Distributed Computer System Resources Control Mechanism Based on Network-Centric Approach

Abstract

In this paper, we present the development of a decentralized mechanism for the resources control in a distributed computer system based on a network-centric approach. Intially, the network-centric approach was proposed for the military purposes, and now its principles are successfully introduced in the other applications of the complex systems control. Due to the features of control systems based on the network-centric approach, namely adding the horizontal links between components of the same level, adding the general knowledge control in the system, etc., there are new properties and characteristics. The concept of implementing of resource control module for a distributed computer system based on a network-centric approach is proposed in this study. We, basing on this concept, realized the resource control module and perform the analysis of its operation parameters in compare with resource control modules implemented on the hierarchical approach and on the decentralized approach with the creation of the communities of the computing resources. The experiments showed the advantages of the proposed mechanism for resources control in compare with the control mechanisms based on the hierarchical and decentralized approaches.One of the main components of resource control system (RCS) of DCS is a scheduler which coordinates the solution of the tasks that arrive for processing to the distributed system by placing them on available computing resources [7]. It is important to meet the requirements of the user and to avoid the idles for the computational resources [8]. In [7] there are identified the following key characteristics of the DCS scheduler: the run time of the task, DCS performance and the response time of the system.In recent years, there was devoted much attention to the study of decentralized resource control techniques for the distributed computer systems to improve the performance of the DCS in the context of its scalability and reliability [9]. In [9] are described the comparative analysis of the existing decentralized distributed computer systems and the mechanisms that are used for resource control. This study contains a comparative analysis of mechanisms of resource control of DCS in different contexts.Additionally, [10,11,12] presents more recent developments of DCS resources control systems, that are implemented on a decentralized basis.The development in [11] uses the exchanges of network messages between DCS compute nodes for the organization of distributed computing. The messages of several types come, each of which provides a specific functions of the algorithm for the resources allocation. This planning framework supports several heuristic algorithms and it also provides the load balancing for the DCS nodes.In [12] there is considered the DCS, which consists of heterogeneous computing elements (CE), which are interconnected by a computer network. In addition to CE, the system has passive nodes, the so-called "Bulletin boards", which are used to allocate users tasks and to collect some service in...

Protecting Hybrid Information Transmission Network from Natural and Anthropogenic Hazards

“…The reason for increasing the motivation of this work is the existing theoretical developments on the hybrid network [13][14][15][16][17] and planned improvements on protection against cyber threats in the information system.…”

Load Balancing Approaches for Web Servers: A Survey of Recent Trends

“…Various task based scheduling exists for improving the efficiency of grid environment [37][38][39]. An approach for selecting the fittest resource is presented in reference [37].…”

The Methods to Improve Quality of Service by Accounting Secure Parameters

“…These methods use various queue management, reservation and feedback algorithms to reduce the negative impact of queues to acceptable to user's level [3][4][5][6]. Experimental and numerical studies conducted in recent decades indicate that many multiservice networks traffic has self-similar properties [2,[7][8][9][10]. Self-similar traffic causes significant delays and packet losses, even if the total intensity of all flows is far from the maximum allowable values [7,10,11].…”