WO2016208014A1

WO2016208014A1 - Management computer and method for switching system configuration

Info

Publication number: WO2016208014A1
Application number: PCT/JP2015/068261
Authority: WO
Inventors: 淳平清時; 義之田畑; 竹内　伸也
Original assignee: 株式会社日立製作所
Priority date: 2015-06-24
Filing date: 2015-06-24
Publication date: 2016-12-29

Abstract

[Problem] To switch remote paths when one of a plurality of remote paths becomes highly loaded in a system configuration into which data is written via a remote path between storage devices. [Solution] When there is a data write request from a host computer, a primary storage device writes data to the primary volume of the primary storage device, and then transfers data via a first remote path and writes data to the secondary volume of a secondary storage device. When there is a data write request from the host computer, the secondary storage device writes, to the primary volume, data transferred via a second remote path from the secondary storage device, and then transfers data via the second remote path and writes to the secondary volume. A management computer selects a storage device to which data from the host computer is written so that the amounts of data transfer of the first remote path and second remote path are equalized.

Description

Management computer and system configuration switching method

The present invention relates to a management computer and a system configuration switching method, and is suitably applied to a management computer and a system configuration switching method that automatically switch a remote path according to a load status of the remote path.

Conventionally, in a computer system having a host device and a storage system, an environment for recognizing the environment of the host device and enabling access to a logical volume has been automatically constructed (for example, Patent Document 1). Also, in a computer system that has an executing host device and a standby host device and has fault tolerance, if a failure occurs in one system, the system is switched to another system that was waiting and continues processing. (For example, Patent Document 2).

In addition, recently, a technology for making data redundant in units of volumes between two primary and secondary storage devices has been disclosed. With this technology, the host device can be switched from the primary system (execution system) to the secondary system (standby system) without changing the correspondence between the host apparatus and the logical volume. Specifically, even when the host device is switched from the primary system to the secondary system, the data is first transferred to the primary volume via the remote path from the secondary storage system connected to the secondary host to the primary storage system. After that, the data is written to the secondary volume forming a copy pair with the primary volume. This makes it possible to switch the host path from the primary system to the secondary system while maintaining the correspondence between the host apparatus and the logical volume without changing the volume configuration or setting.

Japanese Patent No. 5020601 JP 2006-189963 A

However, with the technology described above, the remote path is automatically switched when a failure occurs in a certain system, but the remote path is automatically switched even if one of the multiple remote paths becomes heavily loaded. There was a problem that resources could not be used effectively.

The present invention has been made in consideration of the above points. In a system configuration in which data is written via a remote path between storage apparatuses, when any one of a plurality of remote paths becomes highly loaded, it is automatically performed. A management computer capable of switching a remote path and a system configuration switching method are proposed.

In order to solve this problem, in the present invention, a management computer that manages the performance of a storage system that makes data requested to be written by a host computer redundant in units of volumes in a primary storage device and a secondary storage device The management computer refers to the data transfer amount, a memory for storing a data transfer amount of data transmitted / received between the host computer and the primary storage device or the secondary storage device. A CPU and a network interface connected to the host computer, the primary storage device or the secondary storage device, and the primary storage device receives a data write request from the host computer, After writing data to the primary volume of the primary storage device, When data is transferred via a remote path and data is written to the secondary volume of the secondary storage device, the secondary storage device receives the data write request from the host computer. After the data transferred from the device via the second remote path is written to the primary volume, the data is transferred via the second remote path and written to the secondary volume, and the management computer When the data transfer amounts of the first remote path and the second remote path are acquired, and the data transfer amount exceeds a predetermined threshold, the first remote path and the second remote path A storage device to which data is written from the host computer is selected so that the data transfer amount is leveled. Computer is provided.

In order to solve this problem, in the present invention, a management computer that manages the performance of a storage system that makes data requested to be written by a host computer redundant in units of volumes in a primary storage device and a secondary storage device The management computer includes a memory for storing a data transfer amount of data transmitted / received between the host computer and the primary storage device or the secondary storage device, and the data A CPU that refers to the transfer amount; and a network interface that is connected to the host computer, the primary storage device, or the secondary storage device, and the host computer issues a data write request to the primary storage device. If there is, the primary volume of the primary storage device After the data is written, the step of transferring the data via the first remote path and writing the data to the secondary volume of the secondary storage device, and the step of writing the data from the host computer to the secondary storage device When there is a write request, after the data transferred from the secondary storage device via the second remote path is written to the primary volume, the data is transferred via the second remote path. Writing to the secondary volume, the management computer obtaining the data transfer amount of the first remote path and the second remote path, and the management computer determining that the data transfer amount is a predetermined threshold value. So that the data transfer amount of the first remote path and the second remote path is leveled, Characterized in that it comprises the steps of selecting a data write destination storage device from strike computer, the system configuration switching method is provided.

According to the present invention, in a system configuration in which data is written via a remote path between storage devices, when one of a plurality of remote paths becomes a heavy load, the remote path can be automatically switched and used. Effective resource utilization.

