FindCloudHost.com > Cloud Web Hosting News

Engineers at Stanford Develop Software that Could Bring Down Cloud Computing Costs

March 13, 2014
by Staff Writer

Engineers at Stanford University have developed software that is designed to bring down the cost of cloud computing. The engineers, Christos Kozyrakis, associate professor of electrical engineering and computer science, and Christina Delimitrou, who is taking a doctoral degree in electrical engineering, have developed a new software solution that increases the efficiency of computer server clusters by around 300%.

With workloads that fluctuate on a daily basis, data centers need to retain resources in reserve to cater for possible spikes in activity. This means that despite their huge cost, data centers generally only use 20% of their capacity. As the size of data centers increases, the cost of retaining additional capacity grows. Called Quasar, Mr. Kozyrakis and Ms. Delimitrou’s solution allows an algorithm to manage capacity to maximize efficiency by estimating the demands of the various workloads that utilize a data center’s resources and decide the level of resources that should be allocated to specific servers.

Quasar also determines which servers are best suited for specific applications. "Quasar recommends the minimum number of servers for each application and which applications can run best together," explained Ms. Delimitrou. In doing this, utilization rates of 70% have been achieved using a 200-server test bed. That compares favourably with the 20% utilization rates currently achieved.

Quasar has the potential to provide the data center efficiency increases required to fuel cloud computing growth. The two engineers are currently working on improvements to Quasar that will allow data centers with tens of thousands of servers to fully scale operations.

Are you aware of any other innovations that could bring down cloud computing costs? Let us know the details. Add your comments below.

Comment News

Connect with