IP Library Granted Patent US 12,298,913
Granted Patent B2
US 12,298,913 · App. 18/178,953 · Granted May 13, 2025

Storage system

Inventors: Kentaro Shimada (Tokyo, JP); Nobuhiro Yokoi (Tokyo, JP); Masahiro Tsuruya (Tokyo, JP)
Assignee: HITACHI VANTARA, LTD.
G06F12/10
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 12,298,913
App. No.
18/178,953
Granted
May 13, 2025
Kind
B2
Abstract

A protocol chip writes a request from a host apparatus to a shared memory. One of the plurality of processors reads the request from the host apparatus from the shared memory through an address translation unit and writes a response to the request to the shared memory through the address translation unit. The protocol chip reads the response from the shared memory and sends the response to the host apparatus. In the case where a first processor reboots, the first processor performs a reboot process of a first address translation unit but does not perform the reboot process of the shared memory. A second processor reads a first request addressed to the first processor from the host apparatus through a second address translation unit and writes a first response to the first request to the shared memory through the second address translation unit.

Claims (39)

1. A storage system for processing a request from a host apparatus, the storage system comprising:

a first protocol chip configured to control a protocol used for communication with the host apparatus;

a plurality of processors configured to control the storage system;

a first shared memory that can be read and written by the first protocol chip and the plurality of processors; and

a plurality of address translation units each of which connects a corresponding processor of the plurality of processors and the first shared memory, wherein

the address translation units translate addresses used by the respective corresponding processors of the plurality of processors into addresses used to read from or write to the first shared memory,

the first protocol chip writes the request from the host apparatus to the first shared memory,

a first processor among the plurality of processors reads the request from the host apparatus written to the first shared memory by the first protocol chip from the first shared memory through a first address translation unit among the plurality of address translation units that is connected to the first processor and writes a response to the request to the first shared memory through the first address translation unit,

the first protocol chip reads the response to the request from the host apparatus written by the first processor from the first shared memory and sends the response to the host apparatus,

the first processor among the plurality of processors

is configured to perform a reboot process that includes a reboot of the first address translation unit connected to the first processor, and

while the first processor is performing the reboot process, a second processor that is different from the first processor of the plurality of processors is configured to substitute for the first processor by reading a first request addressed to the first processor from the host apparatus written by the first protocol chip from the first shared memory through a second address translation unit connected to the second processor, and writing a first response to the first request to the first shared memory through the second address translation unit.

2. The storage system of claim 1 , wherein

the first processor performs the reboot process of the first address translation unit by transmitting a reset signal to the first address translation unit.

3. The storage system of claim 1 , further comprising:

a first storage controller and a second storage controller, wherein

the first storage controller includes the first protocol chip, the first shared memory, the first processor, the second processor, the first address translation unit, and the second address translation unit; and

the second storage controller includes a second protocol chip configured to control a protocol used for communication with the host apparatus, a third plurality of processors, a second shared memory that can be read and written by the second protocol chip and the third plurality of processors, and a third plurality of address translation units each of which connects a corresponding processor of the third plurality of processors and the second shared memory, wherein

the third plurality of address translation units translate addresses used by the respective corresponding processors of the third plurality of processors into addresses used to read from or write to the second shared memory.

4. The storage system of claim 3 , further comprising:

a plurality of storage nodes capable of communicating with each other, wherein

each of the plurality of storage nodes includes the first storage controller and the second storage controller.

5. The storage system of claim 1 , wherein

the first protocol chip, the first shared memory, and the plurality of address translation units are connected via one or a plurality of switch units.

6. The storage system of claim 1 , wherein

the plurality of processors collectively configured to monitor whether another processor reboots,

the second processor detects a reboot of the first processor, and

the second processor notifies all the processors other than the first processor and the second processor that the second processor will substitute for the first processor.

7. The storage system of claim 1 , further comprising:

a plurality of operation monitoring units, wherein

the plurality of operation monitoring units are collectively configured to monitor a reboot of the plurality of processors,

upon detection of a reboot of a particular processor among the plurality of processors, one of the plurality of operation monitoring units determines, from among other processors that have not rebooted, a respective processor that will read the request from the host apparatus substituting for the particular processor that has rebooted, requests arbitration to all the other operation monitoring units of the plurality of operation monitoring units for determination of the respective processor that will read the request from the host apparatus from the first shared memory substituting for the particular processor that has rebooted, receives acknowledgement of the request for the arbitration for the determination of the respective processor that will read the request from the host apparatus from the first shared memory substituting for the particular processor that has rebooted from all the other operation monitoring units of the plurality of operation monitoring units, and instructs the respective processor to read the request from the host apparatus from the first shared memory substituting for the particular processor that has rebooted.

8. The storage system of claim 1 , wherein

the first shared memory includes a request area for storing a request from the host apparatus and a response area for storing a response to the request for each combination of the first protocol chip and one of the plurality of processors,

the first protocol chip writes the request from the host apparatus to the request area of the corresponding processor,

each of the plurality of processors reads the request from the request area corresponding to each processor,

each of the plurality of processors writes a response to the request to the response area corresponding to each processor,

the first protocol chip reads the response from the response area, and,

while the first processor performs the reboot, the second processor is further configured to read the first request addressed to the first processor from the request area corresponding to the first processor, and write the first response to the first request to the response area corresponding to the first processor.

Assignments (2)
CHANGE OF NAME Recorded Sep 27, 2024
From: HITACHI, LTD.
To: HITACHI VANTARA, LTD.
Reel/Frame 069067/0529 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 9, 2023
From: SHIMADA, KENTARO; YOKOI, NOBUHIRO; TSURUYA, MASAHIRO
To: HITACHI, LTD.
Reel/Frame 062938/0275 →
Priority Claims (1)
JP 2022-128905 · Aug 12, 2022 · national
Continuity (1)
Related Publication 20240054076A1 · Feb 15, 2024
References Cited (3)
US 8423677B2 · Nakamura et al. · 2013 [cited by applicant]
US 8700856B2 · Tanaka et al. · 2014 [cited by applicant]
US 20200133764A1 · Li · 2020 [cited by examiner]