It is a conceptual diagram explaining the outline | summary which concerns on one Embodiment of this invention. It is the schematic explaining the structure of the computer system concerning the embodiment. It is a conceptual diagram which shows the structure of the database managed by the management server concerning the embodiment. It is a chart showing an example of port data transfer amount time-series performance data information according to the embodiment. It is a chart showing an example of host information concerning the embodiment. It is a chart which shows an example of the storage information concerning the embodiment. It is a chart showing an example of copy pair information according to the embodiment. It is a chart which shows an example of the remote path information concerning the embodiment. It is a chart which shows an example of the volume information concerning the embodiment. It is a chart showing an example of port information concerning the embodiment. It is a conceptual diagram explaining the system configuration | structure switching process concerning the embodiment. It is a conceptual diagram explaining the system configuration | structure switching process concerning the embodiment. It is a flowchart which shows the detail of the system configuration | structure switching process concerning the embodiment. It is a flowchart which shows the detail of the system configuration | structure switching process concerning the embodiment. It is a flowchart which shows the detail of the system configuration | structure switching process concerning the embodiment. It is a flowchart which shows the detail of the system configuration | structure switching process concerning the embodiment. It is a conceptual diagram which shows the example of a display of the performance data concerning the embodiment.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

In the following description, the information of the present invention will be described using expressions such as “aaa table”, “aaa list”, “aaaDB”, “aaa queue”, etc., but these information are not necessarily limited to tables, lists, DBs, queues, It may be expressed by other than the data structure. Therefore, “aaa table”, “aaa list”, “aaaDB”, “aaa queue”, etc. may be referred to as “aaa information” to indicate that they are not dependent on the data structure.

Furthermore, in describing the contents of each information, the expressions “identification information”, “identifier”, “name”, “name”, and “ID” are used, but these can be replaced with each other.

In the following description, there is a case where “program” is used as the subject. However, since the program performs processing determined by being executed by the processor using the memory and the communication port (communication control device), the processor is used as the subject. The explanation may be as follows. Further, the processing disclosed with the program as the subject may be processing performed by a computer such as a management server or an information processing apparatus. Further, part or all of the program may be realized by dedicated hardware.

Various programs may be installed in each computer by a program distribution server or a storage medium that can be read by the computer. In this case, the program distribution server includes a CPU and storage resources, and the storage resources further store a distribution program and a program to be distributed. When the distribution program is executed by the CPU, the CPU of the program distribution server distributes the distribution target program to other computers.

(1) Outline of the present embodiment First, the outline of the present embodiment will be described. As described above, in the technology for making data redundant between the primary and secondary storage devices in units of volumes (hereinafter, this technology is referred to as a global-active device or gad), the host device and the logical volume are Standby without waiting for the host system to take over work from the running system (hereinafter also referred to as the primary host system) that is executing the business, without changing the correspondence, in the event of a failure of the primary host system It is possible to switch to a system (hereinafter also referred to as a subordinate host device).

Normally, in global-active device, remote path is automatically switched when a failure occurs. If there is a failure in the primary host device, the master host device is switched to the secondary host device, and the synchronous remote copy described above is executed. That is, when the primary host device is switched to the secondary host device, the I / O from the secondary host device is transferred from the secondary storage device connected to the secondary host device to the primary storage device. The data is first written to the primary volume via the remote path, and then the data is written to the secondary volume that forms a copy pair with the primary volume. To achieve this, the remote path requires two directions, upstream and downstream, and I / O from the host device is performed from the primary volume chassis or from the secondary volume chassis. The direction of the remote path used differs depending on the

In addition, when the host device is configured with a plurality of clusters, a plurality of remote paths for performing data transfer for remote copy between the primary storage device and the secondary storage device are configured. As described above, the remote path is normally switched only when a failure occurs in a certain system. However, when one of the multiple remote paths becomes heavily loaded, the remote path is automatically set. It was not configured to switch, and resources could not be used efficiently. Therefore, in this embodiment, by effectively switching a plurality of remote paths, it is possible to effectively use resources by distributing the load of remote path performance while maintaining the switching function when a failure occurs. Yes.

Specifically, as shown in FIG. 1, in the computer system according to the present embodiment, a plurality of host devices (hereinafter referred to as business hosts that execute a predetermined business) constitute a cluster. ing. For example, in FIG. 1, the business hosts 200A1 and 200B1, the business hosts 200A2 and 200B2, and the business hosts 200A3 and 200B3 constitute a cluster. For example, when the business host 200A1 or 200A2 becomes the primary system and executes (active) the business, the business host 200B1 or B2 is a secondary business host that does not execute the business, and the primary host device becomes a failure The system is on standby (active standby) so that it can take over business. When the business host 200A3 is stopped due to a failure or the like, the business host 200B3 becomes active.

In the following, a case where the business host configures a cluster will be described, but the present embodiment is not limited to such an example. For example, in this embodiment, the primary volume and the secondary volume are connected to one business host, and I / O from the business host is switched from the primary volume housing to the secondary volume housing. May be applied.

Referring to FIG. 1, a description will be given of remote path switching for optimizing the remote path performance in the present embodiment. The business host 200 and the storage system 300 that manages a plurality of volumes are connected to the management server 100 that manages the configuration of the computer system. The hardware configuration and functional configuration of the computer system will be described later in detail. The management server 100 is an example of the management computer of the present invention.

As shown in FIG. 1, first, the management server 100 detects a remote path in which the port data transfer amount reaches the maximum value of the network bandwidth (STEP 01). Here, the port data transfer amount is a data transfer amount of data transferred from a port of one storage system 300 to a port of another storage system 300 via a remote path when the above-described synchronous copy is performed. .

Subsequently, the management server 100 acquires the GAD copy pair information of the volume using the remote path detected in STEP 01 (STEP 02). Then, the management server 100 requests switching that causes the business host using the copy pair acquired in STEP 02 to determine the master-slave relationship, and acquires the port data transfer amount of the remote path after switching (STEP 03).

Then, based on the port data transfer amount acquired in STEP 03, an optimized combination cluster configuration is selected and the cluster configuration is maintained (STEP 04).

