Managing data operations in a quorum-based data replication system

Patent metadata

Field	Value
Publication number	US-10126980-B2
Application number	US-201514698874-A
Country	US
Kind code	B2
Filing date	Apr 29, 2015
Priority date	Apr 29, 2015
Publication date	Nov 13, 2018
Grant date	Nov 13, 2018

Abstract

When a request is received to perform a data operation requiring an interaction with any one of multiple data replicas stored on one or more data storage devices and managed by a quorum-based data management protocol in which completion of a data update is reported to an initiator of the data update when acceptance of the data update is reported by a majority of the data replicas, the data operation is routed to be performed using one of a predefined minority of the data replicas if the data operation requires less than strong consistency, is a read-only data operation, and meets a predefined criterion of being computationally time-intensive or computationally resource-intensive, or routed to be performed using a predefined majority of the data replicas if the data operation requires strong consistency or requires a data write operation or does not meet the predefined criterion.

First claim

What is claimed is:

1. A method for managing data operations in a quorum-based data replication system, the method comprising:
partitioning a plurality of data replicas into two partitions,
wherein one of the partitions includes a predefined minority of the data replicas,
wherein the other of the partitions includes a predefined majority of the data replicas,
wherein each of the data replicas is included in only one of the partitions,
wherein the predefined minority of the data replicas is designated to receive data operation requests that require less than strong consistency and are read-only data operations and require at least a predefined measure of any of computation time and computation resources, and
wherein the predefined majority of the data replicas is designated to receive data operation requests that require strong consistency or require a data write operation or require less than the predefined measure of any of computation time and computation resources;

receiving a request to perform a data operation that requires an interaction with any one of the plurality of data replicas,
wherein the data replicas are stored on one or more data storage devices, and
wherein the data replicas are managed by a quorum-based data management protocol in which completion of a data update is reported to an initiator of the data update when acceptance of the data update is reported by more than half of the data replicas;

determining whether the data operation that requires an interaction with any one of the plurality of data replicas requires less than strong consistency and is a read-only data operation and requires at least the predefined measure of any of computation time and computation resources;
routing the request to one of the predefined minority of the data replicas responsive to determining that the data operation requires less than strong consistency and is a read-only data operation and requires at least the predefined measure of any of computation time and computation resources; and
routing the request to one of the predefined majority of the data replicas responsive to determining that the data operation requires strong consistency or requires a data write operation or requires less than the predefined measure of any of computation time and computation resources.

2. The method of claim 1 wherein the routing the request to one of the predefined minority or predefined majority of the data replicas is performed to data replicas that store Resource Description Framework (RDF) linked data.

3. The method of claim 1 wherein the routing to the predefined majority of the data replicas is performed wherein the predefined majority of the data replicas includes a protocol-designated leader.

4. The method of claim 1 wherein the receiving the request to perform a data operation that requires an interaction with any one of the plurality of data replicas and the routing the request to one of the predefined minority or predefined majority of the data replicas are implemented in any of
a) computer hardware, and
b) computer software embodied in a non-transitory, computer-readable medium.

5. A system for managing data operations in a quorum-based data replication system, the system comprising:
a plurality of data replicas partitioned into two partitions,
wherein one of the partitions includes a predefined minority of the data replicas,
wherein the other of the partitions includes a predefined majority of the data replicas,
wherein each of the data replicas is included in only one of the partitions,
wherein the predefined minority of the data replicas is designated to receive data operation requests that require less than strong consistency and are read-only data operations and require at least a predefined measure of any of computation time and computation resources, and
wherein the predefined majority of the data replicas is designated to receive data operation requests that require strong consistency or require a data write operation or require less than the predefined measure of any of computation time and computation resources;

a request manager configured to receive a request to perform a data operation that requires an interaction with any one of the plurality of data replicas,
wherein the data replicas are stored on one or more data storage devices, and
wherein the data replicas are managed by a quorum-based data management protocol in which completion of a data update is reported to an initiator of the data update when acceptance of the data update is reported by more than half of the data replicas; and

a request router configured to
determine whether the data operation that requires an interaction with any one of the plurality of data replicas requires less than strong consistency and is a read-only data operation and requires at least the predefined measure of any of computation time and computation resources,
route the request to one of the predefined minority of the data replicas responsive to determining that the data operation requires less than strong consistency and is a read-only data operation and requires at least the predefined measure of any of computation time and computation resources, and
route the request to one of the predefined majority of the data replicas responsive to determining that the data operation requires strong consistency or requires a data write operation or requires less than the predefined measure of any of computation time and computation resources.

6. The system of claim 5 wherein the data replicas store Resource Description Framework (RDF) linked data.

7. The system of claim 5 wherein the predefined majority of the data replicas includes a protocol-designated leader.

8. The system of claim 5 wherein the request manager and the request router are implemented in any of
a) computer hardware, and
b) computer software embodied in a non-transitory, computer-readable medium.

9. A computer program product for managing data operations in a quorum-based data replication system, the computer program product comprising:
a non-transitory, computer-readable storage medium; and
computer-readable program code embodied in the storage medium, wherein the computer-readable program code is configured to
partition a plurality of data replicas into two partitions,
wherein one of the partitions includes a predefined minority of the data replicas,
wherein the other of the partitions includes a predefined majority of the data replicas,
wherein each of the data replicas is included in only one of the partitions,
wherein the predefined minority of the data replicas is designated to receive data operation requests that require less than strong consistency and are read-only data operations and require at least a predefined measure of any of computation time and computation resources, and
wherein the predefined majority of the data replicas is designated to receive data operation requests that require strong consistency or require a data write operation or require less than the predefined measure of any of computation time and computation resources,

receive a request to perform a data operation that requires an interaction with any one of the plurality of data replicas,
wherein the data replicas are stored on one or more data storage devices, and
wherein the data replicas are managed by a quorum-based data management protocol in which completion of a data update is reported to an initiator of the data update when acceptance of the data update is reported by more than half of the data replicas,

determine whether the data operation that requires an interaction with any one of the plurality of data replicas requires less than strong consistency and is a read-only data operation

Assignees

• IBM

Inventors

• Laden Guy
• Mandler Benjamin
• Tock Yoav

Classifications

• G06F3/0604

  Improving or facilitating administration, e.g. storage management · CPC title

• G06F3/0659 Primary

  Command handling arrangements, e.g. command buffers, queues, command scheduling · CPC title

• H05K999/99

  Electricity · mapped topic

• G06F3/065

  Replication mechanisms · CPC title

• G06F3/067

  Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS] · CPC title