Computer system
The system dynamically manages journal volume capacity based on operation information to address the challenge of determining optimal size, maintaining efficient use and preventing waste in remote copy operations.
Patent Information
- Application Number
- JP2023073056
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-04-27
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
It is challenging for storage system administrators to determine the appropriate size of a journal volume in remote copy operations due to uncertainties in IO load and network conditions, leading to potential waste of capacity when initial allocations are too large or inefficient use when allocations are too small.
A computer system and method that dynamically manages the capacity of journal volumes based on operation information, expanding or releasing capacity as needed to maintain an appropriate range during remote copy operations.
The system ensures that the storage capacity allocated to journal volumes is maintained within an optimal range, preventing waste and ensuring efficient use throughout the operation period.
Smart Images

Figure 2024158117000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a computer system for remote copying. [Background technology]
[0002] Remote copy is used in a hybrid cloud operation mode that combines on-premise and cloud. A storage system described in Japanese Patent Application Laid-Open No. 2005-18506 is known as an example of remote copy.
[0003] A computer system for remote copying copies and stores update data written from a host server to a primary volume at a primary site (business site) in a journal volume (master journal volume). This data is then copied to a journal volume (restore journal volume) at a secondary site (backup site) asynchronously with the I / O of the primary volume.
[0004] In this way, data written from the host to the primary volume at the primary site is transferred to the secondary site asynchronously with the write request. Writing data from the restore journal volume to the secondary volume completes the remote copy. [Prior art documents] [Patent documents]
[0005] [Patent Document 1] JP 2005-18506 A Summary of the Invention [Problem to be solved by the invention]
[0006] The optimal size of the journal volume is estimated based on the IO load on the primary volume, the network bandwidth connecting the primary site and secondary site, the stability of the network, etc. However, it is inherently difficult for a storage system administrator to determine the appropriate size of the journal volume.
[0007] At the initial design stage of the journal group (JNL), it is difficult to accurately grasp in advance the maximum write amount of the IO load in actual operation, the worst case line conditions, etc. In the first place, there is a possibility of momentary network interruption between the primary site and the secondary site, or temporary degradation of transmission due to the convenience of the network operator.
[0008] In addition, once a block of a given storage capacity is allocated to a journal volume, it remains allocated to the journal volume and the capacity is not reclaimed. On the other hand, if a large capacity is initially allocated to a journal volume due to the difficulty of initial design, IO write pages will be allocated to the journal volume even in a state of temporary degradation of the transfer speed due to a network interruption, etc., and the capacity of the journal volume will be wasted.
[0009] It is desirable that a journal volume is first created with an appropriate capacity, and then that capacity is expanded as necessary. It is also desirable that any excess capacity that is no longer needed is released from the journal volume and put to effective use. It is an object of the present invention to provide a computer system and method that can maintain the storage capacity allocated to the journal volume within an appropriate range during the operation period of remote copy. [Means for solving the problem]
[0010] In order to achieve the above-mentioned object, the present invention provides a computer system comprising a first storage system constituting a primary site, a second storage system constituting a secondary site, and a management computer, wherein the first storage system comprises a primary volume and a primary journal volume, and the second storage system comprises a secondary volume and a secondary journal volume, and the management computer manages remote copying pairing the primary volume, the primary journal volume, the secondary journal volume, and the secondary volume, and expands or releases the capacity of the primary journal volume and / or the secondary journal volume according to operation information of resources related to the remote copying.
[0011] In order to achieve the above-mentioned object, the present invention further provides a method for controlling a storage system in which a management computer manages remote copying between a first storage system constituting a primary site and a second storage system constituting a secondary site, and the management computer dynamically expands or releases and changes the capacity of the journal volume in accordance with operation information of resources related to the operation of the remote copying. Effect of the Invention
[0012] According to the present invention, the storage capacity allocated to the journal volume can be maintained within an appropriate range during the operation of the remote copy. [Brief description of the drawings]
[0013] [Figure 1] FIG. 1 is a block diagram showing a configuration of a computer according to an embodiment of the present invention. [Diagram 2] 2 is a functional block diagram of an asynchronous remote copy in the computer system of FIG. 1. [Diagram 3] FIG. 2 is a block diagram of the memory structure of the storage system. [Figure 4] 4 is a block diagram of a storage structure of a memory of a management computer. [Diagram 5]1 is a block diagram showing a method for managing journal data in a journal volume. [Figure 6A] 13 is an example of a volume management table. [Figure 6B] 1 is an example of a journal management table. [Figure 6C] 1 is an example of a pair management table. [Figure 7A] 13 is an example of a volume operation information management table. [Figure 7B] 13 is an example of an inter-journal performance information management table. [Figure 7C] 1 is an example of an operation information management table of the storage system. [Figure 8A] 13 is an example of a pair configuration management table. [Figure 8B] 13 is an example of a primary / secondary journal configuration management table. [Figure 9] 13 is an example of a log of operations executed on a storage system. [Figure 10A] 13 is an example of a primary / secondary journal usage rate history management table. [Figure 10B] 13 is an example of an inter-journal performance information history management table. [Figure 11] 13 is an example of a determination history tendency management table. [Figure 12] 13 is an example of a determination map. [Figure 13] 13 is a flowchart for creating a journal volume. [Figure 14] 13 is a flowchart of updating the determination history tendency management table. [Figure 15] 15 is a flowchart of updating the determination history tendency management table (continuation of FIG. 14). [Figure 16] 13 is a flowchart of a procedure for changing the capacity of a journal volume depending on the operation status of a remote copy. [Figure 17] 13 is a flowchart for expanding or releasing the capacity of primary and secondary journal volumes based on a determination map. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0014] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. For clarity of explanation, the following description and drawings may be omitted or simplified as appropriate. The present invention is not limited to this embodiment, and any embodiment that meets the concept of the present invention is included in the technical scope of the present invention. Unless otherwise specified, each component may be plural or singular.
