Analysis of New Multi Level Feedback Queue Scheduler for Real Time Kernel

摘要

Most contemporary multi-task real-time systems use a Multi-level Feedback Queue (MLFQ) model to implement their processor scheduler. However, there exist some problems with this model when facing the fast growth of real time systems and handhelds, in which requirements for interactivity and the growth of system loads need to be taken into corresponding consideration. In this research paper, the comparison of different preemptive scheduling algorithms is performed. The simulation of new multi level feedback queue (NMLFQ) is written in C++. NMLFQ simulates the CPU scheduler by using four queues and two different kinds of jobs: regular and batch. The simulation is highly interactive and user-friendly. The simulation results show that the proposed scheduler not only has a good response time but also keeps the system with ideal throughput.