Happy Birthday OpenMP! About 15 years ago, the OpenMP ARB Corporation was incorporated as a Minnesota  non-profit company. Not-many mains stream parallel language can make that claim. I think it was Bjarne who commented that new parallel languages get invented by the hundreds every year. Fortunately, nearly the same number also die every year. Our longevity in this rapidly changing world is a testament to our importance, the efforts of our members, and our intention to keep pace, and even lead the world in adoptions that allow people to take... [More]

Tags:  c++ isocpp.org sc12 4.0 meeting openmp accelerators

In late January, I hosted the OpenMP January 2012 Face-2-Face (F2F) meeting in San Francisco at the Serrano hotel attended by 24 experts. In the words of one expert, it was one of the most productive meeting we have had in a long time. In this meeting, we increased the urgency to drive towards OpenMP 4.0 with subgroups to discuss affinity, taskgroups, accelerators and high level vector language, Fortran 2003 integration, and error model.   Currently, the Affinity subgroup is lead by Christian Terboven, Task subgroup is lead by Federico... [More]

Tags:  future accelerators openmp fortran model affinity error vector taskgroup 2003

At SC11 in Seattle, a large IBM team arrived to help showcase IBM's effort to support High Performance Computing. A team will help with talking to the Scientific Computing Users Group called SciComp or SP-SXXL. I will be mostly responsible for looking after OpenMP. OpenMP has a booth where we hosted many questions from users and discussed future. we have been a  presence in SC for many years and frankly we almost don't have to spend the large sum of money to be there to remind people of our deep connection with High , but we would be... [More]

Tags:  sc11 openmp bof

Its been quite a few months since my last update and I apologize for the lack of blog updates since the last Standard meeting. Things have been hectic indeed. Up to now, I have been enjoying my super-techie life in September of last year and while giving a keynote speech to open the Karlsruhe Institute of Technology (KIT) Steinbuch Centre for Computing which does a great job of showcasing GRID computing, I was asked to become the CEO of OpenMP . First a few things about the KIT school where the audience was very interested in the new C++11... [More]

Tags:  openmp ceo

After almost two years of work, OpenMP 3.1 has been released. Here is the announcement on it and the new specification in pdf online.   As I mentioned in the previous note on the public comment draft , and in this trip report and Steven Perron's posting about the new content, 3.1 offers something for everyone. The main purpose of OpenMP 3.1 was to redefine and extend existing specification, and maintain backward compatibility. This was also a bug-fix release in that it clarified a number of examples in OpenMP 3.0 and fixed about 20 minor... [More]

Tags:  specification parallel 3.1 openmp

I should have sent this a lot earlier. But as previous posts https://www.ibm.com/developerworks/mydeveloperworks/blogs/5894415f-be62-4bc0-81c5-3956e82276f3/entry/the_view_or_trip_report_from_the_feb_2010_openmp_santa_clara_meeting183?lang=en have indicated,  we have been hard at work finishing the OpenMP 3.1 specification. The announcement went out that we are at a public review period right now in February. Although the review is closing soon, you still have some time to hurry over to the OpenMP forum and check out the draft and add comments.... [More]

Tags:  draft openmp specification 3.1

 IBM members are active in designing the feature updates to the new OpenMP specification. We have supported OpenMP since its inception. our latest XL C, C++ and FORTRAN compilers were among the first to fully support the 3.0 specification. Steven has highlighted a few of the last features in this post mostly related to completion of 3.0 support in FORTRAN: https://www.ibm.com/developerworks/mydeveloperworks/blogs/5894415f-be62-4bc0-81c5-3956e82276f3/entry/new_openmp_features10?lang=en At the last meeting which occurred in Feb 1-3 (I know I am... [More]

Tags:  3.1 openmp new features

OpenMP is an add-on to the C, C++, and FORTRAN programming languages that is meant to give the programmer an easy and portable way to parallelize their programs. This is done by adding directives (or pragmas), run-time routines, and environment variables that allow the programmer to control how a program is parallelized. The XL Compilers support OpenMP, and, in this latest release, have added support for two important parts of the OpenMP 3.0 specification, and improved its implementation of another. The final important change is the... [More]

Tags:  fortran c/c++ openmp parallel cppcafe

The following is a a private communication from an IBM engineer Matthew Markland who asked a great question. I do not claim great expertise but I feel that there is enough of an opinion piece that some folks may like to see this discussion or continue it. I have edited the response somewhat but it is largely in tact and reprinted with Matthew's permission. Note I have no insight into PGI or any other product other then what I read in public articles, and as such makes no product claim. Any opinion regarding other company remain necessarily my... [More]

Tags:  openmp cppcafe opencl

Let's get to the good stuff. OpenMP 2.5 did not really specify what constructors should be called with various private/first/lastprivate/threadprivate: In some cases, it did not even specify that they should apply to non-PODs (Plain'ol Data, i.e. C structs). OpenMP 3.0 changed that. Beside specifying non-PODs, it also specified precise rules for the constructor sequence that is in line with what the semantics would require. Firstprivate: For instance, it would specify that a firstprivate for a class type variable should expect an accessible... [More]

Tags:  openmp cppcafe c++

This part will talk about enabling threadprivatization of static class member variables. In 2.5, as a result of ambiguous language, the support for this was inconsistent. In general, it would claim that a threadprivate variable must be namespace, file or block scope. In 3.0, this code is now allowed: class T { public: static int i; #pragma omp threadprivate(i) }; This may seem a trivial change, but for C++, it enables a powerful idiom of singletons and allocators, which all rely on static class member variable. Next posting will... [More]

Tags:  openmp cppcafe c++

This was one of the question asked at SC 08. I will try to answer that here. I will start and add more as I move through the various topics. OpenMP 3.0 had better support for C++ in the following areas: Parallelization of RandomAccess Iterator loops with strict canonical operators Threadprivatization of static class members Unsigned loop control variable support Fully specify constructor call requirement in private/first/lastprivate/threadprivate better match with the C++0x memory model For-Worksharing with Iterator loops: We specifically... [More]

Tags:  openmp cppcafe c++0x memory_model

I should include an obligatory photo of the OpenMP Booth with our CEO Larry Meadows. Many members worked tirelessly, especially Larry at making sure the Booth was put up, taken down and staffed properly. We also met many people who specifically dropped by to see what was going on, in this first year where we had a booth. Many Thanks.

Tags:  conference openmp cppcafe supercomputing

The OpenMP Birds-of-a-Feather session at SC08 was very well attended. The room was full to overflowing, with approximately 60-80 people. while OpenMP had BOFs at SC in prior years, this is actually the first year that OpenMP has had a Booth on the Exhibitor floor as well. The BOF had many elements, including what is new in OpenMP 3.0. They were: 1. Welcome and summary of ARB news - 5 mins - Larry Meadows 2. The three greatest things about OpenMP 3.0 and the three most important things left out of OpenMP 3.0 - 15 mins - Tim Mattson 3. Tasking -... [More]

Tags:  bofs supercomputing cppcafe openmp