Granted Patent B1

US 9,122,627 · App. 13/492,330 · Granted Sep 1, 2015

Inventors: Siew Yong Sim-Tang (Saratoga, CA); Semen Alexandrovich Ustimenko (Novosibirsk, RU)

Assignee: Dell Software Inc.

G06F11/1076G06F11/108G06F11/1088G06F11/1092

Abstract

Claims (19)

1. An erasure encoding method comprising:

providing a plurality of computer systems forming a distributed computer network, each computer system having a management process operating thereon;

providing a plurality of distributed persistent memories, each persistent memory interoperably coupled to at least one of the computer systems and operative to receive constant updates;

distributing via an erasure encoding process a plurality of data sets across the distributed persistent memories, each data set including at least one data block and at least one check sum block;

initializing the management processes to operate as a cluster;

wherein during the initializing of the management processes to operate as a cluster, each management process:

agrees to enter into an initialization mode;

enters into initialization mode;

suspends regular run time processing; and

registers its data-set management-process management responsibility;

identifying the distributed persistent memories associated with each data set of the plurality of data sets;

classifying whether the at least one data block and at least one checksum block of each data set are good;

verifying if there are enough good blocks to recover each data set of the plurality of data sets;

performing a consistency check of the good blocks; and

recovering each data set of the plurality of data sets using the good blocks.

2. The method of claim 1 further comprising, for each management process, registering a data-set management-process management responsibility wherein the data-set management-process management responsibility is determined using a common responsibility distribution algorithm.

3. The method of claim 1 wherein each data set is associated with one management process having responsibility for performing encoding and decoding of the data set.

4. The method of claim 1 further comprising determining by a responsible management process, via a sequence indicator, whether each data block and each checksum block of each data set is modified completely.

5. The method of claim 4 wherein the sequence indicator is a sequence number entered into a header and a trailer of the at least one data block and the at least one checksum block of each data set.

Assignments (28)

Continuity (4)