Embedded technology forms a humanless real time environment for getting efficient and effective results ( Strohbach et al., 2004). This environment supports closed loop system to get responses with high accuracy. The amalgamate of firmware and hardware defines the embedded system to perform a predefined task. The scheduling algorithm is a technique which determines how much time is allocated for threads and processes. Scheduler is a machine that organizes or maintains schedules. Real time scheduling algorithms are used for various applications and services (He et al., 2019). The main use of scheduling algorithms is to minimize the starvation and Non-deterministic Polynomial (NP) complete problems (Öztop et al., 2018). The scheduling algorithms are classified as static and dynamic. Dynamic scheduling algorithms are based on planning and best efforts. These algorithms are mainly utilized for the platform of multiprocessors to achieve better response than the normal conditions ( Abdelhafez et al., 2019). However, this platform provides optimum solutions to NP complete problem. The system performance can further be enhanced by selecting the appropriate scheduling algorithm ( Bhuiyan et al., 2018). Scheduling toolboxes are used to schedule the tasks on processors. Toolbox task parameters and optimization criterion offered appropriate resources for configuration and defines the scheduling problem using routines.

At any time, the energy consumption is dynamically managed by the Energy aware scheduling algorithm ( Zhang, 2019). In computing, multitasking is the hierarchy execution of various tasks over a span moment and hinders the execution of basic programming tasks ( Ji et al., 2019). This pre-emption technique works based on priority assigned to the tasks available in each segment by sharing common resources to all. In context switching, the current task address of the main program is saved in stack memory before it suspends temporarily the execution of main program when interrupt occurs. If more than one interrupt occurs at a time, interrupts are serviced by processor based on the priorities assigned to them. Remaining code of the main program will be executed by the processor after interrupt service routine program execution is done. This approach has been developed for multi core processors to cover all requirements like low power consumption, execution speed, tardiness task exemption, completion time, long battery life etc. Dynamic Power Management and Voltage Scaling techniques (DPM & DVS) are the two run-time techniques used to enhance the system performance and power consumption ( Bose et al., 2019; Chasapis et al., 2019; Qin et al., 2019). Nowadays, battery based system models are becoming a major important considerations in design. Amalgamate scheduling algorithms are used to allocate the processor time to available tasks in a system for minimizing the execution time of all tasks including tardy tasks etc ( Sreenath & Sukumar, 2013). This improves system's efficiency by calculating the relevant speed of resources, based on the time slots allocated with real time scheduling algorithms. The speed of processors or controllers is initially high and gradually decreases until the deadlines are met, which further helps in saving the power.

Actually, user expects low power consumption and high per formance from any type of system. Power consumption is a main design metric while using battery based systems and the performance of a system can be improved by utilizing the available processors or cores in a system effectively. However, on multicore or processors based system have some difficulties to satisfy the above parameters. If more energy is required to perform any task, it may degrade system performance. Energy consumption during execution of task depends on the number of processors and cores. Workload will be distributed to the available processors or cores in a system based on the types of tasks and precedence constraints. The behaviour of digital computer system is totally different from single processor based systems. The digital computer system requires work migration from one processor / center to another, once priority controls have been taken and communications between processors / cores have been established to share resources. These migrations may take several cycles compared to the single-processor based systems. Optimization and search problems like constrained and unconstrained may be solved using genetic algorithms ( Sahoo & Padhy, 2019).

Programmers eventually faced NP-complete problems while executing the tasks with in a time. The time required to solve the problem increases as the size of the problem increases, which further helps to solve NP-complete problem. NP-complete problems refers "quickly", there is no possible way to find a solution quickly. The NPcomplete problem may be addressed by approximate and effective scheduling algorithms in polynomial time. Time constraint is the basic consideration while scheduling any type of tasks to achieve good performance by taking minimal computation time for tasks execution on multiprocessors. The NP-complete primarily focuses on the tasks flow of execution on multiprocessors / cores by providing interaction directly or indirectly between the tasks with algorithms and NP has drawback while programming the tasks on a digital computer system. Genetic algorithm overcomes the drawback by scheduling the additional tasks with viably and effectively giving high total throughput. The main aim of using scheduling algorithms is to execute the tasks within allocated time, it indicates that all tasks will execute within deadline. These steps will help for executing the tasks with minimal time period and minimizing the power consumption of the system on multiprocessors.

Various scheduling algorithms reported in existing literature are summarized in Table 1.

Table 1. Summarized Report on Scheduling Algorithms

From the Table 1, different approaches for allocation of tasks, scheduling the tasks by fixed or dynamic approaches and reducing the energy consumption using simulators can be examined. The specific list scheduling algorithm allocates the specific tasks to the available processors, which increases the run time speed and improves the performance compared to existing techniques. Various scheduling algorithm simulators are reported in Table 2.

Table 2. Summarized Report on Scheduling Algorithm Simulators

The scheduling algorithm determines the performance of the system by allocating tasks to available processors. Multiprocessor supported embedded systems provide the improved results in terms of real time scenario by real time control for real time applications. We found that many of the research scholars, academicians and students of under & post graduation are initially facing the difficulty while selecting the simulation tool and doing the research on scheduling algorithms. A suitable platform is essential for developing the new scheduling algorithms with the help of existed algorithms to some extent. Traditional real time scheduling algorithms are implemented using standard structure on suitable simulation tools for improved performance in terms of makespan, latency, flow time, utilization, success ratio, throughput and energy consumption. For this concern, we are focused and found that different objectives taken and the satisfied results are obtained in the different articles. We were planned to segregate all the data related to scheduling algorithms and simulation tools for reducing the complexity of selecting simulation tool along with the awareness of existed scheduling algorithms. This article will help to some extent for the new to do research. The main objective of this review is to find the optimum solution on multi-processor based system using scheduling algorithms in supported simulation tools.

This literature survey used for implementing the novel scheduling algorithms by amalgamate or modify the algorithms as per requirements with the help of existed scheduling algorithms and simulation tools. Finally, it executes the tasks faster and enhances the performance of a system by assigning time slots to available processors for the execution of tasks within a stipulated time period.