As discussed in section 1.3, there are many scheduling algorithms, each with their own parameters. As a result, selecting an algorithm can be difficult. The first problem is to define the criteria to be used to select an algorithm. Criteria are often defined in terms of CPU utilization, latency, response time, or throughput. To select an algorithm, you must first define the relative importance of these elements. Our criteria can include several measures, such as these: • Maximize CPU utilization subject to the constraint that the maximum response time is 1 second • Maximize throughput such that execution time is (on average ) linearly proportional to the total execution time Once the selection criteria have been defined, evaluate the algorithms considered. There are different methods available to evaluate algorithms: deterministic modeling, queuing models, simulation and implementation. Among these methods, simulation is used to evaluate the algorithm. To get a more accurate evaluation of scheduling algorithms, we can use simulations. Running simulations involves programming a model of the computer system. Software data structures represent the main components of the system. The simulator has a variable representing a clock. As the value of this variable increases, the simulator modifies the system state to reflect the activities of devices, processes, and the scheduler. As the simulation runs, statistics indicating the performance of the algorithm are collected and printed. The data needed for the simulation can be generated in several ways. The most common method uses a programmed random number generator to generate processes; CPU burst times, arrivals, departures, etc., according to probability distribution...... middle of paper ...... scheduling multiple processes on different processors is one of the most difficult problems in multi-process operating systems. The real-time short-term scheduler is well suited to multiprocessor scheduling problems. In this project, a short-term scheduler is introduced to schedule tasks in multi-process real-time systems. The scheduler is based on moving events on a logical time scale and simultaneously allocates different processes to different processors in the system. Analysis of the results obtained from the scheduler reveals that minimization of waiting time and maximization of CPU utilization are achieved. It also indicates that the average wait time is removed before a process is processed. The calculated overall summary statistics reveal that the short-term real-time scheduler can be the solution for scheduling processes in a multiprocessor system..