Browse Prior Art Database

Method and System for Simultaneous Management of Availability, Power Consumption, Predictive Failure Analysis, Performance and Virtualization associated with Machines in a Data Center

IP.com Disclosure Number: IPCOM000202462D
Publication Date: 2010-Dec-16
Document File: 2 page(s) / 58K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method and system for simultaneous management of availability, power consumption, predictive failure analysis, performance and virtualization associated with machines in a Data Center.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Method and System for Simultaneous Management of Availability , Power Consumption,

Predictive Failure Analysis, Performance and Virtualization associated with Machines in

a Data Center

A method and system for simultaneous management of availability, power consumption, predictive failure analysis, performance and virtualization associated with machines in a Data Center is disclosed.

The method involves virtualization transparent migration to perform dynamic server consolidation onto a small set of servers to maintain the required minimal levels of performance. Thus, most of those servers that are not needed for the current workload may then be powered off thereby resulting in significant power savings. Further, when servers fail or workload rises, hot spares may be immediately activated and other servers may be powered on to replace them as hot spares. In addition, the method also utilizes predictive failure analysis to trigger preemptive virtual machine migration away from failing servers. When such a server finally fails, the server may typically be idle.

A controlling policy engine collects current performance measurements, actual server failure state, predicted server failures, and static configurations and policies into account. Thereafter, the controlling policy engine outputs a dynamic configuration. The dynamic configuration includes information indicating virtual machines run on physical machines, machines held as hot spares, machines that must have performances adjustments (i.e., clock and memory speed adjustments) and machines are to be powered off. The dynamic configuration is then send to a second entity. The second entity computes a set of actions to bring the server collection into compliance with the sta...