran Library is a comprehensive set of popular mathematical and statistical functions. Software developers can use the pre-written functions included within the library as building blocks in their software applications. This library frees programmers from the necessity of developing their own equivalent functions. Both the library and numerical algorithms within it have a reputation for quality and performance across an extensive range of compilers and operating system environments.
"The IMSL Fortran Library is one of the most widely used math libraries for technical computing applications on Windows and is very popular with our Linux customers," said Douglas Miles, director, The Portland Group. "We've been working with Visual Numerics to bring this valuable resource to PGI's complete line of Windows compilers including support on native Windows, the Windows Subsystem for Unix Applications (SUA), and within applications built using PGI Visual Fortran. "
"We are pleased to be working closely with The Portland Group to bring the IMSL Fortran Library to their latest generation of Windows Fortran compilers," said Tim Leite, Director of Corporate Development at Visual Numerics. "PGI compilers are popular in technical and high-performance computing, a key target market segment for the IMSL Libraries. With this new product we extend to PGI and HPC developers on Windows the same high standards for quality and reliability that IMSL Libraries are known for worldwide. "
PGI compilers and tools are used predominantly on 64-bit and 32-bit Linux, Mac OS X and Microsoft Windows workstations, servers, and clusters based on microprocessors from AMD and Intel. The Portland Group's flagship product suite, PGI Workstation, includes compilers and tools for building, debugging and profiling parallel 64-bit and 32-bit Fortran, C and C++ applications within a UNIX- and Linux- compatible development environment. Key compiler features include automatic and user-directed parallelization and optimizations for extracting maximum performance from the new generation of multi-core processors. In addition to extensive UNIX migration features, the PGI Visual Fortran compiler is available fully integrated with Microsoft's Visual Studio development environment. The key advantages of PGI compilers and tools can be found on The Portland Group web site at www.pgroup.com/about/why_pgi.htm. A trial version of PGI compilers is available for download at www.pgroup.com/support/downloads.php . Registration is required.
The IMSL Numerical Libraries have been the cornerstone of high- performance and deep computing as well as predictive analytics applications in science, technical and business environments for well over three decades. These embeddable mathematical and statistical algorithms, written in C, C#, Java™, and Fortran, are used in a broad range of applications. The IMSL Libraries are regarded as the most sophisticated, flexible, scalable and highly accessible technology available for numerical analysis in the most important mainstream programming environments in use today.
All registered trademarks and marks are the property of their respective owners.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959
PGI CDK Cluster Development Kit for Windows
Now Available From The Portland Group
Portland, OR
January 30, 2008
PGI brings powerful suite of parallel compilers and programming tools to popular Microsoft Windows Compute Cluster Server 2003
The Portland Group®, a wholly-owned subsidiary of STMicroelectronics and a leader in compilers and development tools for high-performance computing, today announced the general availability of its PGI CDK® Cluster Development Kit® for Microsoft Windows Computer Cluster Server 2003 (CCS). The PGI CDK for Windows is a suite of parallel Fortran, C, and C++ compilers and development tools specifically optimized for Windows CCS. PGI compilers and tools are used widely in high- performance computing (HPC), the field of technical computing engaged in the modeling and simulation of ocean modeling, weather forecasting, seismic analysis, crash simulation and other complex processes.
Until now, the PGI CDK has been available only for Linux-based clusters based on 32-bit and 64-bit microprocessors from AMD and Intel. The Portland Group extends to these same platforms support for the Microsoft 64-bit Compute Cluster Server operating systems with this release.
"With an accelerating trend towards platform diversification in the HPC market, PGI remains committed to providing HPC developers with the best performance on their choice of platform," said Douglas Miles, director, The Portland Group. "The HPC community has recognized PGI compilers for delivering world-class performance across a wide spectrum of applications and benchmarks running on workstations and servers based on both major x64 processors, and the introduction of PGI CDK for Windows CCS gives the HPC community another option for cluster programming."
"We are excited to partner with PGI to deliver both world-class performance and ease-of-use for HPC clusters," said Shawn Hansen, director of HPC marketing division at Microsoft. "Now that PGI's market leading tools are available cross-platform, customers can accelerate their productivity by combining their tuned developer environments with the ease-of-use of the Windows HPC platform."
The PGI CDK Cluster Development Kit includes the PGF77®, PGF95™, PGCC®, and PGC++® compilers for the FORTRAN 77, Fortran 95, C, and C++ programming languages, respectively. All PGI compilers feature full native support for OpenMP parallel programming extensions in Fortran, C, and C++; full support for 64-bit addressing; native integrated scalar and vector SSE code generation; and a bundled version of the ACML, a highly optimized library of numeric functions for mathematical, engineering, scientific, and financial applications.
In addition, PGI CDK includes PBDBG®, an OpenMP and MPI parallel symbolic debugger and PGPROF®, an OpenMP and MPI parallel graphical performance profiler. Both tools include support for MSMPI, the Windows implementation of the de facto standard Message Passing Interface commonly used for running programs in parallel on clusters. MSMPI is included with Windows CCS. PGDBG and PGPROF provide full-featured graphical user interfaces that are fully compatible with the Linux versions, creating an environment that is ideal for programmers porting applications from Linux to Windows.
Included in the PGI CDK for Windows are network floating-license versions of PGI's compilers and tools designed to run under a number of different Windows programming environments including:
- Native, under 64-bit or 32-bit Windows CCS, using a UNIX-like shell command interface that is compatible with UNIX/Linux environments, the traditional operating system of choice in the HPC field.
- Windows Subsystem for UNIX-based applications (SUA). SUA is a source-code compatible subsystem for compiling and running UNIX-based applications on a computer running a Windows Server-class operating system including Windows Compute Cluster Server 2003.
- Visual Studio 2005. Visual Studio is the preeminent graphical software development environment for Windows.
Complete information on the PGI CDK for Windows is available from The Portland Group web site at http://www.pgroup.com. All PGI products are highly optimized for both AMD64 and Intel 64 processors.
All registered trademarks and marks are the property of their respective owners.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959
PGI Compilers and Tools from The Portland Group
Portland, OR
October 29, 2007
64-bit high-performance parallel Fortran, C and C++ Tools support Mac OS X Leopard
The Portland Group®, a wholly-owned subsidiary of STMicroelectronics (NYSE: STM) and leading supplier of compilers for high-performance computing (HPC), today announced the general availability of its PGI(r) Workstation suite of 64-bit high-performance parallel compilers and tools for Intel processor-based Macs running Mac OS X, including the new fully 64-bit enabled Mac OS X Leopard.
PGI develops and markets high-performance C/C++ and Fortran compilers and development tools that are popular and widely used by engineers and scientists in high-performance technical computing (HPTC). Technical computing involves the modeling and simulation of complex processes, such as ocean modeling, weather forecasting, seismic analysis, bioinformatics and other areas. PGI compilers are recognized in the HPC community for delivering world-class performance across a wide spectrum of applications and benchmarks, and they are referenced regularly as the industry standard for performance and reliability. PGI compilers and tools are designed to extract maximum performance from the latest multi-core processors from Intel and AMD.
"PGI is committed to supporting all popular high-performance computing platforms. We see the Intel processor-based Macs gaining momentum among our traditional customer base of scientists and engineers," said Douglas Miles, director, The Portland Group. "By adding Mac OS X support alongside our existing support for Linux and Windows, HPC developers can now target all the primary platforms using a single, uniform software development environment."
According to Net Applications operating system market share data for computers using the Internet (http://marketshare.hitslink.com), for the year ending 31 August 2007, Mac OS X market share was up over 42%, now comprising 6.2% of the market. Shares for the same period for Microsoft Windows Vista and Linux were 6.3% and 0.8% respectively. Windows XP remains the operating system of choice for most Internet users with over an 80% market share. The Mac's gains are largely the result of new users adopting the Intel processor-based systems as the market share for the older PowerPC processor-based platform fell by less than 0.4% over the period.
"Leopard delivers full native 64-bit support to enable applications to take complete advantage of 64-bit processing while still running side by side with existing 32 bit applications and drivers," said Ron Okamoto, Apple's vice president of Worldwide Developer Relations. "With a growing base of HPC tools like PGI Workstation now available, more scientists and engineers are choosing the Mac platform."
Today, PGI compilers and tools are used predominantly on 64-bit and 32-bit Linux and Microsoft Windows workstations, servers, and clusters based on microprocessors from AMD and Intel. The Portland Group's flagship product suite, PGI Workstation, includes compiler and tools for building, debugging and profiling 64-bit and 32-bit Fortran, C and C++ applications. Key compiler features include automatic and user-directed parallelization and optimizations for extracting maximum performance from the new generation of multi-core processors. The key advantages of PGI compilers and tools can be found on The Portland Group web site at www.pgroup.com/about/why_pgi.htm. The free trial version PGI Workstation for Mac OS X is available for download at www.pgroup.com/support/downloads.php. Trial versions of PGI compilers for Linux and Windows are available at the same address. Registration is required.
