Lab Summary

Figure 1

Writing code that executes concurrently can be a challenging proposition that can require a tremendous amount of time and resources. Writing your own work-splitting routines and synchronization implementation may require a great deal of development time and testing.

In order make parallelization an easier task, Microsoft has encapsulated the hardest parts of parallelization into libraries for developers to exploit. By taking full advantage of the Concurrency Runtime, the Parallel Pattern and Asynchronous Agents Libraries; developers will be writing parallel code in no time without having to worry about the pesky details. Furthermore, Visual Studio 10 will be bundled with Performance tools that will help you to efficiently debug your parallel applications and optimize them accordingly.

Regardless of the parallelization implementation of choice, you should start evaluating your applications in order to find out if there are processor intensive tasks that would benefit from parallelization.