×

Distributed storage and communication

  • US 9,026,844 B2
  • Filed: 09/01/2009
  • Issued: 05/05/2015
  • Est. Priority Date: 09/02/2008
  • Status: Active Grant
First Claim
Patent Images

1. A method of storing data, the method comprising:

  • provisioning a first level cluster by separating the data into a first plurality of data subsets, and generating parity data from the first plurality of data subsets such that each data subset of the first plurality of data subsets may be recreated from other subsets of the first plurality of data subsets and the data from the first plurality of data subsets;

    subsequent to provisioning the first level cluster, provisioning a second level cluster by separating the first plurality of data subsets and the parity data from the first plurality of data subsets into a second plurality of data subsets comprising a greater number of data subsets than the first plurality of data subsets, and generating parity data from the second plurality of data subsets such that each data subset of the second plurality of data subsets may be recreated from other data subsets of the second plurality of data subsets and the parity data from the second plurality of data subsets; and

    subsequent to provisioning the second level cluster, storing each data subset of the second plurality of data subsets and the parity data from the second plurality of data subsets in separate storage locations that are remote from each other and accessed over an external network.

View all claims
  • 2 Assignments
Timeline View
Assignment View
    ×
    ×