Wednesday, 3 August 2016

Task Scheduling in Parallel Processing: Analysis

Task scheduling in parallel processing is a technique in which processes are assigned to different processors. Task scheduling in parallel processing use different types of algorithms and techniques which are used to reduce the number of delayed jobs. Now a days there are different kind of scheduling algorithms and techniques used to reduce the execution time of tasks. As task scheduling the NP-hard problem and no one can say the about the best algorithm proposed so in this paper we will review some of the task scheduling algorithms and other techniques.

Parallel processing

Parallel processing is dividing the process into multiple processes and execute them concurrently by the use of more than one CPU or processor. Before dividing the process it is checked whether the process is divisible or not, if it is not then the process is executed as a whole and if it is divisible then these processes can be mapped among the processors separately, after execution these processes are reassembled and finally the processed is completed as shown in Figure 1.

Parallel processing is used due to some reasons, i.e. it provide concurrency, save time, solve larger problems, maximize load balancing and make a good use of parallel hardware architecture . In multiprocessor environment parallel processing has two kinds of processorsheterogeneous and homogeneous , in heterogeneous the processors are of different kind of speed and cost while in homogenous there are same kind of processors in all perspective  as shown in Table 1. By adding extra processors it is possible to reduce the execution time of a task.

No comments:

Post a Comment