Harvesting spare storage in a data center

A method performed by a computing device for selecting servers for storage of replicas of data, the method comprising:

• determining a central processing unit (“CPU”) utilization for each of a plurality of servers;
• determining a reimaging (“REI”) rate for each of the plurality of servers;
• calculating a replica accessibility classification for each of the plurality of servers using the CPU utilization and REI rate as input;
• selecting a first primary tenant having a first replica accessibility classification;
• selecting a first server for storage of the data such that the first server hosts the first primary tenant;
• selecting a second primary tenant having a second replica accessibility classification that is different from the first replica accessibility classification;
• selecting a second server for storage of the data such that the second server hosts the second primary tenant;
• and
• storing the data on the first server and the second server.

A computer-storage media having computer executable instructions embodied thereon, which when executed by a computing device, cause the computing device to perform a method for selecting servers for storage of replicas of data, the method comprising:

• determining a central processing unit (“CPU”) utilization for each of a plurality of servers;
• calculating a replica accessibility classification for each of the plurality of servers using the CPU utilization as input;
• selecting a first primary tenant having a first replica accessibility classification;
• selecting a first server for storage of the data such that the first server hosts the first primary tenant;
• selecting a second primary tenant having a second replica accessibility classification that is different from the first replica accessibility classification;
• selecting a second server for storage of the data such that the second server hosts the second primary tenant;
• and
• storing the data on the first server and the second server.