Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Methodology for Isolating Defective Compute Elements from the System Environment

IP.com Disclosure Number: IPCOM000234794D
Publication Date: 2014-Feb-05
Document File: 2 page(s) / 30K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a methodology for temporarily isolating one or more compute elements from an enterprise and server system, in order to remove the disruption to the management console of the computer system until the time when the defective hardware can be repaired or replaced. The methodology effectively removes any requirement to physically remove the compute element except for repair or replace.

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

Page 01 of 2

Methodology for Isolating Defective Compute Elements from the System Environment

A console application that is used to manage and control a system of hardware

elements presents a summarized state or status of the installed hardware elements. From this single point of control, a compute element that is unstable or defective surfaces as a red flag in this summarized status. Once the nature of the defective hardware is understood (or even before), it may be desirable to remove or hide this red flag in the summarized status until the defective hardware is serviced.

The problem is that the current methodologies are not equipped to temporarily remove unstable, defective, or undesirable compute elements from usage by the overall computer system until the hardware has been repaired, replaced, or is once again needed in the configuration.

The compute elements that plug into the chassis of an enterprise system and server system can exhibit qualities of instability and volatility, indicating problems with either the hardware or the firmware of the compute element. To temporarily manage these undesirable qualities, one option is to physically remove or unplug the compute elements. This requires physical access to the system hardware, which is not

always possible or desirable. Another possibility is to allow the red flag to surface,

but run the risk of not being informed at the system level, when other compute

elements fail due to the summarized system status masking the notification of any new errors. The management of a computer system can be disturbed or disrupted by an unstable or defective compute element.

The novel contribution is a methodology for temporarily isolating one or more compute elements in order to remove the disruption to the management console of the computer system, until the time when the defective hardware can be repaired or replaced.

Isolating the compute element from the control and management of an enterprise system prevents the defective compute element from affecting the overall state, status, and management of the computer system. An isolated compute element is

unavailable for exploitation by the system, even though it is plugged and entitled for use by the customer. The compute element is logically isolated rather than physically isolated from the hosting system.

Isolating a compute element is achieved by the customer through an overt action on the console application and is visually indicated to the customer as in "service" mode. By isolating a compute element, it is temporarily r...