It seems that the world of academia is coming up with a new way for cloud platforms — which offer individual computer users and companies access to powerful computers and software applications hosted by remote groups of servers — a kind of early warning system that can predict bottlenecks that slow operating speed and then address them before they become a major problem for businesses.
The research comes from the University of North Carolina.
“Previously, something bad would happen and you’d be left trying to figure out what took place. Often, you’d be unable to recreate the exact conditions that created the problem,” says Dr. Xiaohui (Helen) Gu, an assistant professor of computer science and co-author of a paper describing the new research, in an article I read recently. “However, if you can predict an anomaly, you are able to track the exact conditions that are leading up to a problem, diagnose what is wrong and put corrective actions into place much more quickly.”
The list of things that can go wrong and interfere with smooth operation of hosting services, cloud computing and data centers is long. Just a few: shrinking user capacity and host failures — resulting in violations of service agreements, financial penalties, even loss of clientIs.
In order to be on target predicting abnormalities, UNC researchers created a series of models that examine system activity in a variety of different contexts. In other words, the models are able to determine what constitutes normal behavior under a lot of different circumstances. Since the models do a good job of defining normal behavior, they are able to pretty accurately identify abnormal behavior, said the article.
Gee, this sounds an awful lot like what we at Monitis do for customers of cloud services providers, and it’s called Web Load Tester. With this service we help companies ensure their sites web pages continue to work as designed when more visitors than expected to show up and buy your products. Heavy user traffic is never a bad thing, unless your web system is under too much stress. Web LoadTester helps you determine how your system responds to that traffic. The service is available 24/7 to ensure your websites and applications are ready for any number of visitors whenever they arrive.
Glad to see UNC and Monitis are in sync working on ensuring smooth running of cloud services, even if our efforts are for different sets of audiences.