Welcome RapidMind!

Excited to have RapidMind parallelization experts on board at Intel and on our Ct team! 

Come and join many Intel parallelization experts at the Intel Developer Forum (IDF) September 22-24, 2009 in San Francisco, including folks from the RapidMind and Ct technology team. Hear more about our plans to work with Ct technology and RapidMind. There will be a class by Intel’s Anwar Ghuloum and Amanda Sharp at IDF on Ct, which will soon to be available in beta for you to try out yourself. Ct folks, including me, will also be at Supercomputing 2009, where we will be hosting a Ct related BoF. If you’d like to be involved, please contact me here. Hopefully we’ll see you at these exciting events – your thoughts and feedback on our technology is always welcome!

URL: http://feedproxy.google.com/~r/IntelBlogs/~3/Kb4ekHSL9Zs/

Intel(r) Media SDK beta announced

At SIGGRAPH last week we announced the beta version of the Intel(r) Media SDK to simplify media developers access to video encode and decode (as well as pre-processing). In my previous blogs I had asked about the importance of something like this, and well, we&aposve now got a beta version that you can play around with. A single API that takes the complexity out of hardware accelerated DXVA2 decode, and offers accelerated encode for future Intel silicon. Using the API will help to guarantee that when hardware acceleration is available it will be used, and if not will fall back to optimized software libraries. You can find the beta and more information here. I&aposm looking forward to hearing your fedback!

URL: http://feedproxy.google.com/~r/IntelBlogs/~3/XO29SNPKpsQ/

A Faculty Revisit on Parallel Programming Curriculum – Episode 1

On the 8th, 9th and 10th of July, in Wuhan University, a campus renowned for her beautiful hill and lake side scenery, we had a faculty workshop on curricula development. The workshop was intended to share some new technology updates of parallel programming, with the faculty who either have attended our previous workshops or have been winners of PRC MoE-Intel Model classes, so that they can enhance their existing curricula.

The workshop modules covered go to four categories, threading models consisting of the updated OpenMP threading model (version 3.0) and new module of TBB, threading tools composed of revamped Integrated Architecture and Tools and new module of Intel Parallel Studio, methodology module on Parallel Design Patterns and demonstrative modules of game application parallelization (Destroy the Castle and Smoke Demo). After the modules presentation, we also held a session where faculty presented their experience/successful stories on curricula development and a open discussion among the attendees for any other questions or interests.

The passed family weekend had evicted many details of the event out of my memory while three reflections, in addition to the impression of the beautiful scenery, still stand in my mind and I&aposd like to tell. Firstly, I was kind of "surprised" that the content, though some of them are still at beta, got pretty good interest and reception from the attendees, comparing to our previous workshops. Though not through a formal survey, attendees at the final session told that all modules presented are useful and of interest to them for their curricula update, and would like to follow up consequently. That is really exciting to some one who had to tolerate the dissatisfaction or complaints from the audience each time the faculty workshop was held. And I am eventually not surprised by those feedback when I recall the preparation we&aposve done prior to the workshop – we just did a right thing in inviting the right audience for the right content.

The second image resides in my memory is the topic (argument I&aposd say) on the motivation for the students to get to learn parallel programming after they know the basics of programming. The performance story is one we usually use to attract interest – most students will go and find a job in engineering; if not being able to write performance code to utilize the power of computing platforms, the product they developed will be toasted in the market as their competitors will do a better job. However, that is still some what abstract for those who are still in the Shangri-la before entering the real industry. So finding the (killer) applications for end user (including the students of course) is still The quest for the Holy Grail that had started ever since multi-core platforms appear. And this leads to my following impression.

As my third memory occupation for the workshop, one "surprisingly" close consensus reached in the discussion is that a lot of cases study or examples in different applications, are of big demand. Besides the basic or theoretic materials on parallel programming, examples applying the theory to a real problem can be of great help in getting the attention from students and familiarizing them with the techniques. The examples don&apost have to be uniformed. They actually should various and practical instead, covering a wide range of applications. For students with traditional CS background, parallel algorithms design (like UC Berkley Motifs and development work by ISE) are good; for other students with majors in Information System, alternative examples should be used, e.g. a professor from university in south China is leading a research topic on non-compute-dense applications parallelization. It is a right time to start creating examples/cases study base for world wide academic community now. Faculty with different background and interests can submit the examples he/she creates and get examples from others.

At the end of this month, we&aposll be holding another faculty workshop with the same topics for the faculty from the other parts of China. Looking forward to the feedbacks from them and will report out as consequent episode to this one.

URL: http://feedproxy.google.com/~r/IntelBlogs/~3/5YCVphsbPuo/