IP Library Granted Patent US 11,496,547
Granted Patent B2
US 11,496,547 · App. 17/487,541 · Granted Nov 8, 2022

Storage system node communication

Inventors: Yoshifumi Mimata (Tokyo, JP); Yoshiaki Eguchi (Tokyo, JP); Naoya Hattori (Tokyo, JP)
Assignee: Hitachi, Ltd.
H04L67/06H04L67/01H04L67/108H04L67/1097
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,496,547
App. No.
17/487,541
Granted
Nov 8, 2022
Kind
B2
Abstract

Provided is a storage system capable of avoiding the increase in communication between nodes in the coordination of the file service and the block service. This is a storage system in which a plurality of nodes, which provide a file service for performing I/O in file units and a block service for performing I/O in block units, are connected via a network, and the storage system comprises a management unit which manages the first file processing unit and the second file processing unit as a pair, sets the first file processing unit to be operable, manages the first block processing unit and the second block processing unit as a pair, and sets the first block processing unit to be operable.

Claims (38)

1. A storage system, comprising:

a plurality of nodes, which belong to a cluster and provide a file service for performing I/O (Input/Output) in file units and a block service for performing I/O in block units, are connected via a network, each of the nodes configured by including a storage device which stores data, a plurality of file processing units which are each configured to receive a file I/O request from a file client via the network and convert the file I/O request into a block I/O request in the file service, and a plurality of block processing units which are each configured to perform processing for executing I/O to the storage device based on the block I/O request in the block service;

a management unit configured to:

manage a plurality of pairs of file processing units of the plurality of nodes, each pair of file processing units includes two file processing units in different nodes,

for each of the plurality of pairs of file processing units, set one file processing unit to an active-type and set the other file processing unit to a standby-type,

manage a plurality of pairs of block processing units of the plurality of nodes, each pair of block processing units includes two block processing units in different nodes,

for each of the plurality of pairs of block processing units, set one block processing unit to the active-type and set the other block processing unit to the standby-type, and

set a plurality of combinations of respective pairs of file processing units and respective pairs of block processing, each combination includes one pair of file processing unit and one pair of block processing unit,

wherein each node includes a first file processing unit of a first pair among the plurality of pairs of file processing units and a second file processing unit of a second pair, which is different than the first pair, among the plurality of pairs of file processing units,

wherein each node includes a first block processing unit of a first pair among the plurality of pairs of block processing units and a second block processing unit of a second pair, which is different than the first pair, among the plurality of pairs of block processing units, and

wherein, for a first combination among the plurality of combinations, based on the block I/O request converted by the active-type file processing unit of the file processing units belonging to the first combination, the active-type block processing unit of the block processing units belonging to the combination performs processing for executing I/O to the storage device of a corresponding node.

2. The storage system according to claim 1 , wherein:

the management unit instructs the block processing unit that configures the block processing unit pair belonging to the first combination to generate a logical volume corresponding to the storage device of the corresponding node, and to assign the generated logical volume to the file processing unit that configures the block processing unit pair belonging to the first combination, and

the management unit instructs the file processing unit to which the logical volume is assigned to recognize the generated logical volume.

3. The storage system according to claim 1 , wherein:

the management unit sets the combinations so that the number of combinations is two or more.

4. The storage system according to claim 1 , wherein:

the management unit sets a larger number of combinations as the total capacity of the storage devices corresponding to each of the nodes that configures the cluster is larger.

5. The storage system according to claim 1 ,

wherein the active-type file processing unit of the file processing unit pair and the active-type block processing unit of the block processing unit pair, each of which belongs to a same combination, are in the same node.

6. The storage system according to claim 5 , wherein:

the standby-type file processing unit of the file processing unit pair and the standby-type block processing unit of the block processing unit pair, each of which belongs to a same combination, are in the same node.

7. The storage system according to claim 1 ,

wherein upon a failure occurring in the node having the active-type file processing unit of the file processing unit pair belonging to a same combination, the management unit gives a file processing unit take-over instruction to the standby-type file processing unit of the same pair for taking over the processing of the active-type one, and gives a block processing unit take-over instruction to the standby-type block processing unit of the block processing unit pair belonging to the same combination for taking over the processing of the active-type one.

8. The storage system according to claim 7 , wherein:

based on the file processing unit takeover instruction, the standby-type file processing unit of the file processing unit pair enables the IP address and starts operation.

9. The storage system according to claim 7 , wherein:

the management unit gives the file processing unit take-over instruction after receiving a reply which is in response to the block processing unit take-over instruction and which indicates that the take-over of the processing of the active-type file processing unit is complete.

10. A method of managing a storage system, comprising:

a plurality of nodes, which belong to a cluster and provide a file service for performing I/O (Input/Output) in file units and a block service for performing I/O in block units, are connected via a network, each of the nodes configured by including a storage device which stores data, a plurality of file processing units which are configured to receive a file I/O request from a file client via the network and convert the file I/O request into a block I/O request in the file service, and a plurality of block processing units which are configured to perform processing for executing I/O to the storage device based on the block I/O request in the block service; and a management unit, the method comprising:

managing a plurality of pairs of file processing units of the plurality of nodes, each pair of block processing units includes two block processing units in different nodes;

for each of the plurality of pairs of block processing units, setting one file processing unit to an active-type and setting the other file processing unit to a standby-type;

managing a plurality of pairs of block processing units of the plurality of nodes, each pair of block processing units includes two block processing units in different nodes;

for each of the plurality of pairs of block processing units, setting one block processing unit to the active-type and setting the other block processing unit to the standby-type; and

setting a plurality of combinations of respective pairs of file processing units and respective pairs of block processing units, each combination includes one pair of file processing unit and one pair of block processing unit,

wherein each node includes a first file processing unit of a first pair among the plurality of pairs of file processing units and a second file processing unit of a second pair, which is different than the first pair, among the plurality of pairs of file processing units,

wherein each node includes a first block processing unit of a first pair among the plurality of pairs of block processing units and a second block processing unit of a second pair, which is different than the first pair, among the plurality of pairs of block processing units, and

wherein, for a first combination among the plurality of combinations, based on the block I/O request converted by the active-type file processing unit of the file processing units belonging to the first combination, the active-type block processing unit of the block processing units belonging to the combination performs processing for executing I/O to the storage device of a corresponding node.

Assignments (1)
COMPANY SPLIT Recorded Aug 20, 2024
From: HITACHI, LTD.
To: HITACHI VANTARA, LTD.
Reel/Frame 069518/0761 →
Priority Claims (1)
JP JP2020-130800 · Jul 31, 2020 · national
Continuity (2)
Continuation 17037841 · Sep 30, 2020
Related Publication 20220038526A1 · Feb 3, 2022