Online services such as Gmail, Apple, Facebook, etc, regularly experience disruptions which leave millions of users upset and frustrated. But it’s not just the users that are affected by disruptions in online services – delays and outages have both financial and operational implications for websites too. For instance, a single hour-long episode of disruption can make Amazon lose millions of dollars.
To help prevent prolonged disruptions in Internet services, researcher Olumuyiwa Ibidunmoye has developed new automated, intelligent algorithms that quickly detect problems and help resolve them.
Today, when the number of online services is rapidly growing as well as the number of their users, it’s getting harder and harder to manage various operational issues that inevitably arise in these situations. The main issue is that the problem may be due to a variety of reasons. For example, delays may be caused by inadequate server resources, coding errors or they may be due to competition between hundreds of applications that usually run on the same server.
To prevent, or at least minimize disruptions, the status of systems needs to be continuously measured, problems quickly detected and diagnosed, and intelligent, all-around solutions applied.
That’s where the new smart algorithms come in. According to Ibidunmoye, they automatically uncover symptoms of problems, rank them with minimal human help, and resolve service delays by adjusting the capacity of server resources in response to demand.
Umea University via Phys.org (https://phys.org/news/2017-12-intelligent-algorithms-disruptions-online.html)