×

Method of creating hierarchical indices for a distributed object system

  • US 8,429,198 B1
  • Filed: 06/06/2012
  • Issued: 04/23/2013
  • Est. Priority Date: 07/20/2005
  • Status: Active Grant
First Claim
Patent Images

1. A method of searching a data management system for a data object, the method comprising:

  • receiving a search request for a data object in a data management system, the data object comprising a lookup value, the data management system comprising a plurality of data management servers organized into a plurality of physically-dispersed regions and at least one computer hardware processor associated with at least one data storage device, the at least one data storage device organized into an object oriented data store and an index tree, each physically-dispersed region comprising at least one cluster, the at least one cluster comprising at least one node and a shared storage, the at least one node continuously receiving at least one data stream, the at least one data stream being stored in the object oriented data store, the index tree comprising a root, at least one level of joins, and a plurality of leaves, each leaf associated with a sorted structure, each sorted structure associated with a key, the key being a sorted structure membership key, a cluster membership key, a region membership key or a universe membership key;

    performing a universe membership test against the universe membership key using the lookup value to determine whether the data management system stores the data object;

    for the physically-dispersed regions of the data management system storing the data object, iteratively performing a region membership test against the region membership key using the lookup value to determine the physically-dispersed region storing the data object;

    for the clusters in the physically-dispersed region storing the data object, iteratively performing a cluster membership test against the cluster membership key using the lookup value to determine the cluster storing the data object;

    for the sorted structures in the cluster storing the data object, iteratively performing a sorted structure membership test against the sorted structure membership key using the lookup value to determine the sorted structure storing the data object; and

    iteratively searching entries in the sorted structure storing the data object for the lookup value to locate and return the data object.

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