English | 中文 | Русский | Français
2,856 Posts served
8,606 Conversations started
Posts related to Threading Building Blocks.
After getting lost in the twisty passages of compiler inter-procedural optimization, Robert returns to the simple path and shows a practical example of why it's important to optimize your serial code before parallelizing it.
Join us on February 10, 2010 at 9:00am (Pacific) for a Webinar on the new Intel Software Assessment Tools. Webinar Overview: Intel has developed a new series of fast and free tools designed specifically for Intel Software Partner Program members, enabling them to assess code quickly and easily. This webinar will be an educational session with Intel [...]
The courses are rolling out. Windows C++ developers register now! Intel is offering a one day course on threading and parallelism! Last year I taught a pilot class on threading and parallelism. We had great reviews from those who attended this pilot course. In the post pilot class survey 100% of attendees said they would recommend the course to their peers. We [...]
In December I had fun participating in a couple of web seminars about a product for finding common memory errors as well as thread deadlocks and data races: Intel Parallel Inspector. Many of you might have missed these so let me tell you about them. The first web seminar was one I got to present [...]
Recently, the Intel® Software Partner Program introduced a new software assessment tool for multi-core, the Intel® Concurrency Checker. This tool assesses the multi-core scalability of applications by checking the concurrency. We interviewed an Intel Product Marketing Engineer to discover more about the Intel Concurrency Checker and created a video podcast of the interview. Watch the video podcast [...]
Intel Parallel Amplifier can help identify which synch objects impact performance the most. The Locks and Waits analysis pinpoints synch objects that cause threads to wait and lists them in descending order.
The Intel® Software Partner Program has introduced a set of software assessment tools for two specific Intel technologies for program members. Measure your software's multi-core scaling with the Intel® Concurrency Checker, the software assessment tool for multi-core. Assess your game's playability on systems with Intel® Graphics Technology with the Intel® Graphics Checker, the software assessment [...]
This year we launched intel Parallel Studio. Intel Parallel Studio is designed to improve the productivity and help the Visual C++ developer create good threaded software that scales. It includes components that help express parallelism (Intel Threading Building Blocks), performance libraries, as well as tuning and correctness checking software. Intel Parallel Inspector checks for both memory [...]
Learn more about Intel(R) Parallel Studio Tool suite.
Intel® Parallel Studio Service Pack 1 is now available, adding support for Windows* 7. SP1 is well worth downloading and installing - here are some of the reasons: Parallel Inspector and Parallel Amplifier can be driven (for automating test suites) from the command line now. Bug fixes - of course - not many issues needed fixing, but you [...]
On the way to composing a first thread-parallel version of n-body code, Robert points out parallelization has already been occuring, using the Intel compiler and its vectorization of simple loops.
Let me share some recent practical experience with memory management issues when developing a multi-threaded application. This can probably be a rather common case (as recent post by Roman Dementiev and its follow up discussion demonstrates), and I’d be happy if my experience were helpful for others. Working on CAD Exchanger I am designing one of [...]
Forced to revisit the question of accumulating forces one more time, Robert tests addForce(i,j) and discovers that while accelerations are a little faster, it's not much and a much more complicated story than he realized.
Having found the function that consumes the most time, this episode shows the process of drilling down into the hot source and optimizing it BEFORE going parallel.
Sign up here: http://makebettercode.com/ct_tech/survey.