Consistent Hashing and Real-Time Task Scheduling in Fog Computing

Consistent Hashing and Real-Time Task Scheduling in Fog Computing

Geetha J. J., Jaya Lakshmi D. S., Keerthana Ningaraju L. N.
Copyright: © 2022 |Pages: 17
DOI: 10.4018/978-1-7998-8161-2.ch013
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Distributed caching is one such system used by dynamic high-traffic websites to process the incoming user requests to perform the required tasks in an efficient way. Distributed caching is currently employing hashing algorithm in order to serve its purpose. A significant drawback of hashing in this circumstance is the addition of new servers that would result in a change in the previous hashing method (rehashing), hence, goes into a rigmarole. Thus, we need an effective algorithm to address the problem. This technique has served as a solution for distributed and rehashing problems. Most of upcoming internet of things will have to be latency aware and will not afford the data transmission and computation time in the cloud servers. The real-time processing in proximal distance device would be much needed. Hence, the authors aim to employ a real-time task scheduling algorithm. Computations referring to the user requests that are to be handled by the servers can be efficiently handled by consistent hashing algorithms.
Chapter Preview
Top

2. Literature Review

This is a paper (Li, G et al. 2019) that studied resource scheduling problem in fog computing by applying the FCAP algorithm to cluster fog resources. It is initiated by narrowing the range of user requirements for matching resources. They also proposed the RSAF algorithm to accomplish the task of scheduling resources. From the experimental analysis, the objective function value of the FCAP algorithm is seen to have faster convergence speed compared to that of the FCM algorithm. Moreover, the proposed RSAF algorithm will be efficient in matching various user requests with the appropriate resource categories quickly and also enhance user satisfaction. They have left dynamic changes of resources for their future work to produce new scheduling strategy to improve the utilization of resources and ensure user satisfaction. This gave us the motivation to work on the real time task scheduling algorithm.

The main objective of this paper (Choudhari, T et al. 2018) is to propose a system which has a Fog layer sandwiched between the client and the Cloud layer to cater to the applications and users which are having very low tolerance for latency. The higher priority tasks are to be processed at first in the Fog layer. If all the data centers in the Fog layer are busy at that point of time, then the high priority task is propagated to the next Cloud layer for timely execution without affecting the dependent task. The data centers present in this Fog layer of a particular region, it can communicate with each other to check availability of servers and for load balancing protocol. This paper added to our knowledge regarding task scheduling at Fog level for prioritized task.

This paper (Xu, X et al. 2018) aims to achieve dynamic load balancing for each type of computing node found in the fog and cloud layers by employing a dynamic resource allocation method, named DRAM. It also aims in achieving load balancing. A system framework is presented and load balancing is the analyzed accordingly with the fog nodes. They implement the DRAM method based on the static resource allocation and dynamic resource scheduling for the various fog services. As a result of the obtained experimental evaluations and comparison analysis that they were able to carried out, the verified the validity of their proposed method. This paper gave insights on the approaching dynamic resource allocation of incoming user requests and applications.

Complete Chapter List

Search this Book:
Reset