Storage system and node management method

Patent metadata

Field	Value
Publication number	US-11237746-B2
Application number	US-202016810128-A
Country	US
Kind code	B2
Filing date	Mar 5, 2020
Priority date	Jul 29, 2019
Publication date	Feb 1, 2022
Grant date	Feb 1, 2022

Abstract

In a storage system, attributes of nodes are specified as first attribute nodes and second attribute nodes, the I/O processing in the nodes is executed as a plurality of processes including front-end I/O processing executed by any of the first attribute nodes and back-end I/O processing executed by the first and second attribute nodes having a storage device that stores data relating to the I/O processing.

First claim

What is claimed is:

1. A storage system that configures a storage cluster using a plurality of nodes, the nodes comprising:
a storage device that stores data;
a memory; and
a processor that runs a program configured to perform I/O processing on the storage device,
wherein attributes of the nodes are specified as first attribute nodes and second attribute nodes, and
the I/O processing in the nodes is executed as a plurality of processes including front-end I/O processing executed by any of the first attribute nodes and back-end I/O processing executed by the first and second attribute nodes having the storage device that stores data relating to the I/O processing, wherein
the attribute of the node is specified by a user when adding the node to the storage cluster.

2. The storage system according to claim 1 , wherein
the first attribute nodes are upper nodes for which the attribute at an upper level is specified, the second attribute nodes are lower nodes for which the attribute at a lower level is specified, and the upper nodes include a management node, and
the management node specifies the upper node responsible for the front-end I/O processing in the I/O processing of the lower node with respect to the storage device.

3. The storage system according to claim 2 , wherein
a storage pool provided from the storage cluster is managed by being divided into a plurality of levels of tiers, and
the processor of the upper node executes tier control to control any tier of the storage pool in which the storage area provided by the storage device of each of the nodes is to be handled in accordance with the attribute of each of the nodes.

4. The storage system according to claim 3 , wherein
the tiers of the storage pool include a high tier for storage of data with an I/O frequency above a first threshold and a low tier for storage of data with an I/O frequency below a second threshold.

5. The storage system according to claim 4 , wherein
in the tier control, the processor of the upper node handles the storage area provided by the storage device of the lower node only in the low tier.

6. The storage system according to claim 3 , wherein
an eligibility condition based on an I/O frequency is set for each of the tiers of the storage pool, and
in the tier control, the processor of the upper node confirms whether the storage area satisfies the eligibility condition in a current tier, and copies data in the storage area to another tier that satisfies the eligibility condition among the tiers if the storage area that does not satisfy the eligibility condition in the current tier exists.

7. The storage system according to claim 6 , wherein
in the tier control, the processor of the upper node copies data of the storage area that does not satisfy the eligibility condition in a current tier to another tier that satisfies the eligibility condition among the tiers, and then, deletes the data of a copy source.

8. The storage system according to claim 5 , wherein
if a node to be added to the storage cluster has the storage device with low performance below a predetermined standard, the attribute at the lower level is specified for the node.

9. The storage system according to claim 5 , wherein
when a node to be added to the storage cluster does not have an interface required to execute the front-end I/O processing and has an interface required to execute the back-end I/O processing, the attribute at the lower level is specified for the node.

10. A storage system
that configures a storage cluster using a plurality of nodes, the nodes comprising:
a storage device that stores data;
a memory; and
a processor that runs a program configured to perform I/O processing on the storage device,
wherein attributes of the nodes are specified as first attribute nodes and second attribute nodes, and
the I/O processing in the nodes is executed as a plurality of processes including front-end I/O processing executed by any of the first attribute nodes and back-end I/O processing executed by the first and second attribute nodes having the storage device that stores data relating to the I/O processing,
wherein
it is possible to specify a first type of attribute relating to I/O processing and a second type of attribute relating to an accessory function of managing a volume, which is created in a storage pool, as the attribute of each of the nodes, and
the processor of each of the nodes switches the program to be started on the own node in accordance with the first type of attribute and the second type of attribute specified for the own node.

11. The storage system according to claim 1 , wherein
in the node for which the attribute at a top level for management is specified, the memory holds resource management information on all the nodes of the plurality of nodes, and
in the node for which the attribute at an upper or lower level is specified, the memory holds resource management information on the own node.

12. A node management method performed by a storage system that configures a storage cluster using a plurality of nodes,
wherein the nodes include: a storage device that stores data; a memory; and a processor that runs a program configured to perform I/O processing on the storage device,
attributes of the nodes are specified as first attribute nodes and second attribute nodes,
the I/O processing in the nodes is executed as a plurality of processes including front-end I/O processing executed by any of the first attribute nodes and back-end I/O processing executed by the first and second attribute nodes having the storage device that stores data relating to the I/O processing, and wherein
the attribute of the node is specified by a user when adding the node to the storage cluster.

13. The node management method according to claim 12 , wherein
the first attribute nodes are upper nodes for which the attribute at an upper level is specified, the second attribute nodes are lower nodes for which the attribute at a lower level is specified, and the upper nodes include a management node, and
the management node specifies the upper node responsible for the front-end I/O processing in the I/O processing of the lower node with respect to the storage device.

14. The node management method according to claim 13 , wherein
a storage pool provided from the storage cluster is managed by being divided into a plurality of levels of tiers including a high tier for storage of data with an I/O frequency above a first threshold and a low tier for storage of data with an I/O frequency below a second threshold, and
the processor of the upper node executes tier control to control any tier of the storage pool in which a storage area of data provided by the storage device of each of the nodes to the storage pool is to be handled in accordance with the attribute of each of the nodes.

15. The node management method according to claim 14 , wherein
an eligibility condition based on an I/O frequency is set for each of the tiers of the storage pool, and
in the tier control, the processor of the upper node confirms whether the storage area satisfies the eligibility condition in a current tier, and copies data in the storage area to another tier that satisfies the eligibility condition among the tiers if the storage area that does not satisfy the eligibility condition in the current tier exists.

16. The node management method according to claim 12 , wherein
it is possible to specify a first type of attribute relating to I/O processing and a second type of attribute relating to an accessory function of managing a volume

Assignees

• Hitachi Ltd

Inventors

• Hara Akihiro
• Shimozono Norio
• Kawaguchi Tomohiro
• Deguchi Akira

Classifications

• G06F3/067 Primary

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

• G06F2003/0697

  Physics · mapped topic

• G06F3/0665

  at area level, e.g. provisioning of virtual or logical volumes · CPC title

• G06F3/0683

  Plurality of storage devices · CPC title

• G06F3/0644 Primary

  Management of space entities, e.g. partitions, extents, pools · CPC title