Browse Prior Art Database

Hardware Location Probe for Cooling Efficiency Within a Distributed Cloud Environment

IP.com Disclosure Number: IPCOM000232592D
Publication Date: 2013-Nov-19
Document File: 2 page(s) / 26K

Publishing Venue

The IP.com Prior Art Database

Abstract

Described is a method for fully cost optimizing across a cloud environment. Factors such as location, geography, type of cooling solution, climate and season, time of day, and server load would be gathered from the distributed cloud environment to automate the best cost per flop. Additionally, outlier data points would signal either for better distribution or smothering a malignant process such as a Distributed Denial Of Service (DDOS) attack.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 2

Hardware Location Probe for Cooling Efficiency Within a Distributed Cloud Environment

In a distributed cloud environment, resource-intensive tasks can cause temperature of cloud hardware components to rise. Cloud management software currently does not move tasks based on temperature of hardware components.

Moving resource utilization based on power efficiency.

Tied to a regional data center.

Break down intense operations based on heat dissipation on a single node and moving to

multiple nodes to balance cooling costs.

Move tasks within a cloud based on temperature.

    When a temperature rise is detected, tasks running on hot hardware can be moved to cooler hardware. In similar fashion, when a cooler temperature is detected, tasks can be moved from hotter hardware to the cooler hardware.

Monitoring hardware activities for potential intrusion

    In a cloud computing environment, temperature of each of the components is typically monitored, including CPUs, memory, cards, blades, chassis, racks, and server rooms. When a spike in temperature occurs, the cloud management software is notified, which then determines whether the spike in temperature was expected. If the spike in temperature is due to deployment of a task on the hardware that saw the temperature spike, the temperature spike was probably expected by the cloud management software. But, if the temperature spike is caused by a Distributed Denial of Service (DDOS) attack, the temperature spike would be unexpected by the cloud management software. The notification of the temperature spike could thus inform the cloud management software of the DDOS attack, and a system administrator could then be notified. Furthermore, the cloud component (VMs, etc.) can be migrated.

Monitoring heat dissipation within a cloud infrastructure for cross global live migration

    When a spike in temperature is expected, the cloud management software can determine whether there is cooler hardware available, and if so, one or more tasks can be migrated from hotter hardware to cool...