Browse DevX
Sign up for e-mail newsletters from DevX

Tip of the Day
Language: C++
Expertise: Intermediate
Apr 24, 2000



Building the Right Environment to Support AI, Machine Learning and Deep Learning

Designing Applications for Symmetric Multiprocessing Environments

Many modern operating systems support Symmetric Multiprocessing (SMP)—a hardware architecture that uses two or more processors on one machine and enables you to achieve true multitasking. On SMP architectures, multithreaded programs can execute simultaneously, with each thread running on a separate processor. However, to gain the best performance, a multithreaded application has to be designed with extra care. First of all, remember that not every task can be split into separate threads. The bane of poorly designed multithreaded apps is threads that must execute exclusively. It's better to avoid threads in the first place if they can't execute concurrently. Furthermore, you shouldn't allocate a separate thread for a task if that task takes a very short time to execute (a few hundredths of a second or less). The time needed to perform the context switch between threads may outweigh any performance gain in this case.
Danny Kalev
Comment and Contribute






(Maximum characters: 1200). You have 1200 characters left.



Thanks for your registration, follow us on our social networks to keep up-to-date