Abstract
Generally fair share scheduling methods is not suitable for I/O bound and interactive applications whereas priority scheduling supports to execute variety of processes and applications. Stride scheduling is a deterministically fair share scheduling scheme that defines the share of processor time allocating to a client. This paper proposes priority-driven stride scheduling using stochastic modeling in distributed system that improves throughput and reduces response time by applying probabilistic study with the concept of fairness in resource distribution. A model-based simulation study is performed to analyze the transitions of clients onto the processor of this proposed scheme.
Highlights
Distributed system of independent computers seems as a single coherent system [1] in which computing systems put together composed of large number of computing units connected by high speed network
[8] presented an event-based stride scheduling method for time critical collision detection that meets realtime constraints by balancing and prioritizing computation spent on intersection tests without starvation. [9] was adopted symmetrically initiated algorithm in distributed system. [10] based on the performance trends of load sharing algorithms and the recommendations for the selection of a load distribution algorithm. [11], [12 a], [13 b] have studied scheduling policies in distributed systems to find the ways to distribute client processes among various processors in order to achieve fairness and performance like minimizing execution time, minimizing communication, maximizing resource utilization, minimizing response time, average completion time per application and multiple independent tasks with load balancing with other overheads
Stride scheduling is a deterministic fair-share allocation mechanism of resources in which tickets are used to define the share of processor time allocating to a client [6]
Summary
Distributed system of independent computers seems as a single coherent system [1] in which computing systems put together composed of large number of computing units connected by high speed network. [5] Firstly suggested and introduced Stride scheduling which is a deterministic proportional-share scheme to allocate resources for client processes and efficiently supports the flexible resource management concept to schedule other resources as per processor time They introduced novel hierarchical stride scheduling algorithm that achieves better throughput accuracy and lower response time. [11], [12 a], [13 b] have studied scheduling policies in distributed systems to find the ways to distribute client processes among various processors in order to achieve fairness and performance like minimizing execution time, minimizing communication, maximizing resource utilization, minimizing response time, average completion time per application and multiple independent tasks with load balancing with other overheads. The priority-driven stride scheduling system proposes smaller waiting and response times to give controls the resource utilization rates of clients that are proportional to the relative fair shares allocated by server using markov chain model in distributed system. [16] and [17] studied brief of stochastic processes and Markov chain Model
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: International Journal of Advanced Research in Computer Science
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