As described above, the global-active device can switch the host path from the primary system to the secondary system while maintaining the correspondence between the host device and the logical volume without changing the volume configuration or changing the settings. Therefore, it is possible to easily switch the remote path shown in STEP03. Therefore, in this embodiment, the remote path is switched in STEP 03, and the remote path performance is determined to be appropriate in STEP 04, thereby switching to a cluster configuration that provides the optimum remote path performance while continuing the business operation. Is possible.

(2) Configuration of Computer System Next, the configuration of the computer system will be described with reference to FIG. As shown in FIG. 2, the computer system includes a management server 100, business hosts 200A1 to 200A3 (hereinafter also referred to as business host 200A), business hosts 200B1 to 200B3 (hereinafter referred to as business host 200B). The storage system 300A and the storage system 300B. Hereinafter, the business host 200A and the business host 200B may be collectively referred to as the business host 200A, and the storage system 300A and the storage system 300B may be collectively referred to as the storage system 300.

(2-1) Management Server Configuration The management server 100 is mainly connected to the business host 200A, the business host 200B, the storage system 300A, and the storage system 300B to collect configuration information of each device, and the business hosts 200A and 200B. Or the cluster configuration of the business hosts 200A and 200B is changed.

(2-1-1) Hardware Configuration The management server 100 mainly includes a CPU 101, a memory 102, an HDD 103, and an HBA 104.

The CPU 101 functions as an arithmetic processing unit, and controls the operation of the management server 100 according to programs, arithmetic parameters, and the like stored in the memory 102. An HDD (Hard Disk Drive) 103 includes a storage medium, a recording device that records data on the storage medium, a reading device that reads data from the storage medium, a deletion device that deletes data recorded on the storage medium, and the like, and drives a hard disk Then, the program executed by the CPU 101 and various data are stored.

The HBA (Host Bus Adapter) 104 is a communication interface configured by a communication device or the like for connecting to the network, and is connected to the business host 200 or the storage system 300 to transmit / receive data.

The management server 100 also has an input / output device. Examples of input / output devices include a display, a keyboard, and a pointer device, but other devices may be used. Also, as an alternative to the input / output device, a serial interface or an Ethernet interface is used as the input / output device, a display computer having a display or a keyboard or a pointer device is connected to the interface, and the display information is transmitted to the display computer. By receiving the input information from the display computer, the display computer may perform the display, or the input may be replaced by the input / output device by receiving the input.

Hereinafter, a set of one or more computers that manage the storage system and display the display information of the present invention may be referred to as a management system. When the management server 100 displays display information, the management server 100 is a management system, and a combination of the management server 100 and a display computer is also a management system. Further, in order to increase the speed and reliability of management processing, a plurality of computers may realize processing equivalent to the management server 100. In this case, the plurality of computers (in the case where the display computer performs display, display is performed). Management computer) is the management system.

(2-1-2) Functional Configuration The management server 100 mainly executes a host configuration change command program 110, a leveled value calculation program 111, a cluster configuration host device failure detection program 112, and port data transfer as programs executed by the CPU 101. This includes a time-series performance data collection program 113, a host / storage device configuration information collection program 114, a cluster configuration switching program 117, and the like. The HDD 103 stores a database such as port data transfer amount time-series performance data 115 and host / storage device configuration information 116.

The host configuration change command program 110 sends a command to switch the cluster configuration of the business host 200 to the target business host 200. The leveling value calculation program 111 calculates the leveling value of the remote path from the network bandwidth usage rate. Here, the network bandwidth usage rate is a value obtained by dividing the port data transfer amount by the network bandwidth and multiplying by 100, and what percentage of the network bandwidth, which is the maximum value of the transmission capacity, is used as the data transfer amount. It is a value indicating whether or not.

In this embodiment, an absolute value obtained by subtracting 50% from the network usage rate is calculated for each remote path, and the sum of these values is used as the leveling value. The performance of the remote path is not a problem when the port data transfer amount does not occupy a large range with respect to the network bandwidth, that is, when the network usage rate is large. Therefore, it is necessary to mitigate remote paths with a high network usage rate rather than a low network usage rate. Therefore, in the leveling in this embodiment, by selecting a combination in which the network usage rate of the remote path is close to 50%, the port data transfer amount of the remote path with a high network usage rate is suppressed, and each remote path Performance is leveled. The calculation of the leveling value will be described in detail later.

The cluster configuration host device failure detection program 112 collects cluster configuration failure information from the business host 200. The port data transfer amount time series performance data collection program 113 collects the port data transfer amount from the port data transfer amount provision program 310 of the storage system 300. The host / storage device configuration information collection program 114 acquires the configuration information of the business host 200 and the storage system 300 from the business host 200 and the storage system 300, respectively. The cluster configuration switching program 117 determines whether or not the cluster configuration needs to be switched based on the collected port data transfer amount. If it is determined that the cluster configuration switching program 117 is necessary, the cluster configuration switching program 117 performs an instruction switching process for the cluster configuration host configuration change instruction program. Execute.

(2-2) Business Host Configuration The business host 200 writes / reads data to / from the primary volume of the storage system. Further, in response to a cluster configuration change request from the management server 100, the cluster configuration is changed to the primary system or the secondary system.

(2-2-1) Hardware Configuration The business host 200 mainly includes a CPU 201, a memory 202, an HDD 203, and an HBA 204.

The CPU 201 functions as an arithmetic processing unit, and controls the operation of the business host 200 according to programs, arithmetic parameters, and the like stored in the memory 202. An HDD (Hard Disk Drive) 203 includes a storage medium, a recording device that records data on the storage medium, a reading device that reads data from the storage medium, and a deletion device that deletes data recorded on the storage medium, and drives a hard disk The program executed by the CPU 201 and various data are stored.

An HBA (Host Bus Adapter) 204 is a communication interface configured by a communication device or the like for connecting to a network, and transmits / receives data by connecting to the management server 100 or the storage system 300.

(2-2-2) Functional Configuration The business host 200 mainly includes a host configuration information provision program 210 and a host configuration change program 211 as programs executed by the CPU 201.

The host configuration information provision program 210 provides configuration information of the business host 200 to the management server 100 in response to a request from the host / storage device configuration information collection program 114 of the management server 100.

The host configuration change program 211 switches the cluster configuration of the business host 200 in accordance with a command from the host configuration change command program 110 of the management server 100. When the business host 200 is the primary system, the business host 200 switches from the active mode to the active standby mode in accordance with a command from the management server 100 and switches to the secondary business host. Further, when the business host 200 is a secondary system, the business host 200 switches from the active standby mode to the active mode in accordance with a command from the management server 100 and switches to the primary business host.

For example, the business host 200A1 and the business host 200B1 form a cluster, the business host 200A1 is connected to the storage system 300A via the HBA 204 of the business host 200A1, and the business host 200B1 is connected to the storage system via the HBA 204 of the business host 200B1. Assume that it is connected to 300B. When there is a cluster configuration switching command from the host configuration change command program 110 of the management server 100, the business host 200A1 is switched to the business host 200B1 by switching the connection of the HBA 204. That is, the connection between the HBA 204 of the business host 200A1 and the storage system 300A is disconnected, and the storage system 300B is connected via the HBA 204 of the business host 200B1.

(2-2-3) Database Configuration Next, a database configuration managed by the management server 100 will be described with reference to FIGS. As shown in FIG. 3, the port data transfer amount time series performance data 115 of the management server 100 includes port data transfer amount time series performance data information 151. The host / storage apparatus configuration information 116 includes host information 161, storage information 162, copy pair information 163, remote path information 164, volume information 165, and port information 166.

The port data transfer amount time series performance data information 151 is information for managing the data transfer amount for each port in time series. As shown in FIG. 4, the time 1511, the port ID 1512, and the data transfer amount 1513 are associated with each other. Yes. The time 1511 is the time when the data transfer amount is measured, the port ID 1512 is information for identifying the port, and the data transfer amount 1513 is the measured data transfer amount. For example, in FIG. 4, it can be seen that the data transfer amount of the port ID “Port_1” is “40 Mbyte / sec” at the time “T1”.

The host information 161 is information for managing the configuration and operation of the business host 200, and as shown in FIG. 5, a host ID 1611, a cluster configuration host ID 1612, and an operation state 1613 are associated with each other. The host ID 1611 is information for identifying the business host 200, the cluster configuration host ID 1612 is information for identifying the business host 200 constituting the cluster, and the operating state 1613 is information indicating the operating state of the business host 200. , “Operating”, “standby” or “stop” are set. For example, in FIG. 5, it can be seen that the host ID “HostA1” and the host ID of the host constituting the cluster are “HostB1”, and “HostA1” is in the “operating” state.

The storage information 162 is information for managing the configuration of the storage system 300, and is associated with a storage system ID 1621, a volume ID 1622, and a port ID 1623 as shown in FIG. The storage system ID 1621 is information for identifying the storage system 300, the volume ID 1622 is information for identifying a logical volume, and the port ID 1623 is information for identifying a port. For example, in FIG. 6, it can be seen that the volume ID “Volume_1” of the storage system ID “StorageA” is accessed through the port ID “Port_1”.

The copy pair information 163 is information for managing a copy pair of a volume. As shown in FIG. 7, the copy pair information 163 is a copy pair ID 1631, a storage system (primary side) 1632, a storage system (secondary side) 1633, and a volume ID (primary side). 1634, volume ID (secondary side) 1635, copy type 1636, remote path ID (primary side) 1637, and remote path ID (secondary side) 1638 are associated with each other.

The copy pair ID 1631 is information for identifying a copy pair. The storage system (primary side) 1632 indicates identification information of the primary storage system 300. The storage system (secondary side) 1633 indicates identification information of the storage system 300 on the secondary side. The volume ID (primary side) 1634 indicates the identification information of the primary side volume. The volume ID (secondary side) 1635 indicates identification information of the secondary side volume. The copy type 1636 indicates information on the copy type. The remote path ID (primary side) 1637 indicates identification information of the primary side remote path. The remote path ID (secondary side) 1638 of the remote side indicates identification information of the remote side remote path.

For example, in FIG. 7, the copy pair ID “Pair_001” has a copy pair of “Storage_1” (primary side) “Volume_1” (primary side) and “StorageB” (secondary side) “Volume_5” (secondary side). It can be seen that the copy type is “GAD” (global-active device) and the copy is executed using the remote paths “Path1” (primary side) and “Path2” (secondary side).

The remote path information 164 is information for managing the configuration of the remote path, and as shown in FIG. 8, the remote path ID 1641, the primary storage system (primary side) 1642, the primary side port name (primary side) 1643, the secondary Side storage system (secondary side) 1644, secondary side port name (secondary side) 1645, and network bandwidth 1646 are associated with each other.

The remote path ID 1641 is information for identifying the remote path. The primary storage system (primary side) 1642 indicates identification information of the primary storage system 300. The primary side port name (primary side) 1643 indicates the identification information of the primary side port. The identification information of the secondary side storage system (secondary side) 1644 and the secondary side storage system 300 is shown. The secondary side port name (secondary side) 1645 indicates identification information of the secondary side port. A network band 1646 indicates the network band of the remote path.

For example, in FIG. 8, the remote path ID “Path1” connects “Port_1” (primary side) of “StorageA” (primary side) and “Port_4” (secondary side) of “StorageB” (secondary side), It can be seen that the network bandwidth of “Path1” is “100MB”.

The volume information 165 is information for managing the volume configuration, and is associated with a volume ID 1651, a storage system 1652, and an associated host ID 1653 as shown in FIG. The volume ID 1651 is volume identification information. The storage system 1652 is identification information of a storage system that provides the volume. The related host ID 1653 is identification information of the business host 200 associated with the volume.

For example, in FIG. 9, it can be seen that the volume ID “Volume_1” is provided by “StorageA” and is associated with “Host_A1”.

The port information 166 is information managed by the port configuration, and is associated with a port ID 1661, a storage system 1662, and a port role 1663 as shown in FIG. The port ID 1661 is information for identifying a port. The storage system 1662 is identification information of the storage system 300 having the port. The port role 1663 is information on a role assigned to the port, and for example, “RCU Target”, “Initiator”, and the like are set.

For example, in FIG. 10, it can be seen that the port ID “Port_1” is provided in “StorageA”, and the port role of the port is “RCU Target”.

(2-3) Storage System Configuration Returning to FIG. 2, the storage system 300 interprets the command transmitted from the business host 200 and executes read / write into the storage area of the disk device. In the above-mentioned global-active device, information written in the primary volume is transmitted to the storage system 300 that manages the secondary volume. The management server 100 is also provided with remote path configuration information in the global-active device, time-series data of the port data transfer amount, and the like.

(2-3-1) Hardware Configuration The storage system 300 mainly includes a controller 301 and a disk 320 or a disk 330. The controller 301 provides necessary information to the management server 100 in response to a command from the management server 100, or executes data read / write to the disk 320 or the disk 330 in response to a request from the business host 200. To do.

The storage system 300 includes a plurality of

disks

320 or 330, and one RAID (Redundant Arrays of Independent Disks) group is configured by one or a plurality of hard disk drives, and physical storage provided by one RAID group. One or more logical volumes are defined on the area. Then, this logical volume is provided to the business host 200. The business host 200 can write and read data by sending a predetermined command to the logical volume.

As described above, in the global-active device, when the primary business host 200 is switched to the secondary business host 200, I / O from the secondary business host 200 is transferred to the secondary business host 200. Data is first written to the primary volume via the remote path from the connected secondary storage system 300 to the primary storage system 300, and then the secondary volume forms a copy pair with the primary volume. Data is written to

Therefore, in the global-active device, the primary volume composed of the disk 320 managed by the primary storage system 300 is a volume on which the business host 200 reads and writes data, and the disk is managed by the secondary storage system 300. The secondary volume configured from 330 is a volume from which the business host 200 reads data.

Further, the data copy between the storage systems 300 is transferred via the

remote path

401 or 402. In order to realize the global-active device according to the present embodiment, at least the remote path 401 and the remote path 402 require remote paths in the upstream and downstream directions. The object of the present embodiment is to maintain an optimal cluster configuration so that the transfer amount of each remote path connecting between the enclosures of the storage system 300 is leveled.

(2-3-2) Functional Configuration The storage system 300 mainly includes a port data transfer amount providing program 310 and a storage device configuration information providing program 311 as programs executed by the controller 301.

The port data transfer amount providing program 310 provides configuration information of the storage system 300 to the management server 100 in response to a request from the host / storage device configuration information collection program 114 of the management server 100.

The storage device configuration information provision program 311 provides the configuration information of the storage system 300 to the management server 100 in response to a request from the port data transfer amount time series performance data collection program 113 of the management server 100.

(3) Details of System Configuration Switching Process Next, details of the system configuration switching process according to the present embodiment will be described with reference to FIGS. 11 and 12 are conceptual diagrams for explaining the system configuration switching process. 13 to 16 are flowcharts showing details of the system configuration switching process. FIG. 17 is a display example of performance data displayed on the display screen of the client terminal.

Hereinafter, a computer system having the cluster configuration shown in FIG. 11 will be described as an example. As shown in FIG. 11, four cluster configurations are formed by one management server 100 and eight business hosts 200.

Specifically, the business host 200A1 and the business host 200B1 form a cluster, the business host 200A1 is associated with the volume 320 (P1) of the primary storage system 300A, and the business host 200B1 is associated with the secondary storage system 300B. A volume (S1) is associated. The business host 200A2 and the business host 200B2 form a cluster, the business host 200A2 is associated with the volume 320 (P2) of the primary storage system 300A, and the business host 200B2 is associated with the volume of the secondary storage system 300B (S2). ) Are associated. The business host 200A3 and the business host 200B3 form a cluster, the business host 200A3 is associated with the volume 320 (P3) of the primary storage system 300A, and the business host 200B3 is associated with the volume of the secondary storage system 300B (S3). ) Is associated. The business host 200A4 and the business host 200B4 form a cluster, the business host 200A4 is associated with the volume 320 (P4) of the primary storage system 300A, and the business host 200B4 is associated with the volume of the secondary storage system 300B (S4). ) Is associated.

The business host 200A1, the business host 200A2, the business host A3, and the business host A4 are in an operating state (active), and the business host 200B1, the business host 200B2, and the business host B4 are in a standby state (active standby), and the business host B3 Is assumed to be stopped due to a failure.

Further, the storage system 300A and the storage system 300B copy data using three

remote paths

401, 402, and 403. The upper limit of the network bandwidth of each remote path is 100 MB / second. The remote path 401 is an upstream remote path and is used for the business hosts 200A1 to 200A2.

The remote path 402 and the remote path 403 are downstream remote paths. The remote path 402 is used for the business hosts 200B1 and B2. That is, it is used for data copying between copy pairs of the volume 320 (P1) and the volume 330 (S1), and between copy pairs of the volume 320 (P2) and the volume 330 (S2). The remote path 403 is used for the business hosts 200B3 and B4. That is, it is used for data copying between copy pairs of the volume 320 (P3) and the volume 330 (S3) and between copy pairs of the volume 320 (P4) and the volume 330 (S4).

FIG. 12 shows a state in which the cluster configuration composed of the business host 200A1 and the business host 200B1 is reversed by the system configuration switching process described later. That is, the business host 200A1 in the operating state is switched to the standby state, and the business host 200B1 in the standby state is switched to the operating state.

As shown in FIG. 12, when the cluster configuration is switched, the business host 200B1 becomes the primary system, and I / O from the business host 200B1 to the storage system 300B is forwarded from the secondary storage system 300B1 via the remote path 402. Is transferred to the storage system 300A1 on the side, and data is first written to the volume 320 (P1) on the primary side. Thereafter, the data is transferred from the primary storage system 300A1 to the storage system 300B1 via the remote path 402, and the data is written to the secondary volume 330 (S1).

Next, details of the system configuration switching process in the management server 100 will be described with reference to FIGS. As shown in FIG. 13, the port data transfer amount time-series performance data collection program 113 of the management server 100 receives the remote path used in the global-active device configuration from the port data transfer amount provision program 310 of the storage system 300. The port data transfer amount is acquired (S101). Then, the port data transfer amount time-series performance data collection program 113 stores the port data transfer amount acquired in step S101 in the port data transfer amount time-series performance data information 151 (S102).

The cluster configuration switching program 117 executes the cluster configuration switching determination process based on the acquired and port data transfer amount (S103). Details of the cluster configuration switching determination processing in step S103 will be described with reference to FIG.

As shown in FIG. 14, the cluster configuration switching program 117 determines whether the port data transfer amount has reached the network bandwidth (S201). When the threshold value of the port data transfer amount is set in advance by the user, in the determination process in step S201, the determination may be made based on the user setting value instead of the network band.

If it is determined in step S201 that the port data transfer amount has reached the network bandwidth, the cluster configuration switching program 117 acquires information on all copy pairs using the detected remote path ( S202). Then, the cluster configuration switching program 117 identifies the business host 200 whose cluster configuration is to be changed based on the copy pair information acquired in step S202, and uses the remote path whose port data transfer amount reaches the network bandwidth. All cluster configuration switching processing is executed (S203).

On the other hand, if it is determined in step S201 that there is no remote path whose port data transfer amount has reached the network bandwidth, the processing is terminated.

Next, the cluster configuration switching process in step S203 will be described in detail. As shown in FIG. 15, the cluster configuration switching program 117 performs cluster configuration for all combinations of business hosts 200 using the specified remote path. The process which changes is performed.

Hereinafter, a case where the cluster configuration of the business host 200 is n will be described, and the business hosts constituting the cluster will be described as host 1, host 2,. In this case, the process of steps S212 to S225 is repeated by changing the configuration of the host 1, the process of steps S213 to S221 is repeated by changing the configuration of the host 2, and the configuration of the host n is changed. The processes from S214 to S217 are repeated. However, if there is a failure in the cluster configurator of the host whose cluster configuration is to be changed, the cluster configuration cannot be switched, so the cluster configuration is not changed.

The cluster configuration switching program 117 acquires failure information of the cluster configuration of the host n (S214). If there is a failure in the cluster configuration target device of the host n, the cluster configuration cannot be switched. Therefore, the failure information of the cluster configuration is first acquired to determine whether the cluster configuration can be switched.

The cluster configuration switching program 117 determines whether there is a failure in the cluster configuration unit of the host n (S215). If it is determined in step S215 that there is no failure in the cluster configuration unit of host n, the cluster configuration of host n is reversed between operation and standby (S216). Then, a leveling value calculation process described later is executed (S217).

On the other hand, if it is determined in step S215 that there is a failure in the cluster configuration unit of host n, the next step is executed without executing the processing in steps S215 and S216.

Subsequently, the cluster configuration switching program 117 acquires failure information of the cluster configuration of the host 2 (S218).

Then, the cluster configuration switching program 117 determines whether there is a failure in the cluster configuration device of the host 2 (S219). If it is determined in step S219 that there is no failure in the cluster configuration unit of the host 2, the cluster configuration switching program 117 reverses the cluster configuration of the host 2 between operation and standby (S220). Then, the cluster configuration switching program 117 executes a leveling value calculation process described later (S221).

On the other hand, if it is determined in step S219 that the cluster configuration unit of host n has a failure, the cluster configuration switching program 117 executes the next step without executing the processing of step S220 and step S221.

Subsequently, the cluster configuration switching program 117 acquires failure information of the cluster configuration of the host 1 (S222).

The cluster configuration switching program 117 determines whether there is a failure in the cluster configuration unit of the host 1 (S223). If it is determined in step S219 that there is no failure in the cluster configuration unit of the host 2, the cluster configuration switching program 117 reverses the cluster configuration of the host 1 between operation and standby (S224). Then, the cluster configuration switching program 117 executes a leveling value calculation process to be described later (S225).

Next, with reference to FIG. 16, the leveling value calculation process executed by the leveling value calculation program 111 executed in step S217, step S221 and step S225 will be described.

As shown in FIG. 16, the leveling value calculation program 111 acquires the port data transfer amount of the remote path detected by the host n (S231). Subsequently, the leveling value calculation program 111 acquires the port data transfer amount of the remote path that is to be performed after the configuration change of the host n (S232).

Then, the leveled value calculation program 111 acquires the network bandwidth of the remote path detected by the host n (S233). Subsequently, the leveled value calculation program 111 acquires the network bandwidth of the remote path used after the configuration change of the host n (S234). The network standby acquired in step S233 and step S234 acquires the user setting value when the threshold value of the port data transfer amount is set in advance by the user.

Then, the leveling value calculation program 111 calculates the network usage rate (%) before and after the configuration change by the following formula 1 (S235).

Port data outflow / network bandwidth (user setting value) (Equation 1)

Then, the leveling value calculation program 111 calculates an absolute value by subtracting 50% from the network usage rate calculated in step S235, and adds them together (S236).

For example, in FIG. 12, it is assumed that the port data data transfer amount of the remote path 401 is 100 (MB / second) and the network bandwidth 100 (MB / second) has been detected in step S201. In step S202, information on all copy pairs using the remote path 401 is acquired. In FIG. 12, the acquired copy pair information includes volume 320 (P1) and volume 330 (S1), volume 320 (P2) and 330 volume (S2), volume 320 (P3), volume 330 (S3), and volume 320. (P4) and volume 330 (S4).

Then, the business host 200 whose cluster configuration is to be changed is determined based on the acquired copy pair information. That is, it is determined which of the cluster configurations of the business host 200 that uses the acquired copy pair is reversed.

Before determining which cluster configuration is to be reversed, the failure information of the cluster configuration machine of the business host 200 is acquired to determine whether the cluster configuration can be reversed. If the cluster configuration machine is faulty and cannot be operated, it is determined which cluster configuration is to be switched by excluding the cluster configuration. For example, in FIG. 12, since the business host 200B3 has a fault and is in a “stopped” state, the cluster configuration of the business host 200A3 is maintained without switching.

Next, the cluster configuration of the business host 200A1 in which no failure has occurred in the cluster configuration machine is reversed between operation and standby. FIG. 12 shows a state in which the cluster configuration of the business host 200A1 is reversed. The business host 200A1 is in a standby state (active standby), the business host 200B1 is in an operating state (active), and the volume of the storage system 300B from the business host 200B1. The I / O issued to 330 (S1) is written to the volume 320 (P1) of the storage system 300A via the remote path 402 (upward direction), and then via the remote path 402 (downward direction). , Written in the volume 330 (S1).

Then, the upstream and downstream port data transfer amounts of the

remote paths

401, 402, and 403 after switching the cluster configuration are acquired. In FIG. 12, for example, the port data transfer amount of the remote path 401 is 80 (MB / second), the port data transfer amount of the remote path 402 is 20 (MB / second), and the port data transfer amount of the remote path 403 is 0 (MB). / Sec).

Then, using the above formula 1, the network bandwidth usage rate is calculated from the acquired port data transfer amount and the network bandwidth. Assuming that the network bandwidth of the remote paths 401 to 403 is 100 (MB / second), the network bandwidth of each remote path is as follows.

(Network bandwidth usage rate of remote path 401)
= 80 (MB / sec) / 100 (MB / sec) = 80%
(Network bandwidth usage rate of remote path 402)
= 20 (MB / sec) / 100 (MB / sec) = 20%
(Network bandwidth usage rate of remote path 403)
= 0 (MB / sec) / 100 (MB / sec) = 0%

Then, the absolute value is calculated by subtracting 50% from the calculated network bandwidth usage rate, and these are added together to calculate the leveled value. In the above example, since the absolute value of the remote path 401 is 30, the absolute value of the remote path 402 is 30, and the absolute value of the remote path 403 is 50, the following leveling value is calculated.

Leveled value = 30 + 30 + 50 = 110

As shown in FIG. 15 above, for all combinations of cluster configuration switching (except for the cluster configuration in which a failure has occurred), the above switching process is executed to calculate a leveling value. In the example of FIG. 12, since one cluster configuration is excluded, the above processing is executed for seven combinations that are combinations of switching of the three cluster configurations. Then, a combination of cluster configurations having the lowest leveling value is selected from the calculated leveling values, and the cluster configuration is changed so as to maintain the cluster configuration. For example, in the example of FIG. 12, when a leveling value 90 lower than the leveling value 110 described above is calculated, the cluster configuration is changed to a leveling value 90 combination.

As described above, when changing the cluster configuration in the global-active device, the main cluster configuration is maintained without changing the volume configuration or changing the settings while maintaining the correspondence between the business host 200 and the logical volume. The system and subordinate system can be reversed. Therefore, as described above, for all cluster configurations that use remote paths whose port data transfer amount reaches the network bandwidth, switching is performed by reducing the load on the entire system, and all cluster configurations are combined. It is possible to estimate the performance of the remote path.

FIG. 17 is a display example 500 of performance data displayed on the display screen of the client terminal. From the display example 500, the user can confirm the port data transfer amount of the remote path used in the global-active device and the limit value of the network bandwidth.

In the display example 500 of FIG. 17, the analysis target column 510 displays the copy group name where the remote path is used, the analysis organization, the primary storage system and the secondary storage system using the remote path.

Then, the network column 520 displays the limit value of the network bandwidth of the remote path, the port data transfer amount of the remote path 00 and the remote path 01, the response time, etc. in a graph. The user can confirm the details of the performance of the current remote path with reference to the graph displayed in the network column 520.

(4) Effects of this Embodiment According to the above embodiment, when a data write request is made from the business host (host device of the present invention) 200 to the primary storage system (storage device of the present invention) 300 After writing the data to the primary volume of the primary storage system (storage device), the data is transferred via the first remote path, and the data is transferred to the secondary volume of the secondary storage system (storage device) 300. When there is a data write request from the business host (host device of the present invention) 200 to the secondary storage device, the data is transferred from the secondary storage system 300 (storage device) via the second remote path. After the written data is written to the primary volume, the data is transferred via the second remote path to the secondary volume. In the global-active device that writes to the volume, the management server 100 acquires the port data transfer amount of the first remote path and the second remote path, and the port data transfer amount is a predetermined threshold (predetermined network bandwidth or user). Storage system (storage) to which data is written from the business host 200 (host device) so that the port data transfer amounts of the first remote path and the second remote path are leveled Device) 300 is selected.

As a result, in a system configuration in which data is written via a remote path between storage devices, if any of the multiple remote paths becomes heavily loaded, the remote path is automatically switched to ensure that the available resources Effective use can be achieved.

100 Management Server 110 Host Configuration Change Command Program 110 Leveling Value 111 Leveling Value Calculation Program 112 Cluster Configuration Host Device Failure Detection Program 113 Port Data Transfer Amount Time Series Performance Data Collection Program 114 Storage Device Configuration Information Collection Program 115 Port Data Transfer Amount Time series performance data 116 Storage device configuration information 117 Program 151 Port data transfer amount time series performance data information 161 Host information 162 Storage information 163 Copy pair information 164 Remote path information 165 Volume information 166 Port information 200 Business host 210 Host configuration information provision program 211 Host configuration change program 300 Storage system 310 Port data transfer amount providing program 311 Storage device Configuration information providing program

Claims

A management computer that manages the performance of a storage system that makes the data requested to be written by the host computer redundant on a volume basis between the primary storage device and the secondary storage device,
Management computer
A memory for storing a data transfer amount of data transmitted / received between the host computer and the primary storage apparatus or the secondary storage apparatus;
A CPU for referring to the data transfer amount;
A network interface connected to the host computer, the primary storage apparatus or the secondary storage apparatus;
Have
When there is a data write request from the host computer, the primary storage device writes the data to the primary volume of the primary storage device and then transfers the data via the first remote path. Write data to the secondary volume of the storage device on the secondary side,
When there is a data write request from the host computer, the secondary storage apparatus writes the data transferred from the secondary storage apparatus via the second remote path to the primary volume. Transfer data via the second remote path, write to the secondary volume,
The management computer is
When the data transfer amounts of the first remote path and the second remote path are acquired, and the data transfer amount exceeds a predetermined threshold, the first remote path and the second remote path A management computer, wherein a storage device to which data from the host computer is written is selected so as to level the data transfer amount.
The management computer according to claim 1, wherein the predetermined threshold is a limit value of a network bandwidth or a preset setting value.
The host computer has a cluster configuration in which a primary host computer and a secondary host computer are paired,
The primary storage device is connected to the primary host computer, the secondary storage device is connected to the secondary storage device,
If there is a write from the primary host to the primary storage, after writing the data to the primary volume of the primary storage device, the data is transferred via the first remote path, and the secondary side Write data to the secondary volume of the storage device
When there is a write from the secondary host to the secondary storage, after the data transferred from the secondary storage device via the second remote path is written to the primary volume, the second volume Transfer data via remote path, write to the secondary volume,
The management computer is
When the data transfer amounts of the first remote path and the second remote path are acquired, and the data transfer amount exceeds a predetermined threshold, the first remote path and the second remote path The management computer according to claim 1, wherein the cluster configuration is switched so that the data transfer amount is leveled.
When the host computer is composed of a plurality of clusters including a plurality of pairs of the primary host computer and the secondary host computer,
The management computer is
The data transfer amount of the first remote path and the second remote path is acquired, the cluster configuration using the remote path whose data transfer amount exceeds a predetermined threshold is specified, and the cluster configuration is switched. The management computer according to claim 3, wherein the management computer obtains a data transfer amount.
The management computer is
The management computer according to claim 4, wherein a cluster configuration in which the data transfer amount is leveled is selected based on the data transfer amount after the switching of the cluster configuration.
The management computer is
6. The data transfer amount after switching the cluster configuration is obtained for all combinations of a plurality of cluster configurations, and the cluster configuration is changed so that the data transfer amount is leveled. The listed management computer.
The management computer is
The combination of the cluster configurations is selected from among the combinations of the cluster configurations, so that the usage rate of the remote path is close to a predetermined value so that the data transfer amount is leveled. The listed management computer.
The management computer is
The cluster configuration in which either the primary host computer or the secondary host computer has a failure among a plurality of cluster configurations is excluded from the combination that changes the cluster configuration. Management computer.
The host computer and the storage device are connected by a host bus adapter,
The management computer is
The management computer according to claim 8, wherein the host computer is instructed to switch connection to change a cluster configuration of the host computer.
A system configuration switching method in a management computer that manages the performance of a storage system that makes data requested to be written by a host computer redundant on a volume basis between a primary storage device and a secondary storage device,
The management computer includes a memory that stores a data transfer amount of data transmitted and received between the host computer and the primary storage device or the secondary storage device, a CPU that refers to the data transfer amount, A host computer, a network interface connected to the primary storage apparatus or the secondary storage apparatus,
If there is a data write request from the host computer to the primary storage device, after writing the data to the primary volume of the primary storage device, transfer the data via the first remote path, Writing data to the secondary volume of the secondary storage device;
If there is a data write request from the host computer to the secondary storage device, the data transferred from the secondary storage device via the second remote path is written to the primary volume, and Transferring data via a second remote path and writing to the secondary volume;
The management computer obtaining data transfer amounts of the first remote path and the second remote path;
Data from the host computer is used so that the management computer equalizes the data transfer amounts of the first remote path and the second remote path when the data transfer amount exceeds a predetermined threshold. Selecting a storage device for writing; and
A system configuration switching method.