[0015] In the following description, various information may be described using the expression "xxx table", for example, but the various information may be expressed in a data structure other than a table. In order to indicate that the various information does not depend on the data structure, the "xxx table" may be referred to as "xxx information".
[0016] In addition, in the following description, when describing elements of the same type without distinguishing between them, reference signs (or common parts of the reference signs) may be used, and when describing elements of the same type with distinction between them, the element ID (or the element reference sign) may be used.
[0017] In the following description, a "computer system" is a system including one or more computers. Therefore, a "computer system" may be one computer, may be multiple computers, or may include devices other than computers in addition to the computers. The one or more computers typically include at least one physical computer. The one or more computers may also include at least one virtual computer.
[0018] Furthermore, in the following explanation, a "storage management device" may be made up of one or more computers. Specifically, for example, if the management computer has a display device and displays information on its own display device, the management computer may be a management system.
[0019] Also, for example, when a management computer (e.g., a server) transmits display information to a remote display computer (e.g., a client) and the display computer displays the information (when the management computer displays information on the display computer), a system including at least the management computer of the management computer and the display computer may be the storage management device.
[0020] In the following explanation, processing may be described with a "program" or its process as the subject, but since a program is executed by a processor or a controller (e.g., a CPU) to perform a defined process using storage resources (e.g., memory) and / or a communication interface device (e.g., a communication port) as appropriate, the subject of the processing may also be the processor. The processor operates as a functional unit that realizes a specified function by operating according to the program. Devices and systems that include a processor are devices and systems that include these functional units.
[0021] In the following description, a "storage device" may be, for example, an HDD or an SSD. Different types of devices may be mixed in the storage system.
[0022] 1 is a block diagram showing the configuration of a computer system 100 according to an embodiment of the present invention. The computer system 100 comprises a first storage system 110a and a second storage system 110b. Because asynchronous remote copy is performed between these storage systems, the storage system 110a is sometimes called the primary storage system, and the storage system 110b is sometimes called the secondary storage system. The computer system 100 further comprises a management computer 130, a host 150a (host 1), and a host 150b (host 2).
[0023] The storage system 110a (110b) has a storage device 116a (116b) which is a data storage device, a port 113a (113b) for connecting to a management network 122 to which a management computer 130 is connected, a memory 112a (112b), and a CPU: processor 111a (111b) connected to these elements.
[0024] Furthermore, the storage system 110a (110b) includes a port 114a (114b) for connection to a storage network 121 to which the host 150a (host 1) and the host 150b (host 2) are connected.
[0025] The port 114a (114b) performs interface processing for data input / output between the storage system 110a (110b) and the host 150a (150b), and may be, for example, an HBA.
[0026] The port 113a (113b) performs interface processing for data input / output between the storage system 110a (110b) and the management computer 130, and may be, for example, a NIC. The networks 121, 122 may be, for example, an FCSAN.
[0027] The storage system 110a (110b) provides the host 150a (150b) with one or more logical volumes from / to which data can be read and written. The storage system 110a (110b) receives an I / O command (e.g., a write command or a read command) specifying a logical volume from the host 150a (150b), and processes the I / O command.
[0028] In response to a write command or read command input from the host 150a (150b), the storage system 110a (110b) reads or writes data at an address location in a specified logical volume specified in the write command or read command.
[0029] The storage system may be composed of one server (node), or multiple servers (nodes) may be organized into one cluster and the cluster may be regarded as one storage system. The management computer, storage system, and host may be operated on physical / virtual servers in an on-premise data center. They may also be operated in a cloud environment provided by a public cloud. In the on-premise case, the storage device may be a physical HDD, SSD, etc. In the public cloud case, it may be a virtual device provided by the public cloud vendor (EBS, S3, etc. in the case of AWS). The management computer may be included in the storage system.
