Computer system, computer system management method, and program

Patent metadata

Field	Value
Publication number	US-9892183-B2
Application number	US-201314426996-A
Country	US
Kind code	B2
Filing date	Apr 16, 2013
Priority date	Apr 16, 2013
Publication date	Feb 13, 2018
Grant date	Feb 13, 2018

Abstract

A computer system, comprising a plurality of computers coupled to one another via a network, the computer system being configured to execute a task using a database constructed from the plurality of computers, the plurality of computers including a first computer and a second computer, the second computer being configured to execute: data replication for receiving the recovery request, for reading data, and for transmitting the read data to the first computer as first replicated data; and update processing for determining, in a case where a command to update the data is received in a recovery state, the sequence number of the update command, for updating predetermined data, and for transmitting the updated predetermined data as second replicated data, the first computer or the second computer being configured to control a write order of the first replicated data and the second replicated data by the first computer.

First claim

What is claimed is:

1. A computer system, comprising:
a plurality of computers coupled to one another via a network, the computer system being configured to execute a business task using a database constructed from storage areas included in each of the plurality of computers,
the database being configured to store a plurality of pieces of data, each data includes information identifying the data, a value of the data, and a sequence number indicating an execution order of an event in the database,
the plurality of pieces of data being allocated in a distributed manner across the each of the plurality of computers for each management area determined by applying a partitioning algorithm for a distributed data store to the information identifying the data,
the each of the plurality of computers including:
a data management module configured to manage the allocated data;
a data control module configured to determine the sequence number of an operation on the allocated data; and
a recovery control module configured to transmit replicated data of data included in a predetermined management area to a newly added computer,
the plurality of computers including a first computer configured to transmit a recovery request and a second computer configured to receive the recovery request,
the second computer being configured to execute:
data replication upon receiving the recovery request from the first computer, including: shifting a state of the second computer to a recovery state, reading at least one piece of data from the database based on the sequence numbers of the data stored in the second computer, and for transmitting the read at least one piece of data to the first computer as first replicated data; and
update processing for determining, in a case where an update command to update predetermined data stored in the second computer is received by the second computer while in the recovery state, a sequence number of the update command that is greater than a latest sequence number, for updating predetermined data based on the update command, and for transmitting the updated predetermined data as second replicated data,
wherein the each of the plurality of computers respectively hold histories associating the sequence numbers and content of an event on the database with each other,
wherein the recovery request includes information indicating a management range of a processing target,
wherein in the data replication,
the recovery control module of the second computer is configured to:
hold the latest sequence number as a replication sequence number based on the histories in a case of receiving the recovery request; and
read from the plurality of pieces of data included in the management range of the processing target at least one piece of data that is not transmitted and includes the sequence number older than the replication sequence number, and transmit the read at least one piece of data to the recovery control module of the first computer as the first replicated data,
wherein in the update processing,
the data management module of the second computer updates the predetermined data based on the update command, and outputs to the recovery control module of the second computer an instruction to transmit the first replicated data, and
the recovery control module of the second computer transmits the first replicated data to the recovery control module of the first computer based on the transmission instruction,
wherein at least one of the first computer and the second computer being configured to control a write order indicating an order of writing the first replicated data and the second replicated data by the first computer, and
wherein the first computer being configured to execute write processing for writing, based on the write order, the first replicated data and the second replicated data to the storage areas constructing the database.

2. The computer system according to claim 1 ,
wherein in the data replication,
the recovery control module of the second computer is configured to:
obtain an exclusive lock of the management range of the processing target;
release the obtained exclusive lock after reading the at least one piece of data to the recovery control module of the first computer as the first replicated data,
wherein in the update processing,
the data management module of the second computer is configured to:
obtain the exclusive lock of the management range of the processing target;
and
release the obtained exclusive lock after updating the predetermined data based on the update command, and output to the recovery control module of the second computer an instruction to transmit the first replicated data, and
wherein in the write processing, the recovery control module of the first computer writes, based on the received order, the first replicated data and the second replicated data to the storage areas constructing the database.

3. The computer system according to claim 1 ,
wherein the each of the plurality of computers includes an operation storage area for temporarily storing the first replicated data and the second replicated data,
wherein in the write processing,
the recovery control module of the first computer is configured to:
store the received first replicated data and the received second replicated data in the operation storage area;
retrieve from the database and the operation storage area the first replicated data and the second replicated data including information identifying the same data;
refer to the sequence number included in the retrieved first replicated data and the sequence number included in the retrieved second replicated data, and select the replicated data including the newest sequence number; and
write the selected replicated data to the storage areas constructing the database.

4. The computer system according to claim 3 ,
wherein the data stored in the database further includes a deletion flag indicating whether the data is deletion target data,
wherein in the update processing, the data management module of the second computer transmits deletion target data set the deletion flag as the first replicated data to the recovery control module of the first computer in a case where the update command is a command to delete data, and
wherein in the write processing, the data management module of the first computer deletes, of the plurality of pieces of data written to the database, data set the deletion flag from the storage areas constructing the database.

5. The computer system according to claim 2 ,
wherein the each of the plurality of computers holds configuration information indicating the management range to be managed by the each of the plurality of computers as a master and the management range to be managed by the each of the plurality of computers as a slave,
wherein in the data replication,
the recovery control module of the second computer instructs, after the first replicated data including the replication sequence number has been transmitted, the data control module of the second computer to determine a participation sequence number as the sequence number of a participation event for adding the first computer to the plurality of computers, and
the data control module of the second computer is configured to:
determine, based on a distributed consensus algorithm, the participation sequence number by communicating to and from the data control module of the each of the plurality of the computers;
hold a history associating the determined participation sequence number and content of the participation event with each other; and
output the determined participation sequence number to the recovery control module of the second computer,
wherein in the update processing,
the data management module of the second computer is

Assignees

• Hitachi Ltd

Inventors

• Hasegawa Toshiyuki
• Yamamoto Nobuyuki

Classifications

• G06F17/30864

  Physics · mapped topic

• G06F17/30362

  Physics · mapped topic

• G06F17/30309

  Physics · mapped topic

• G06F17/30117

  Physics · mapped topic

• G06F17/30581 Primary

  Physics · mapped topic