Processor Scheduling in High-Performance Computing (HPC) Environment

Processor Scheduling in High-Performance Computing (HPC) Environment

Annu Priya (Birla Institute of Technology, Ranchi, India) and Sudip Kumar Sahana (Birla Institute of Technology, Ranchi, India)
Copyright: © 2020 |Pages: 29
DOI: 10.4018/978-1-7998-4742-7.ch009


Processor scheduling is one of the thrust areas in the field of computer science. The future technologies use a huge amount of processing for execution of their tasks like huge games, programming software, and in the field of quantum computing. In real-time, many complex problems are solved by GPU programming. The primary concern of scheduling is to reduce the time complexity and manpower. Several traditional techniques exit for processor scheduling. The performance of traditional techniques is reduced when it comes to the huge processing of tasks. Most scheduling problems are NP-hard in nature. Many of the complex problems are recently solved by GPU programming. GPU scheduling is another complex issue as it runs thousands of threads in parallel and needs to be scheduled efficiently. For such large-scale scheduling problems, the performance of state-of-the-art algorithms is very poor. It is observed that evolutionary and genetic-based algorithms exhibit better performance for large-scale combinatorial and internet of things (IoT) problems.
Chapter Preview


The science of scheduling mechanism is not new, the very first citation of scheduling mechanism is seen in past 3000-year old by Egyptian. Sun Tzu is researched a scheduling strategy for military perspective subject in 2500 year ago. In morden days, New and Emerging applications in Automotive and IoT space needs diversity of processors to deliver the best-in-class elastic performance/power. The effectiveness of the system depends on the scheduling algorithm which divides the given problem into the tasks and assigns them onto the given processors. These days in the field of Industry, the development of the scheduling tools using GUI continuously increases. Now scheduling mechanism is require in every field such as the business sector, cooperate etc. In the cooperate sector, scheduling mechanism is achieved by controlled project performance and effective tools that are supported by skilled project management and project scheduler. In microcomputer environment system, a skilled cadre of scheduler is used to control scheduling of expensive data. This scheduler helps to organize the data correctly. In real-time, Scheduling strategies centralize the different fields. In the field of business, the Business Intelligence (BI) require skilled tools in real-time to create plans and structure the complex tasks processing by using scheduling technique. These skilled tools are efficiently very reliably to handle the complex calculation of large ongoing business tasks processing. The highly precise algorithms are process on skilled scheduler for scheduling task and also in another field of service operation. These scheduling algorithms are incorporate with other core operation to accomplish the high-level goal of scheduling in the field of BI applications. But there numerous drawbacks of the BI scheduling techniques such as:

  • 1.

    Highly expensive

  • 2.

    Less reliable

  • 3.

    Less credibility

The cost of individual skilled tools is very high. For this purpose, business application tools is not used in our day to day life. Significantly, Scheduling is not only use in the field of business but also uses in the field of games scheduling to minimizing the objective. The most of the gaming club is dealing with game scheduling the problems because of the various conflicting constraint between the different users. Scheduling the game is most challenging task in game clubs (like Arsenal, Manchester United, etc. footwall teams). The Game scheduling helps to generate a schedule based on multiple teams, multi-programs to achieve objectives. Generating schedule for game scheduling with their condition is difficult to solve. In this scheduling processes are self-decision making players which are assigned uniformaly to the machine to minimize load chosen by machine with activation cost. Further scheduling mechanism is used in field of computer applications where scheduling is used for minimization of execution time of the process and also sequence the job. The CPU scheduled is required to achieve time schedule in advanced mathematical models. Now, researcher focused on construction of the scheduling algorithm for large and complex problems. The operating system has two very important functionality i) CPU scheduling and ii) process scheduling. CPU scheduling provides a way to determine which process is assign to the computer power for processing while another process is on hold condition. The process scheduling strategy by using which process manager handles running process from the CPU and the selection of another process. It is very essential part of the multiprogramming operating system. The process has various states which is shown below in Figure 1.

Figure 1.

Process state of CPU various states


Complete Chapter List

Search this Book: