JP2014115929A - Replication system, replication processing method and replication processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a replication system which does not cause processing delay even in the event of a defect in one storage part out of a plurality of storage parts.SOLUTION: The replication system 1 includes: a plurality of storage parts 4 storing same data; a storage part control part 2 which, in receiving a request for update of the data in the plurality of storage parts 4 from the outside, selects one storage part out of the plurality of storage parts 4 in accordance with a prescribed rule, updates the data in the selected storage part, and issues a replication command; and a replication part 3 performing replication for updating the data in the remaining storage parts, other than the selected storage part, out of the plurality of storage parts 4, which correspond to the updated data in the selected storage part, in response to the replication command.

Description

  The present invention relates to a replication system, a replication processing method, and a replication processing program, and more particularly, to a replication system, a replication processing method, and a replication processing program that are less affected by a failure in a storage unit.

  In the computer-related field, replication means replication by providing redundancy to hardware such as a storage unit (hereinafter also referred to as “disk”, “volume”, or “storage medium”) that constitutes a computer system. A generic term for technologies that improve the reliability of a computer system.

  Replication is roughly divided into data replication in which one volume is shared by multiple volumes, and calculation replication in which one calculation is executed by multiple hardware. In the following explanation, replication refers to data replication. Shall be pointed to.

  FIG. 3 is a block diagram showing a configuration of a general replication system 15. Referring to FIG. 3, the replication system 15 is connected to an information processing apparatus 18 such as a server, and includes a master storage unit 17-1, a replication storage unit 17-2, and a storage unit control unit 16. The replication storage unit 17-2 stores a copy (copy) stored in the master storage unit 17-1.

  For example, the storage unit control unit 16 updates the corresponding data in the master storage unit 17-1 or adds new data to the master storage unit 17-1 in accordance with the data update or data registration instruction from the information processing device 18. Register data. Thereafter, the storage unit control unit 16 performs the same data update or registration process on the replication storage unit 17-2 so as to reflect the data update or registration in the master storage unit 17-1.

  The information processing apparatus 18 is configured to perform data update or registration processing only on the master storage unit 17-1, and cannot perform these processing on the replication storage unit 17-2. .

  Patent Document 1 discloses a technique related to replication. Specifically, Patent Document 1 discloses a copy processing technique that does not lose data in data transfer between a plurality of sites separated by a long distance when a disaster occurs. In this copy processing technique disclosed in Patent Document 1, when data is copied from a production site to a remote site, the data is also copied to the local site.

JP 2006-072635 A

  However, in the replication system 15 shown in FIG. 3, when a failure occurs in the master storage unit 17-1, the storage unit control unit 16 notifies the information processing device 18 to that effect, and as a result, the information processing device 18 Data update processing is interrupted. Thereafter, when the failure in the master storage unit 17-1 is resolved, the storage unit control unit 16 copies the data in the replication storage unit 17-2 to the master storage unit 17-1. Next, the storage unit control unit 16 notifies the information processing apparatus 18 that the failure has been resolved, and the information processing apparatus 18 resumes the data update process.

  As described above, in the replication system 15 shown in FIG. 3, when a failure occurs in the master storage unit 17-1, the information processing apparatus 18 must interrupt the data update process, resulting in a processing delay of the entire system. There is a problem that cannot be avoided.

  In the system described in Patent Document 1, data is copied to a plurality of sites in preparation for the loss of data due to a disaster or the like. Nothing is disclosed about the processing delay.

  An object of the present invention is to provide a replication system, a replication processing method, and a replication processing program that solve the above-described problems and do not cause adverse effects such as processing delay due to a failure in a storage unit.

The replication system of the present invention
A plurality of storage units for storing the same data;
When a request for updating data in the plurality of storage units is received from outside, one storage unit is selected from the plurality of storage units according to a predetermined rule, and the data in the selected storage unit is updated. And a storage control unit that issues a first replication instruction;
In response to the first replication instruction, data in each of the remaining storage units excluding the selected storage unit among the plurality of storage units, the updated data in the selected storage unit And a replication unit that executes replication to update the data corresponding to.

Further, the replication processing method of the present invention includes:
Used in replication systems with multiple storage units that store the same data,
When a request for updating data in the plurality of storage units is received from the outside, according to a predetermined rule, one storage unit is selected from the plurality of storage units, and the data in the selected storage unit is updated. And an update step for issuing a first replication instruction;
In response to the first replication instruction, data in each of the remaining storage units excluding the selected storage unit among the plurality of storage units, the updated data in the selected storage unit And a replication step for performing replication to update data corresponding to the.

