Browse Prior Art Database

A Method For Availability Detection of Dynamically Discovered and Uniquely Identified Managed Resources Based on Events

IP.com Disclosure Number: IPCOM000022355D
Original Publication Date: 2004-Mar-10
Included in the Prior Art Database: 2004-Mar-10
Document File: 2 page(s) / 61K

Publishing Venue

IBM

Abstract

Disclosed is a method by which managed resources can be dynamically discovered, uniquely named and their availability information derived by a managing system as events representing state changes on the resources are received. This method provides availability data for dynamically discovered resources without collision of names and is used in Event Management, Service Level Agreement breech and thend analysis as well as data warehouse reporting.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

A Method For Availability Detection of Dynamically Discovered and Uniquely Identified Managed Resources Based on Events

Motivation: I n the IT management arena, resources of interest to the organization are monitored for state transitions, and usually these state transitions are reported as events to management systems. Being able to derive availability information about these resources from events is an important capability that provides input to data warehouse reports and service level agreement breech and trend analysis. The collection and computation of availability information about large amount of resources which many times can have colliding names is a challenge for management solutions because: - querying or polling managed resources for their current state can be performance prohibitive. A event based notification of state changes is required - the monitoring of managed resources state transitions requires the definition of each one of these resources the system. This definition usually requires manual intervention, which can become cumbersome, as the typical number of monitored resources in these environments are in the order of thousands to millions. A dynamic way of discovering these resources is required. - many resources may have the same name, which when conglomerated in a management system, e.g., a data warehouse, could lead to collision and incorrect availability data reporting. For instance, a resource "servlet" named "BuyStock" may be deployed in 3 application servers running in 3 different machines, and each one of these 3 instances of "BuyStock" must have its own availability information. A dynamic method of building unique or hierarchical distinguished names is required.

Overview: In a setup phase metadata is input to the system to describe the type of resource to be recognized and how its information (including unique name and availability) can be derived from an event. When an event arrives at the Event Management System, the event is inspected to try to find a match with any of the metadata and when found a representation of such managed resource is dynamically created (if not already existent). For instance, if you are interested in managing nodes in a network, it is not necessary to define each one of the thousands of nodes, but only provide initial metadata about this type of resource such that its definition can be derived from a state change event. In addition, unique names and availability information (e.g., this resou...