All registered trademarks and marks are the property of their respective owners.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959
The Portland Group Announces PGI Release 7.1 of its
Optimizing Compilers for Multi-core 64-bit Processors
Portland, OR
October 25, 2007
Targeting CPUs with four or more cores, latest PGI release brings MPI debugging and profiling to the Linux desktop and delivers record-setting x64 benchmark performance on new Quad-Core AMD Opteron Processors
The Portland Group®, a wholly-owned subsidiary of STMicroelectronics (NYSE: STM), today announced the general availability of Release 7.1 of its suite of Fortran, C and C++ compilers and development tools. PGI® compilers and tools are used widely in high-performance computing (HPC), the field of technical computing engaged in the modeling and simulation of complex phenomena, such as ocean modeling, weather forecasting, seismic analysis, bioinformatics and other areas. PGI compilers, which convert software programs into the binary instructions that a computer can understand, are recognized in the HPC community for delivering world-class performance across a wide spectrum of applications and benchmarks, and they are referenced regularly as the industry standard for performance and reliability.
The latest PGI 7.1 compilers and tools deliver all existing best-practices software components for developing highly efficient parallel applications to multi-core Linux desktops. For the past 10 years, applications for parallel supercomputers have been developed almost exclusively using the de facto standard Message Passing Interface (MPI) programming model or OpenMP directive-based extensions to Fortran, C and C++. HPC developers now face the daunting task of adapting these MPI and OpenMP applications to run effectively on clusters of multi-core CPUs, while general-purpose developers face the substantial challenge of adapting serial legacy applications to effectively use multiple cores. The coming wave of multi-core processors will require developers to leverage the existing knowledge base of software development from HPC.
PGI 7.1 delivers comprehensive support for both OpenMP and MPI, including graphical OpenMP and MPI debugging and profiling tools, in all PGI product configurations for Linux. This includes PGI's lowest-cost configuration: PGI C/C++ Workstation for a single academic user, which is priced at $299 USD for a permanent license. Until now, such a comprehensive parallel-programming environment was available only on dedicated HPC or cluster systems at a cost of thousands of dollars per seat. PGI's latest offering makes all of these capabilities plus state-of the-art automatic parallelization available and affordable to virtually any developer with a need to adapt applications to run in parallel on multi-core processors.
"We continue to invest heavily in delivering performance-oriented compilers and tools to our core science and engineering users while making a concerted effort to enable parallel application development by more general-purpose developers," said Douglas Miles, director, The Portland Group. "With Release 7.1, we are delivering our fastest Fortran compiler yet, and also C and C++ compilers that show performance gains of more than 10% versus PGI 7.0 on standard benchmarks running on the latest multi-core processors. We are taking a practical approach to helping our customers cross the multi-core divide by delivering increasingly better compile-and-go performance as part of a complete parallel programming toolkit for systems ranging from dual- or quad-core laptops and desktops to the world's fastest supercomputers."
"HP and its customers value PGI's practical approach to optimizing software for multi-core processors," said Ed Turkel, Manager of Product Marketing for High Performance Computing at HP. "HP is delighted to have PGI participate in our Multi-core Optimization Program. We recognize the significant value that PGI's comprehensive set of processor-independent compilers and tools brings to our joint customers in tuning applications on HP's servers and clusters using the latest multi-core processors from both AMD and Intel."
The PGI 7.1 compilers have enabled AMD to publish SPECfp_rate_base2006 performance results on Quad-Core AMD Opteron 2.5 Ghz processors that are over 27% faster than the best published performance results on the same benchmark running on Intel Xeon X5365 Quad Core 3.0 Ghz processors using Intel 10.1 compilers.* "With Release 7.1, PGI has made a significant advance in bringing the performance of their C and C++ compilers up to the same high level that we have come to expect from their Fortran compilers," said Michael Goddard, senior director, AMD Performance Center of Excellence. "PGI and AMD have been cooperating throughout 2007 to ensure PGI compilers are highly tuned for the new micro-architecture and features of AMD's latest Quad-Core AMD Opteron processors."
PGI Release 7.1 also marks the introduction of two new products: optimizing PGI compilers and tools for Intel processor-based Apple Macintosh systems, and a compiler and tools suite for Windows that includes the first commercially available debugger to support debugging of MSMPI applications on Microsoft's flagship HPC product, Windows Compute Cluster Server (CCS). PGI 7.1 includes numerous under-the-hood compiler optimizations, and it features improved support for what already is the most comprehensive developer-tools offering for migration of UNIX applications to AMD and Intel processor-based systems. PGI is unique in offering developers the option of migrating from UNIX to Linux or to Windows x64 using either a UNIX-like development environment on native Windows or through Microsoft's Subsystem for UNIX-based Applications environment. PGI 7.1 for Windows platforms includes the ability to use all of the PGI compilers and tools on both native Windows and from within Windows SUA (Subsystem for UNIX-based Applications).