Furthermore, the replication processing program of the present invention provides:
In a computer having a plurality of storage units for storing the same data,
When a request for updating data in the plurality of storage units is received from the outside, according to a predetermined rule, one storage unit is selected from the plurality of storage units, and the data in the selected storage unit is updated. And an update process for issuing a first replication instruction;
In response to the first replication instruction, data in each of the remaining storage units excluding the selected storage unit among the plurality of storage units, the updated data in the selected storage unit Replication processing that executes replication to update data corresponding to the.

  In the present invention, even when a failure occurs in a storage unit among a plurality of storage units, it is not necessary to interrupt the update process of data requested by the information processing apparatus, so that a processing delay is not caused. There is an effect.

It is a block diagram which shows the structure of the 1st Embodiment of this invention. It is a block diagram which shows the structure of the 2nd Embodiment of this invention. It is a block diagram which shows the structure of the general replication system relevant to this invention.

  Next, embodiments of the present invention will be described in detail with reference to the drawings.

<First Embodiment>
FIG. 1 is a block diagram showing the configuration of the first exemplary embodiment of the present invention.

  Referring to FIG. 1, the present embodiment includes a replication system 1, a plurality of storage units 4, and an information processing device 5.

  The replication system 1 stores the same data as the storage unit control unit 2 that controls the exchange of data between itself and the information processing apparatus 5 (referred to as “data access” in FIG. 1). A plurality of storage units 4.

  The storage unit control unit 2 includes a replication unit 3 that processes replication.

  When the information processing device 5 updates the data in the storage unit 4, the storage unit control unit 2 first selects one of the plurality of storage units 4 according to a predetermined rule. The data in the storage unit 4 is updated first. At this point, the information processing apparatus 5 recognizes that the data update has been completed.

  Here, as the predetermined rule, for example, there is a method of selecting one storage unit at random, a method of selecting one storage unit 4 having the largest processing capacity according to the processing capacity of each storage unit 4 and the like. . Needless to say, the predetermined rule is not limited to these methods.

  When the data in the first storage unit 4 is updated in this way, the storage unit control unit 2 instructs the replication unit 3 to perform replication in which the corresponding data in the remaining storage units 4 is similarly updated.

  The replication unit 3 receives the above-described instruction and updates the data in the remaining storage unit 4.

  Next, processing when a failure occurs in a certain storage unit 4 will be described.

  Now, assume that a failure occurs in a certain storage unit 4. In this case, the storage unit control unit 2 issues an instruction to the replication unit 3 so as to stop the replication for the storage unit in which the failure has occurred (hereinafter referred to as “failure storage unit”).

  The replication unit 3 then stops replication related to the failure storage unit according to the instruction.

  In such a case, when an update request for data in the storage unit 4 is received from the information processing device 5, the storage unit control unit 2 first stores a fault memory from among the plurality of storage units 4 according to the predetermined rule. One storage unit 4 other than the storage unit 4 is selected, and the data in the selected storage unit 4 is updated first.

  Thereafter, the storage unit control unit 2 instructs the replication unit 3 to perform replication in which the corresponding data in the remaining storage unit 4 is similarly updated. In response to this, the replication unit 3 similarly updates the corresponding data in the storage unit 4 other than the failure storage unit, and does not perform replication on the failure storage unit.

  As described above, according to the present embodiment, even when a failure occurs in a certain storage unit 4, there is no need to interrupt the data update process requested from the information processing apparatus 5, so that no processing delay is caused. There is.

  The reason is that the storage control unit 2 selects one storage unit 4 excluding the failure storage unit, performs the requested update on the data in the selected storage unit 4, and the replication unit 3 This is because the update replication is performed on the storage unit 4 other than the failure storage unit.

<Second Embodiment>
Next, a second embodiment based on the above-described first embodiment will be described.

  In the following, characteristic portions according to the present embodiment will be mainly described. In the present embodiment, the same reference numerals are given to the same components as those of the first embodiment described above. The overlapping description of these components will be omitted.

  FIG. 2 is a block diagram showing the configuration of the second exemplary embodiment of the present invention.

  Referring to FIG. 2, the present embodiment includes a replication system 6, a plurality of storage units 4, and an information processing device 5.

  The replication system 6 further includes a measurement unit 8 that measures the processing load in the plurality of storage units 4, and a storage unit control that controls data exchange between the information processing device 5 and the storage unit 4 in accordance with the processing load. Part 7.

  The measuring unit 8 measures the processing load of each of the plurality of storage units 4.

  In this embodiment, when an update request for data in the storage unit 4 is received from the information processing device 5, the storage unit control unit 7 selects the storage unit 4 with the lowest processing load based on the processing load measured by the measurement unit 8. The selected data in the storage unit 4 is updated first.

  Thereafter, the storage unit control unit 2 instructs the replication unit 3 to perform replication in which the corresponding data in the remaining storage unit 4 is similarly updated.

  In addition, since the process when a failure occurs in a certain storage unit 4 is the same as that in the first embodiment, the description thereof is omitted.

