STUDY OF MULTITHREADING ABILITY OF CENTRAL PROCESSING UNIT

Abstract

In computer architecture, multithreading is ability of a central processing unit (CPU) or a single core within a multi-core processor to implement multiple processes or threads concurrently, appropriately supported by operating system. it plea differs from multiprocessing, as within multithreading processes & threads have to share resources of a single or multiple cores: computing units, CPU caches, & translation lookaside buffer (TLB). Multiprocessing systems include multiple complete processing, multithreading targets to increase utilization of a single core by using thread-level as well as instruction stage parallelism. As two approaches are complementary, they are sometimes combined within systems within multiple multithreading CPUs & within CPUs within multiple multithreading cores. A scheduler might aim at one of several goals, for example, maximizing throughput, minimizing response time, or minimizing latency, maximizing fairness. In practice, these goals often conflict, thus a scheduler would implement a acceptable agreement. Preference is given to any one of concerns mentioned above, depending upon user's needs & objectives. Objective of research is to enhance efficiency of scheduling dependent task using enhanced multithreading.