Back to results
US 10,768,851 B2Granted
Instant-ready active-active storage nodes
AssigneeEMC IP Holding Company LLC
InventorsDeepak Vokaliga, Michael Specht, Mohammed Asher
Life of this patent
Estimating the term…
Abstract
In a data storage system in which a first storage array and a second storage array maintain first and second replicas of a production volume, the replicas are made discoverable and accessible while inconsistent. Each storage array maintains an invalid track list of inconsistencies. Initially, all tracks are marked as invalid. While background synchronization is eliminating inconsistencies, accesses to invalid tracks are resolved by exchanging data associated with IOs and updating the invalid track lists based on IO bias and other factors.
Claims20
Apparatus comprising:
- a first storage array comprising a plurality of interconnected computing nodes, each of the computing nodes comprising at least one processor and non-transitory memory, and a plurality of groups of data storage drives, each group of data storage drives connected with one of the computing nodes, wherein the first storage array maintains a first replica of a production volume comprising contiguous logical block addresses that map to non-contiguous addresses of the data storage drives;
- a second storage array comprising a plurality of interconnected computing nodes, each of the computing nodes comprising at least one processor and non-transitory memory, and a plurality of groups of data storage drives, each group of data storage drives connected with one of the computing nodes, wherein the second storage array maintains a second replica of the production volume comprising contiguous logical block addresses that map to non-contiguous addresses of the data storage drives;
- and
- program code stored on the non-transitory memory of the first storage array and the second storage array, the program code comprising:
- instructions that converge differing characteristics of the first replica and the second replica;
- instructions that cause the first replica and the second replica to be discoverable and accessible to hosts while the first replica is inconsistent with the second replica;
- and
- instructions that resolve accesses to extents of data that are inconsistent between the first replica and the second replica based at least in-part on access bias, where the first storage array has preferential bias over the second storage array.
5 more elements
A method comprising:
- in a network comprising:
- a first storage array comprising a plurality of interconnected computing nodes, each of the computing nodes comprising at least one processor and non-transitory memory, and a plurality of groups of data storage drives, each group of data storage drives connected with one of the computing nodes, wherein the first storage array maintains a first replica of a production volume comprising contiguous logical block addresses that map to non-contiguous addresses of the data storage drives;
- and
- a second storage array comprising a plurality of interconnected computing nodes, each of the computing nodes comprising at least one processor and non-transitory memory, and a plurality of groups of data storage drives, each group of data storage drives connected with one of the computing nodes, wherein the second storage array maintains a second replica of the production volume comprising contiguous logical block addresses that map to non-contiguous addresses of the data storage drives:
- converging differing characteristics of the first replica and the second replica;
- causing the first replica and the second replica to be discoverable and accessible to hosts while the first replica is inconsistent with the second replica;
- and
- resolving accesses to extents of data that are inconsistent between the first replica and the second replica based at least in-part on access bias, where the first storage array has preferential bias over the second storage array.
5 more elements
Dependencies are inferred from the claim text.