[0030] The management computer 130 has a port 133 connected to the network 122, a storage device 134, a memory 132, and a CPU: processor 131 connected to these elements. The port 133 performs interface processing for data input / output between the management computer 130 and the storage system 110a (110b).
[0031] The CPU 131 executes predetermined processes using various programs and data stored in the memory 132 and the storage device 134, thereby controlling the operation of the port 133 and performing overall control of the management computer 130. The management computer 130 manages the logical volumes provided to the storage system 110a (110b) and the host 150a (150b).
[0032] The host 150a (150b) has a port 152a (152b) connected to the network 121, a storage device 158a (158b), a memory 154a (154b), and a CPU (processor 156a (156b) connected to these elements. The port 152a (152b) performs interface processing for data input / output to / from the storage system 110a (110b). The CPU 156a (156b) controls the operation of the port 152a (152b) and performs overall control by executing predetermined processing using various programs and data stored in the memory 154a (154b) and the storage device 158a (158b).
[0033] 2 shows a functional block diagram of asynchronous remote copy performed between a primary storage system 110a at the primary site and a secondary storage system 110b at the secondary site. Asynchronous remote copy is a function that transfers data written from a host to a storage system at a business site (primary site) to a storage system at a backup site (secondary site) asynchronously with a write request.
[0034] In a remote copy system consisting of a primary site 110a and a secondary site 110b, a first remote copy pair consisting of a first consistency group (CTG1) 160 and a second remote copy pair consisting of a second consistency group (CTG2) 162 are set.
[0035] A primary volume PVOL1 and a (master) journal PJNL1 (170a) are set in the primary site 110a of the first consistency group (CTG1) 160. A journal is also called a journal group. The journal PJNL1 (170a) includes a (master) journal volume (JVOL1).
[0036] A secondary volume SVOL1 and a (restore) journal SJNL1 (170b) are set in the secondary site 110b of the CTG1 (160). The journal SJNL1 (170b) includes a (restore) journal volume (JVOL2).
[0037] Update data written from the host 150a (150b) to the primary volume (PVOL1) is copied and stored in the primary journal volume (JVOL1). The update data (journal data) in the journal volume (JVOL1) is copied to the secondary journal volume (JVOL2) of the secondary site 110b asynchronously with I / O to the primary volume (PVOL1).
[0038] Update data in the secondary journal volume (JVOL2) is written to the secondary volume (SVOL1) of the secondary site 110b. A remote copy pair of PVOL1-JVOL1-JVOL2-SVOL1 is set in CTG1.
[0039] Primary volumes (PVOL2, PVOL3) and a primary journal PJNL2 (180a) are set in the primary site 110a of the second consistency group (CTG2) 162. The primary journal PJNL2 includes a primary journal volume (JVOL3).
[0040] And, secondary volumes (SVOL2, SVOL3) and secondary journal SJNL2 (180b) are set in the secondary site 110b of CTG2 (162). The secondary journal SJNL2 includes a secondary journal volume JVOL4. In CTG2, a remote copy pair of PVOL2-JVOL3-JVOL4-SVOL2 and a remote copy pair of PVOL3-JVOL3-JVOL4-SVOL3 are set.
[0041] 3 is a functional block diagram of the storage structure of memory 112a (112b) of storage system 110a (110b). Bitmap management program 300 is a program for managing the block management bitmap 310 (FIG. 5) of the journal volume. Volume management program 302 is a program for managing the configuration of volumes, such as creation and deletion.
[0042] The journal management program 304 is a program for managing journals, journal volumes, creation, deletion, etc. The pair management program 306 is a program for managing pairs, such as creating and deleting volume pairs.
[0043] The operation information management program 308 is a program for measuring, saving and managing operation information of resources (hardware and software) related to remote copy operations, such as CPUs and volumes.
[0044] The block management bitmap 310 is a bitmap for managing whether journal data is transferred, not transferred, or unused for a journal volume (FIG. 5), and is updated by the bitmap management program 300. The volume management table 312 is a table for managing configuration information of volumes, and is updated by the volume management program 302.
[0045] The journal management table 314 is a table for managing configuration information of journals, and is updated by the journal management program 304. The pair management table 316 is a table for managing configuration information of pairs, and is updated by the pair management program 306.
[0046] The device performance information management table 318 is a table of performance information for the CPU and memory of the storage system, and is updated by the performance information management program 308. The volume performance information management table 320 is a table for managing volume performance information, and is updated by the performance information management program 308. The inter-journal performance information management table 322 is a table of inter-journal performance information, and is updated by the performance information management program 308. The operation log 324 is a table of logs of operations performed on the storage system.
[0047] 4 is a functional block diagram of the storage structure of the memory 132 of the management computer 130. A configuration management program 400 manages the configuration of the managed storage systems 110a, 110b. A trend management program 402 manages the trends of performance information of each of the multiple managed resources of the storage systems 110a, 110b. An information collection and update program 404 collects and updates configuration information and performance information from the managed storage systems 110a, 110b.
[0048] The pair configuration management table 406 is a table for managing the remote copy pair configuration of the managed storage systems 110a, 110b, and is updated by the configuration management program 400. The primary / secondary journal configuration management table 408 is a table for managing the configuration of the primary journal (PJNL) and secondary journal (SJNL) of the managed storage system, and is updated by the configuration management program 400.
[0049] The primary / secondary journal usage rate history management table 410 is a table for managing the historical trends of the usage rates of the primary journal (PJNL) and secondary journal (SJNL) of the managed storage system, and is updated by the trend management program 402. The inter-journal performance information history management table 412 is a table for managing the history of performance information between the primary journal and secondary journal of the managed storage system, and is updated by the trend management program 402.
[0050] The device operation information history management table 414 is a table for managing the history of the device operation information management table 318, and is updated by the trend management program 402. The judgment history trend management table 416 is a judgment table for changing, adjusting, modifying, correcting, etc. the configuration including the attributes, specifications, and operation of the primary and secondary journal volumes, and is updated by the trend management program 402. The judgment map 418 is a control table for how the configuration management program 400 changes the configuration of journals and journal volumes when multiple events occur in resources related to remote copy operations. This may be prepared in advance or updated as necessary by user input.
[0051] 5 is a block diagram showing a method of managing journal data in a journal volume. The journal volume JVOL receives data written to a primary volume and functions as a buffer for sending the data to a secondary volume.
[0052] Journal data (update data and metadata) is sequentially extracted from the first memory address L of the journal volume and sent to the secondary storage system 110b. The journal volume includes a transferred area where journal data has been transferred to the secondary journal volume, an area where journal data is stored but not yet transferred to the secondary journal volume, and an unused area.
[0053] The block management bitmap 310 stores a value indicating whether each block of the journal volume JVOL is in use or unused. If the corresponding block is in use, a bit value of "1" is set, and if it is unused, a bit value of "0" is set. The journal volume is divided into a number of blocks of a common, predetermined fixed size and managed. Each block is large enough to store a number of journal data.
[0054] The configuration management program 400 of the management computer 130 references the block management bitmap 310 based on the operating status trends of resources related to or affecting remote copy operations, and expands or frees up the capacity of the journal volume.
[0055] The configuration management program 400 releases from the journal volume a part or all of the transferred area of the journal volume to reduce the storage capacity remaining allocated to the journal volume. The configuration management program 400 can also release a part of the unused area of the journal volume.
[0056] 6A is an example of the volume management table 312. The volume management table 312 has a volume resource ID, capacity, and its attributes. The attributes include those for normal volumes accessed from a host, and those used as journal volumes.
[0057] 6B is an example of the journal management table 314. The journal management table 314 has information on the ID of the journal resource, the ID (VOLID) of the journal volume included in the journal, the capacity of the journal volume (JVOL capacity), its status (Normal, Error), and the usage rate of the journal volume.
[0058] 6C shows an example of the pair management table 316. The pair management table 316 has a remote copy pair ID (resource ID), the ID of the primary volume included in the pair (PVOLID), the ID of the primary journal volume (PJNLID), the ID of the secondary volume (SVOLID), the ID of the secondary journal volume (SJNLID), a group ID when multiple pairs are managed in a group (CTG), and a status indicating the state of the pair.
[0059] The statuses include, for example, PAIR, where pair data continues to be copied normally, SUSPEND, where copying is temporarily stopped, PFUL, where the journal volume is about to overflow, and PSUE, where an abnormality has occurred due to some error.In addition, since it is assumed in these tables that the volume and journal IDs are globally unique, the ID indicating the device is omitted, but if the volume and journal IDs are unique within the device, information indicating the primary and secondary device IDs may be included.
[0060] 7A is an example of the volume performance information management table 320, which includes the volume ID, information indicating the type of metric (performance information), time, and the performance information value at each time. The volume performance information management table 320 records values every minute, but the time interval is not limited to this.
[0061] FIG. 7B is an example of an inter-journal performance information management table 322, which records the journal ID (PJNLID) of the primary site 110a, the journal ID (SJNLLID) of the secondary site 110b, information indicating the metric type, and the value at each time.
[0062] 7C is an example of the device (storage system) operation information management table 318, and shows the IDs of the CPU and memory as one of the resources of the storage system, the type of metric, time, and the metric value at that time. The memory write pending rate indicates the proportion of data that accumulates because it cannot be written to memory as the amount of write data increases. This is an index showing that the load on the storage system is high. The metrics are not limited to the values shown here. For example, volume response time, drive operation rate for each drive, memory cache hit rate, CPU, memory, and drive operation rates for each node in the storage cluster, etc. may also be used.
[0063] Fig. 8A is an example of a pair configuration management table 406. The pair configuration management table 406 basically has the same information as the pair management table 316 of the storage system. Fig. 8B shows an example of a primary / secondary journal configuration management table 408. The primary / secondary journal configuration management table 408 has the same information as the journal management table 314 of the managed storage systems 110a, 110b. It basically has the same column information as the journal management table 314 of the storage system.
[0064] Fig. 9 is an example of an operation log 324 executed for the storage system 110a (110b). The operation log may have multiple values when multiple resources are the target of an operation (for example, an array). The specified parameters are values required for the operation content, such as the size when creating a volume. The operation log may be in table format or a list of text. The columns (values) recorded as a log are not limited to the example in Fig. 9.
[0065] The management computer 130 collects operation information of resources related to the remote copy operation from the storage systems 110a, 110b, and creates a history of the operation information as a history management table. The management computer 130 calculates the trend from the collected operation information history, and based on this trend, determines the remote copy operation status, for example, the trend of the usage rate of the primary journal volume and / or the secondary journal volume (an increasing or decreasing trend of the usage rate), and based on this, the capacity of the journal volume can be dynamically changed in accordance with the remote copy operation. As a result, the storage capacity allocated to the journal volume can be maintained within an appropriate range during the operation period of the remote copy.
[0066] 10A is an example of a primary / secondary journal usage rate history management table 410. The primary / secondary journal usage rate history management table 410 holds the history of usage rates (journal volumes) by time for each of multiple journals PJNL1, PJNL2, SJNL1, and SJNL2 of storage systems 110a and 110b, which are targets managed by the management computer.
[0067] The usage rate of a journal volume refers to the ratio of the untransferred area of journal data and the transferred area to the total capacity of the journal volume. The usage rate (utilization rate) of each journal and journal volume is an example of operation information of resources related to remote copying.
[0068] Fig. 10B is an example of an inter-journal performance information history management table 412. This has a history of data transfer speed (load on the network 121) as inter-journal performance information (metrics) for each combination of a primary journal (PJNL ID) and a secondary journal (SJNLL ID). The metrics are not limited to the values described here. For example, they may be the response time between journal volumes, or the IO load (IOPS) on a journal volume. The data transfer speed between multiple journals is an example of performance information for resources related to remote copying.
[0069] Fig. 11 is an example of the judgment history trend management table 416. In Fig. 11, the primary journal volume, the primary storage system, the secondary storage system, and the network between the primary journal volume and the secondary journal volume are each an example of resources related to remote copy.
[0070] The judgment refers to a judgment for identifying the load trend in the resource operation information (hereinafter simply referred to as "trend"). The trend is, for example, "normal" or "performance degradation" in FIG. 12. The management computer 130 changes the configuration of the journal volume based on the operating state trend for the remote copy operation of the storage system. Therefore, the management computer 130 can dynamically change the configuration of the primary and secondary journal volumes according to the operation status of the remote copy, and can continuously keep the storage capacity allocated to the journal volume within an appropriate range during the operation period of the remote copy.
[0071] The judgment history trend management table 416 includes the ID of the storage system (DKC) (DKC1: primary storage system), the resource ID of the storage system, and resource operation information (metrics). For a metric, if there is a resource related to the resource ID, the related resource ID is recorded. The trend value is a base value for judgment and is an operation information value, and may be, for example, an average of peak values for a certain recent period (e.g., one day).
[0072] 11, the CPU utilization rate of a storage system and the write pending rate of memory are also examples of resource operation information related to remote copy. The resource operation information includes at least one of the load of the first storage system, the load of the second storage system, the IO load on the journal volume, the utilization rate of the journal volume, and the load of the network between the first storage system and the second storage system, which have been explained above.
[0073] 12 shows an example of the judgment map 418. The judgment map is configured in a table format. The judgment map 418 includes multiple combinations of trends in the operation information of each of multiple resources related to remote copying, and a scenario for expanding or releasing the capacity of the journal volume for each combination. Examples of the "tendency" of each operation information are "constant" and "increasing" for the IOPS of the PJNLVOL, "normal" for the load of the primary DKC (primary storage system) 110a, "normal", "performance degradation", and "upper limit reached" for the network status between journals (JNL NW status), and "normal" and "high load" for the load of the secondary DKC (secondary storage system) 110b.
[0074] The determination map 418 includes the trends of the operation information of each of a plurality of resources. This trend is determined by the management computer by comparing the "trend value" shown in Fig. 11 with the current operation information. The storage system (DKC) load is, for example, the CPU utilization rate and the cache write pending rate.
[0075] The JNL NW (inter-journal network: 122) status indicates, for example, the value of the data transfer speed between the primary and secondary journals. The judgment map includes remote copy operational adjustments according to scenarios such as trends and developments of whether the usage rates of the primary journal volume and secondary journal volume will increase or decrease in the future based on a combination of trends in multiple resource operation information. The remote copy operational adjustments include changes to the journal volume configuration (expanding or releasing the capacity of the journal volume, etc.).
[0076] The management computer 130 executes a configuration change of the journal volume according to the judgment map. The management computer 130 may handle the configuration change of the journal volume according to this judgment map, or may execute logic equivalent to the judgment map based on a flowchart.
[0077] Fig. 13 is an example of a flowchart for creating a journal volume. It is assumed that a primary volume (PVOL) and a secondary volume (SVOL) are created in each of the primary and secondary storage systems 110a and 110b. Fig. 13 is executed when starting a remote copy operation.
[0078] 13 starts when a user accesses the management computer 130. The user instructs the configuration management program 400 of the management computer 130 to create a remote copy pair (S1300). In response, the configuration management program 400 instructs the primary and secondary storage systems 110a, 110b to create a journal volume (S1302).
[0079] Upon receiving this instruction, the volume management programs 302 of the primary and secondary storage systems each create a journal volume (S1304). At this time, the capacity of the journal volume may be the minimum required capacity, may be the same capacity as the volume (PVOL), or may be preset by the user as an appropriate value.
[0080] Next, the configuration management program 400 of the management computer 130 instructs the primary and secondary storage systems to create a journal (S1306). In response, the journal management programs 304 of the primary and secondary storage systems create a journal and register it in the journal management table 314 (S1308). Next, the configuration management program 400 instructs the pair management programs 306 of the primary and secondary storage systems to create a remote copy pair (S1310).
[0081] Then, the pair management program 306 of the primary and secondary storage systems creates a remote copy pair from the information in the volume management table 312 (Figure 6A) of the volumes (including the journal volumes), and registers this in the journal management table 314 (Figure 6B) and the pair management table 316 (Figure 6C) (S1312), thereby ending the flowchart.
[0082] Next, the management computer 130 collects information from the storage systems 110a, 110b and updates the above-mentioned table (FIG. 4). The management computer 130 updates the judgment history information (judgment history trend management table 416: FIG. 11) according to the flowchart in FIG. 14. The management computer 130 executes the flowchart in FIG. 14 at predetermined time intervals.
[0083] The information collection and update program 404 refers to the volume management table 312, journal management table 314, pair management table 316, volume performance information management table 320, and inter-journal performance information management table 322 from the primary and secondary storage systems 110a, 110b (S1400).
[0084] Next, the information collection and update program 404 updates (S1402) the pair configuration management table 406, the primary and secondary journal configuration management table 408, the primary and secondary journal usage rate history management table 410, the inter-journal performance information history management table 412, and the device performance information history management table 414. Then, the information collection and update program 404 updates (S1404) the judgment history trend management table 416 based on the flowchart of FIG.
[0085] 15 is a flowchart for updating the determination history trend management table 416 (FIG. 11). The trend management program 402 updates the determination history trend management table 416 based on each management table updated based on the flowchart of FIG.
[0086] The trend management program 402 calculates trend values for device operation information (CPU utilization, memory write pending rate, drive operation rate, etc.) of the primary and secondary storage systems, the IOPS of the primary and secondary journal volumes, the amount of data transferred between journals, etc., for a certain period of time in the past (e.g., the past day), and registers these in the judgment history trend management table 416 (S1500).
[0087] The trend value may be, for example, the average value excluding the maximum and minimum peak values of each metric (operation information). The reason for excluding the maximum and minimum values is to eliminate extreme values. The peak value in this case is, for example, the value of the metric when the difference between consecutive periods of operation information is positive.
[0088] Next, the trend management program 402 compares the trend value (recorded value) recorded in the judgment history trend management table 416 with the currently calculated performance information (current value) (S1502). The trend management program 402 calculates the difference between the recorded value and the current value, and compares the difference with a threshold value (S1504). If there is at least one metric whose difference exceeds the threshold value, the judgment is affirmative and the process proceeds to step S1506. If the judgment is negative, the trend management program 402 determines that there has been no substantial change in the performance information of the resources related to the remote copy operation, and ends the flowchart without changing the configuration of the journal volume.
[0089] The threshold value may be a fixed value (e.g., 5% of the trend value recorded in the judgment history trend management table 416) that is determined in advance, or the user may input a predetermined value each time. The threshold value may be different for each resource operation information (metric).
[0090] S1504 may be performed as follows. A weight is set for each piece of performance information of a plurality of resources, and the trend management program 402 multiplies the difference (S1502) by the weight for each piece of performance information. The trend management program 402 may then sum up the values thus obtained for all pieces of performance information, and compare the sum with a predetermined value.
[0091] The configuration management program 400 assigns a weight to each of the operation information, thereby making it possible to rank the influence of a plurality of operation information on a change in the configuration of the journal volume. Therefore, it is possible to comprehensively determine whether or not the configuration of the journal volume needs to be changed in accordance with the operation of the remote copy.
[0092] When a storage system has multiple journal volumes, the threshold may be different for each journal volume. By changing the threshold, the timing at which the capacity of the journal volume is changed can be changed for each journal volume.
[0093] When the management computer 130 judges whether or not it is necessary to change the capacity of a plurality of journal volumes in the storage system, the priority of the operational information may be changed for each of the plurality of journal volumes.
[0094] For example, when there are multiple journal volumes, the management computer 130 may execute a volume configuration change decision by prioritizing the CPU utilization rate for a first journal volume, and may execute a volume configuration change decision by prioritizing the memory write pending rate for a second journal volume. In this way, it is possible to determine whether or not a volume configuration change is necessary depending on the differences in the attributes of the journal volumes.
[0095] Furthermore, when there are multiple journal volumes whose difference in trend value exceeds the threshold, the trend management program 402 executes S1506 for the journal volume with the largest X value based on the priorities X (highest priority), Y (second) and Z (third) of multiple pieces of performance information, for example, performance information X, Y and Z (X=CPU utilization, Y=IOPS, Z=transfer speed).When there are multiple journal volumes with the same X value, the trend management program 402 compares the Y values and executes S1506 for the journal volume with the largest Y.
[0096] Next, the configuration management program 400 executes S1506 for the remaining journal volumes with the largest value of Y, and finally executes S1506 for the journal volume with the largest value of Z. S1506 is executed for all journal volumes in this order of priority.
[0097] If the trend management program 402 judges S1504 as positive, the flow chart proceeds to S1506. The configuration management program 400 takes action to change the capacity of the journal volume in accordance with the operation status of the remote copy (S1506). This will be explained in detail in the flow chart of FIG. 16. Furthermore, after the flow chart is terminated via S1506, the trend management program 402 returns to the flow chart of FIG. 14 and can update the judgment history trend management table 416 in the state after the configuration change of the journal volume has been made (S1404).
[0098] Upon receiving an instruction from the configuration management program 400, the journal management program 304 executes the process shown in the flowchart of Fig. 16 to change the configuration of the journal volume (increase or decrease the capacity). The journal management program 304 refers to the volume operation information management table 320 shown in Fig. 7A to determine whether the IOPS of the primary journal volume (PJNLVOL) is zero (S1600).
[0099] If the result of the step of S1600 is affirmative, the journal management program 304 proceeds to S1602. In S1602, the bitmap management program 300 refers to the bitmap management table (FIG. 5) of the primary journal volume (PJNLVOL), and releases the allocation of the pages of the blocks (pages) of PJNLVOL for which journal data has already been transferred to the secondary journal volume (SJNLLVOL) from the primary journal volume (PJNLVOL).
[0100] If the journal management program 304 judges in S1600 that there is no error, it refers to the inter-journal performance information management table 322 (FIG. 7B) and judges whether or not there is a failure in the network between the primary journal and the secondary journal (S1604). For example, the configuration management program 400 refers to the volume performance information management table 320 and the inter-journal performance information management table 322, and if the Status of the primary and secondary journals is not Error, and the IOPS of the primary volume (PVOL) is not zero, but the data transfer rate between the journal volumes is zero, it judges there is a network NW error (S1604 / YES).
[0101] At this time, the pair management program 306 will, over time, change the status of the remote copy pair in the pair management table 316 (Figure 6C) from "pair" to "suspend" to stop the remote copy, and the journal management program 304 will not perform configuration changes to the journal volume such as capacity release or capacity expansion as in S1602.
[0102] If the journal management program 304 denies S1604, it proceeds to S1606 and checks the operation log 324 (Figure 9) to see if the most recent operation has affected the configuration of the SVOL and prevented the SVOL from achieving its desired performance (e.g., creating a snapshot).
[0103] If the journal management program 304 judges in the affirmative in S1606 that such an operation has been performed, it instructs the volume management program 302 to restrict the inflow of IO to the primary volume (PVOL) and ends the flowchart (S1608). At this time, when the configuration change of SVOL is resolved, remote copy is immediately resumed, so no configuration change is made to the journal volume.
[0104] If the result in S1606 is negative, the journal management program 304 proceeds to S1610 and checks whether or not there is a failure in at least one of PJNLVOL, SJNLLVOL, and SVOL by referring to the volume operation information management table 320 (FIG. 7A). If the result in S1610 is positive, the configuration management program 400 ends the flowchart.
[0105] When a failure occurs in the primary journal volume (PJNLVOL), remote copy is not performed, and so the journal management program 304 ends the flowchart without performing any processing to change the configuration of the journal volume. When a failure occurs in the secondary journal volume (SJNLVOL), the usage rate of the primary journal volume (PJNLVOL) increases, but because a failure suspend occurs and remote copy is not performed, the journal management program 304 ends the flowchart without performing any processing to change the configuration of the journal volume.
