Method for optimizing performance and power usage in an archival storage system by utilizing massive array of independent disks (MAID) techniques and controlled replication under scalable hashing (CRUSH)

A computer-readable medium having computer-executable instructions for performing a method of drive management and data placement in an archival storage system which includes a set of drives, said method comprising:

• mapping a plurality of redundant data stripes onto the set of drives;
• selecting a first active data stripe from the plurality of redundant data stripes, the first active data stripe being located on a first subset of the set of drives;
• placing the first subset into a first power state;
• placing a second subset of the set of drives into a second power state, the second power state being a lower power state than the first power state;
• writing a first set of data to the first active data stripe;
• and
• when the first active data stripe contains an amount of data that is smaller than a maximum data capacity value of the first active data stripe, selecting a second active data stripe from the plurality of redundant data stripes, the second active data stripe being located on a third subset of the set of drives, the third subset including at least one drive of the second subset.

An archival storage system, comprising:

• a set of disk drives;
• means for mapping a plurality of redundant data stripes onto the set of drives;
• means for selecting a first active data stripe from the plurality of redundant data stripes, the first active data stripe being located on a first subset of the set of drives;
• means for placing the first subset into a first power state;
• means for placing a second subset of the set of drives into a second power state, the second power state being a lower power state than the first power state;
• means for writing data to the first active data stripe on the first subset;
• and
• when at least a portion of the data has been written to the first active data stripe and the portion of the data is smaller than a maximum data capacity value of the first active data stripe, implementing Controlled Replication Under Scalable Hashing to select a second active data stripe from the plurality of redundant data stripes, the second active data stripe being located on a third subset of the set of drives, the third subset including at least one drive of the second subset,
• wherein the system implements Massive Array of Independent Disks techniques for drive management and implements Controlled Replication Under Scalable Hashing for data placement.