All PGI products include the ability to generate PGI Unified Binary(tm) executable files. The PGI Unified Binary enables developers to leverage the latest processor innovations from both AMD and Intel while treating x64 as a single platform, maximizing flexibility and eliminating the need to target and optimize for two separate processors. Evaluation copies of the new PGI compilers are available from The Portland Group web site at www.pgroup.com. Registration is required.
All registered trademarks and marks are the property of their respective owners.
* SPEC® and the benchmark name SPECfp® are registered trademarks of the Standard Performance Evaluation Corporation. Competitive benchmark results stated above reflect results published on www.spec.org as of October 11, 2007. The comparison presented above is based on the best published performance on competing AMD and Intel Quad-Core platforms. Complete results for the referenced AMD processors are at http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20070908-01998.html, and for the referenced Intel processors at http://www.spec.org/cpu2006/results/res2007q3/cpu2006-20070821-01886.html.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959
PGI Compilers Target Quad-Core AMD Opteron Processors
Portland, OR
August 7, 2007
Optimizing multi-core development tools from The Portland Group enable migration and tuning of applications for Quad-Core AMD Opteron processors
The Portland Group®, a wholly-owned subsidiary of STMicroelectronics and leading supplier of compilers for high-performance computing (HPC), today announced that PGI compilers and development tools now generate code targeted at Quad-Core AMD Opteron™ processors, code-named "Barcelona". PGI and AMD are cooperating to provide these compilers to independent software vendors and customers interested in migrating and tuning applications for AMD's upcoming native quad-core processor.
PGI develops and markets high-performance parallel C/C++ and Fortran compilers and application debugging and performance profiling tools that are widely used by engineers and scientists in high-performance technical computing (HPTC). PGI compilers and tools are designed to extract maximum performance from the latest multi-core processors from both AMD and Intel. All PGI compilers support generation of PGI Unified Binary executables, which enables developers to leverage the latest processor innovations from both AMD and Intel while treating x64 as a single platform, maximizing flexibility and eliminating the need to target and optimize for two separate processor platforms.
"The comprehensive effort by AMD to engage with vendors like PGI and supply us with all the information and support we need to effectively optimize for their forthcoming processors has been outstanding," said Douglas Miles, director, The Portland Group. "The performance tuning we are building in to the PGI compilers in support of Quad-core AMD Opteron will show significant benefits for PGI's core science and engineering customers and hopefully for the wider AMD developer community as well."
Built on AMD's revolutionary Direct Connect Architecture which improves overall system performance and efficiency by eliminating bottlenecks inherent in traditional front-side bus architectures, the Quad-Core AMD Opteron processor will also introduce several major enhancements that the PGI compilers leverage for improved compile-and-go performance: smart code selection to use the full 128-bit wide floating-point units and avoid merge dependencies; low-overhead inline parallel regions to extend efficient auto-parallelization from dual-core to quad-core; alignment of hot loops to take advantage of the expanded 32 byte code fetch window; highly-optimized bit and string library intrinsics that leverage new ABM and SSE4a instructions; instruction scheduling and selection for improved latency and bandwidth; modified software pre-fetching to complement the Level-1 data cache prefetch hardware; and memory hierarchy optimizations to reduce memory access-related conflicts between cores and to improve throughput efficiency.
"As the shift towards multi-core processor environments accelerates, both PGI and AMD recognize the need for better tools that will help the developer community easily and efficiently leverage the computing performance benefits of dual- and quad-core processors," said Earl Stahl, vice president of software development, AMD (NYSE: AMD). "The new PGI compilers and development tools will enable developers to optimize their applications for Quad-Core AMD Opteron processors, allowing them to exploit the significant core and cache enhancements and unprecedented levels of efficiency made possible through AMD's native design, and ultimately deliver greater levels of performance and efficiency for end-users."
Software developers can generate code targeted at Quad-Core AMD Opteron processors using the latest version 7.0-7 release of the PGI compilers and tools. Version 7.1 of the PGI compilers and tools, scheduled for general availability in Fall 2007, will include additional features and optimizations for Quad-Core AMD Opteron.
All registered trademarks and marks are the property of their respective owners.
Media Contacts
Mike Markowitz
STMicroelectronics
michael.markowitz@st.com
(212) 821 8959