Browse Prior Art Database

Automated Reaction Triggered by External Events in Cloud Environments

IP.com Disclosure Number: IPCOM000237782D
Publication Date: 2014-Jul-10
Document File: 4 page(s) / 173K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system of cloud infrastructures that automatically react to the detection of external events in order to prevent severe impacts to the cloud operations and information technology center hardware.

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

Page 01 of 4

Automated Reaction Triggered by External Events in Cloud Environments

Multiple growth countries exhibit a strong growing demand in information technology (IT), which has led to an increase in the

number of data centers in such countries. In contrast to major market countries, environmental problems can be severe in growth countries and endanger the operation of IT centers. Environmental issues can also occur in major countries, but with a much lower probability. In addition, general hardware vintage effects such as bad part numbers might also endanger IT center operations. These threats are classified as external events, and can cause (but are not limited to):

• Corrosion of electronic components e.g. caused by Sulfide Acid • Hardware failures caused by high temperature • Hardware failures due to vibration • Low air humidity causing electrostatic discharge (ESD) defects
• Hardware vintage issues (e.g., specific hardware part number with systematic problems due to manufacturing issues or due to a bad design)

External events in general endanger a data center entirely or a large amount of the hardware components IT center hardware.

Prior art addresses the following issues with respect to the stated threats:

• Detection of external events by means of sensors (e.g., temperature, humidity, corrosion, etc.) • Alarms in case the sensors detects out-of-specification conditions
• Detection of upcoming potential failures on single hardware components

The novel contribution is a system of cloud infrastructures that automatically react to the detection of external events in order to prevent severe impacts to the cloud operations. The system is a set of modules for automated reaction triggered by external events in cloud environments.

The Cloud Management Module is responsible for the management of the underlying cloud. It receives the reactive task from the detection engine (described below). It identifies and then triggers the defined reactive tasks for the individual systems.

The Bi-Directional Storage Management Module is an extension of the existing Storage Management Module. It receives tasks from the Cloud Management Module and then executes the provided reactive tasks to prevent severe impact.

The Detection Engine uses sensor networks and statistics to detect external events. The Detection Engine collects the hardware location and the failure data, which includes:

• Failure type
• Failure classification

1


Page 02 of 4


• Aggregation of the failures

• Determine total set failure rates (N, mean, sigma)
• Determine subset failure rates (N, mean, sigma)

The Detection Engine then triggers an automatic alert.

The overall schematic of the novel solution is outlined in Figure 1. The Detection Engine is placed in the back-end of the cloud infrastructure. It can manage a single or multiple cloud solutions. It interacts with the C...