Desmond File

Blog archive

Intel's Parallel Studio Goes Beta

Time was, even bloated software got faster thanks to constantly increasing processor clock speeds. Those days may be gone, but developers are finding they can take advantage of increasingly powerful multi-core CPUs to speed their apps. The problem is, nowadays developers have to work at it.

Intel lead evangelist James Reinders has been urging developers to start working on mastering parallel programming concepts for years. Intel has long offered cross-platform products like Threading Building Blocks and OpenMP to help native-code Fortran and C/C++ developers parallelize their applications.

Now Intel has teamed with Microsoft to cook up a suite of native code parallel programming solutions that could help Reinders get his message out to Windows developers. Called Intel Parallel Studio, the suite will ultimately consist of four components: Parallel Composer, Parallel Inspector, Parallel Amplifier and Parallel Advisor.

Parallel Composer went to beta last week and is a compiler, a set of parallel libraries and debugging extensions. It's likely to be released in final form in the second quarter of 2009, according to Reinders.

Following soon after are the Intel Parallel Inspector and Amplifier modules, both slated to go to beta around the end of January. The two modules promise to break parallel dev bottlenecks and improve code quality. Developers can use Parallel Inspector to sleuth out bugs that might otherwise not surface in the course of regular testing and QA. Amplifier provides an intuitive performance analyzer for identifying and resolving bottlenecks.

Reinders said the effort is to bring matured parallel tools and concepts to Windows-based developers, and that it's "absolutely critical" to parallel programming uptake. "A lot of parallel programming is about understanding how to write a lock, create a thread or manage a thread pool," he said, noting that with better tooling, all that low-level work "goes away."

In 2010, the final piece of the Parallel Studio -- Parallel Advisor -- will add in-depth code analysis and debugging to help developers focus on areas of code that will yield the greatest benefit.

"A lot of time we spend time adding parallelism to a program and not understanding fully the implications," Reinders said. "Computing the implications of [parallelism] is enormous. It's like computing all the moves on a chessboard. Parallel Advisor can tell you if adding parallelism somewhere will introduce problems with data structures somewhere else."

Are you planning to aggressively adopt parallel programming practices, and if not, why? E-mail me at [email protected].

Posted by Michael Desmond on 12/16/2008


comments powered by Disqus

Featured

  • Uno Platform Wants Microsoft to Improve .NET WebAssembly in Two Ways

    Uno Platform, a third-party dev tooling specialist that caters to .NET developers, published a report on the state of WebAssembly, addressing some shortcomings in the .NET implementation it would like to see Microsoft address.

  • Random Neighborhoods Regression Using C#

    Dr. James McCaffrey from Microsoft Research presents a complete end-to-end demonstration of the random neighborhoods regression technique, where the goal is to predict a single numeric value. Compared to other ML regression techniques, advantages are that it can handle both large and small datasets, and the results are highly interpretable.

  • As Some Orgs Restrict DeepSeek AI Usage, Microsoft Offers Models and Dev Guidance

    While some organizations are restricting employee usage of the new open source DeepSeek AI from a Chinese company due to data collection concerns, Microsoft has taken a different approach.

  • Useful New-ish Features in .NET/C#

    We often hear about the big new features in .NET or C#, but what about all of those lesser known, but useful new features? How exactly do you use constructs like collection indices and ranges, date features, and pattern matching?

  • TypeScript 5.8 Beta Speeds Program Loads, Updates

    "TypeScript 5.8 introduces a number of optimizations that can both improve the time to build up a program, and also to update a program based on a file change in either --watch mode or editor scenarios."

Subscribe on YouTube