Communications of the ACM
Researchers Craft Program to Stop Cloud Computer Problems Before They Start
Credit: DCPR
North Carolina State University (NCSU) researchers have developed software that prevents performance disruptions in cloud computing systems by automatically identifying and responding to abnormal activities before they develop into larger problems.
The program analyzes the amount of memory being used, network traffic, central processing unit (CPU) usage, and other data in a cloud computing infrastructure to determine what behaviors can be classified as normal. The software defines normal behavior for every virtual machine in the cloud and then looks for changes that could affect the system's ability to provide service to its users. If the program identifies a virtual machine that is acting abnormally, it runs a diagnostic that can determine which metrics are affected without exposing other data.
"If we can identify the initial deviation and launch an automatic response, we can not only prevent a major disturbance, but actually prevent the user from even experiencing any change in system performance," says NCSU professor Helen Gu. She notes that once the system is operational, it uses less than one percent of the CPU load and 16 megabytes of memory. During testing, the system identified 98 percent of anomalies.
From NCSU News
View Full Article
Abstracts Copyright © 2012 Information Inc., Bethesda, Maryland, USA
No entries found