• March 2014
    M T W T F S S
    « Feb   Apr »
  • Advertisements

Stanford Engineers Create a Software Tool to Reduce the Cost of Cloud Computing

Stanford Report (CA) (02/28/14) Tom Abate 

Stanford University professor Christos Kozyrakis and doctoral student Christina Delimitrou have developed Quasar, a cluster management system they say can triple server efficiency and provide reliable service at all times, thereby reducing data center operating expenses. Data centers run applications such as search services for consumers or data analysis for businesses, placing varying demands on the data center and using different amounts of server capacity. Cluster management tools apportion the workload and assign applications to specific servers, based on input from software developers about the capacity the applications will require. Developers often overestimate when reserving server capacity, with the cumulative impact leading to significant excess capacity. Quasar begins by asking about the performance that applications require, then ensures adequate server capacity to meet various requirements. “We want to switch from a reservation-based cluster management to a performance-based allocation of data center resources,” Kozyrakis says. Quasar uses an algorithm similar to popular movie and book recommendation algorithms to suggest the minimum number of servers for each application and which applications can run together most effectively. Boosting data center efficiency will be essential for the growth of cloud computing, because electricity demands that threaten to overwhelm power plants preclude the option of continually adding servers.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: