Software system for inventory and assessment of the wear of computing machines from a network of grid data centers

Abstract

The efficient functioning of grid data centers is one of the most important factor that preoccupies the academic world (professors, researchers, PhD students and so on). The trends, in this continuous growth, of the use of grid technologies in vast and different fields lead to the necessity of the development of methods of optimizing the efficiency of the grid data centers infrastructure. The main objective of the paper consists in developing a computer aided management system for keeping track and assessment of wear off of computing machines from a network of grid data centers (more accurately said, it is a “software system for decision support”). The system will be able to perform real time alarming of the personnel of data centers for possible incidents of for the rising possibility of abnormal functioning of this data centers. This objective will be accomplished through the development of a database that will store data about the hardware equipment components of the grid cluster behavior, data collected form the starting point of the implementation, in real time. The data is acquired through many sensors attached to the components of the cluster servers, like the hard disks, memories and processing units. All this data will be available through a secure web portal, the access to this data being available only to authorized personnel. The database will provide the needed information and data to the decision support software system. The implementation of this software will lead to a growth in functionality of the grid clusters through forecasting the possible malfunctions that will allow the optimized scheduling of hardware maintenance, software reinstall and configuration. More than that, will function as an emergency response system for the management of the network of grid data centers.