  As described above, the present embodiment has the same effects as those of the first embodiment and the effect that the data update process is completed early.

  The reason is that, in response to the data update request from the information processing device 5, the data in the storage unit or the like having the lowest processing load is updated, and at that time, the information processing device 5 has completed the update of the requested data. This is because, thereafter, replication is performed in which the corresponding data in the remaining storage unit 4 is similarly updated.

  Note that the replication systems in the first and second embodiments may be realized by a computer system that operates under program control, or may be realized by dedicated hardware.

DESCRIPTION OF SYMBOLS 1 Replication system 2 Storage part control part 3 Replication part 4 Storage part 5 Information processing apparatus 6 Replication system 7 Storage part control part 8 Measurement part 15 Replication system 16 Storage part control part 17-1 Master storage part 17-2 Replication storage part 18 Information processing device

Claims (9)

A plurality of storage units for storing the same data;
When a request for updating data in the plurality of storage units is received from the outside, according to a predetermined rule, one storage unit is selected from the plurality of storage units, and the data in the selected storage unit is updated. And a storage control unit that issues a first replication instruction;
In response to the first replication instruction, data in each of the remaining storage units excluding the selected storage unit among the plurality of storage units, the updated data in the selected storage unit And a replication unit that executes replication for updating data corresponding to the replication system. The storage unit control unit
When a failure occurs in a storage unit of the plurality of storage units, when a request for updating the data in the storage unit is received from the outside, the failure of the plurality of storage units according to a predetermined rule A second storage unit that selects one storage unit excluding the fault storage unit that is the storage unit in which the error occurred, updates the data in the selected storage unit, and stops replication for the fault storage unit. Issue replication instructions for
The replication unit
In response to the second replication instruction, data in each of the remaining storage units excluding the selected storage unit and the failure storage unit among the plurality of storage units, and in the selected storage unit The replication system according to claim 1, wherein replication is performed to update data corresponding to the updated data. A measuring unit that measures a processing load in each of the plurality of storage units;
The replication system according to claim 1 or 2, wherein a storage unit having the lowest processing load is selected as a storage unit selected according to the predetermined rule. In a replication system comprising a plurality of storage units for storing the same data,
When a request for updating data in the plurality of storage units is received from the outside, according to a predetermined rule, one storage unit is selected from the plurality of storage units, and the data in the selected storage unit is updated. And an update step for issuing a first replication instruction;
In response to the first replication instruction, data in each of the remaining storage units excluding the selected storage unit among the plurality of storage units, the updated data in the selected storage unit And a replication step for executing replication for updating data corresponding to the replication processing method. When a failure occurs in a storage unit of the plurality of storage units, when a request for updating the data in the storage unit is received from the outside, the failure of the plurality of storage units according to a predetermined rule A second storage unit that selects one storage unit excluding the fault storage unit that is the storage unit in which the error occurred, updates the data in the selected storage unit, and stops replication for the fault storage unit. An update step in the event of a failure to issue a replication instruction,
In response to the second replication instruction, data in each of the remaining storage units excluding the selected storage unit and the failure storage unit among the plurality of storage units, and in the selected storage unit The replication processing method according to claim 4, further comprising: a replication step in the event of a failure that executes replication for updating data corresponding to the updated data. A measurement step of measuring a processing load in each of the plurality of storage devices;
The replication processing method according to claim 4, further comprising: a selecting step of selecting a storage unit having the lowest processing load as the storage unit selected according to the predetermined rule. When a computer having a plurality of storage units that store the same data receives a request for updating the data in the plurality of storage units from the outside, one storage unit of the plurality of storage units is set according to a predetermined rule. Selecting and updating the data in the selected storage unit and issuing a first replication instruction;
In response to the first replication instruction, data in each of the remaining storage units excluding the selected storage unit among the plurality of storage units, the updated data in the selected storage unit A replication processing program that executes replication processing for executing replication to update data corresponding to the. When a failure occurs in a storage unit of the plurality of storage units, when a request for updating the data in the storage unit is received from the outside, the failure of the plurality of storage units according to a predetermined rule A second storage unit that selects one storage unit excluding the fault storage unit that is the storage unit in which the error occurred, updates the data in the selected storage unit, and stops replication for the fault storage unit. Update process at the time of issuing a replication instruction,
In response to the second replication instruction, data in each of the remaining storage units excluding the selected storage unit and the failure storage unit among the plurality of storage units, and in the selected storage unit 8. The replication processing program according to claim 7, further comprising: causing the computer to execute a replication processing at the time of executing replication for updating data corresponding to the updated data. A measurement process for measuring a processing load in each of the plurality of storage units;
9. The replication processing program according to claim 7, further comprising: causing the computer to execute selection processing for selecting a storage unit having the lowest processing load as a storage unit selected according to the predetermined rule.

Images