Thinking about How Much Fun It Is to Develop Multi-Threaded Apps

We've just published a new article by Dibyendu Roy, Rethinking Multi-Threaded Design Principles, Part 2. In this article, Dibyendu presents and overview of the laws and principles that relate to application speed-up due to multithreading, then he talks about locks and non-blocking operations and their impact. If you've worked with multithreaded development, your probably familiar with Amdahl's law, which defines the theoretical speed-up of an application when part of its code is parallelized. In Dibyendu's article, he presents the speed-up using this equivalent representation: Speed Enhancement = 1 / (Fs + ((1 - Fs) / N)) Here, Fs is the fraction of the program that is running sequentially, and N is the number of cores or processors. Here's Dibyendu's explanation: if N is 1, we get no speedup, but as N increases (tends to infinity, (1 - Fs)/N becomes 0), the speed is totally dependent on the 1/Fs part - which leads to the conclusion that, to improve execution speed of an application, we need to figure out ways or patterns to make more code execute in parallel. This really is a very interesting equation, and it illustrates why sometimes beginning developers are stunned to find out that a performance improvement they spent days or weeks working on ultimately has little impact on the application's actual performance. The question is: how do you "compute" Fs, the fraction of the program that is...

Date: April, 14 2010
Url: http://www.java.net/blog/editor/archive/2010/04/14/thinking-about-how-much-fun-it-develop-multi-threaded-apps

Others News