[0106] In the case of a failure in the secondary volume (SVOL), the usage rate of both the primary journal volume and the secondary journal volume increases, but since a failure suspension occurs and remote copy is not performed, the journal management program 304 ends the flowchart.
[0107] On the other hand, if the journal management program 304 rejects S1610, that is, if PJNLVOL, SJNLVOL, and SVOL are all normal, the remote copy operation continues, and the capacity of the journal volume is expanded or released S1612 (FIG. 17) based on the decision map of FIG. 12, and then the flowchart ends.
[0108] Fig. 17 is a flowchart for the configuration management program 400 of the management computer 130 to expand or release the capacity of the primary and secondary journal volumes based on the judgment map of Fig. 12. The trend management program 400 reads resource operation information by referring to the management table described above, and compares it with the trend values of Fig. 11 to determine the trends for each of the load of the primary journal volume (PJNLVOL IOPS), the load of the primary storage system (primary DKC), the state of the inter-journal network (inter-JNL NW), and the load of the secondary storage system (DKC), and compares the combination of the multiple trends with the judgment map of Fig. 12.
[0109] In S1700, the configuration management program 400 determines, in accordance with the map in FIG. 12, whether the "action" corresponding to the combination of a plurality of evaluations is one in which the usage rate threshold of the journal volume (JNLVOL) should be checked.
[0110] If the determination is affirmative, that is, if it is determined that remote copying is in a state in which the usage rate of the journal volume is about to be increased, the configuration management program 400 proceeds to S1702.
[0111] The configuration management program 400 checks whether the usage rate of the primary journal volume (PJNLVOL) and / or secondary journal volume (SJNLVOL) is equal to or greater than a threshold value according to the map in Fig. 12. In Fig. 12, "P / SJNLVOL" is both the primary journal volume and the secondary journal volume. The threshold value may be determined in advance, such as 80%.
[0112] If the configuration management program 400 denies S1702, although the usage rate of the journal volume tends to increase, the journal volume still has room to store journal data, and it ends the flowchart by determining that there is no need to expand the capacity of the journal volume. The reason why the configuration management program 400 performs S1702 is to prevent the limited memory capacity of the storage system from being wasted on the journal volume if the capacity of the journal volume is expanded without limit.
[0113] The configuration management program 400 proceeds to S1704 where it affirms S1702 and judges whether the current journal volume capacity is equal to or less than a predetermined value, for example, the capacity of the primary volume. The reason for making such a judgment is to prevent the size of the journal volume from increasing indefinitely in order to operate remote copy stably. If the configuration management program 400 denies S1704, it judges that the capacity of the journal volume is large and cannot be expanded any further, and ends the flowchart.
[0114] If S1704 is answered in the affirmative, the configuration management program 400 calculates the expansion size of the journal volume capacity. The size to be expanded may be determined from past trends, or may be a certain percentage of the current capacity, or a fixed value. The configuration management program 400 proceeds to S1710, and causes the journal management program 304 to expand the journal volume capacity by the calculated size.
[0115] If the configuration management program 400 denies S1700, that is, if it determines that remote copying is not likely to increase the usage rate of the journal volume, it proceeds to S1712. The configuration management program 400 checks whether the "action" is a transfer completion check and page release.
[0116] If the configuration management program 400 judges S1712 to be positive, that is, if it determines that remote copy is in a state in which the usage rate of the journal volume is about to be reduced, it proceeds to S1714. In S1714, the configuration management program 400 causes the bitmap management program 300 to refer to the bitmap management table (FIG. 5) and release the transferred area from the journal volume, and then ends the flowchart.
[0117] The configuration management program 400 determines that the answer to S1712 is No, that is, that the remote copy is not in a state in which the usage rate of the journal volume is about to be reduced, and ends the flowchart.
[0118] As described above, the management computer allocates a specified storage capacity to each of the primary journal volume and the secondary journal during the initial design of the remote copy, and after the remote copy begins, the allocation of capacity to the primary journal volume and / or the secondary journal volume is performed according to the execution status of the remote copy, so that even when the remote copy is performed, the storage capacity of the journal volume can be managed efficiently.
[0119] Furthermore, in the process of executing a storage system control method in which the management computer 130 manages remote copying between a first storage system 110a constituting the primary site and a second storage system 110b constituting the secondary site, the management computer dynamically expands or releases and changes the capacity of the journal volume in accordance with operation information of resources related to the remote copy operation, so that the storage capacity of the journal volume can be maintained appropriately during the operation period of the remote copy.
[0120] The judgment map may be different between the primary journal volume and the secondary journal volume, that is, the content of the response to the combination patterns of the multiple trends (patterns of expanding and releasing the capacity of the journal volume) may be different. Furthermore, when there are multiple journal volumes in one storage system, the judgment map may be different for each journal volume. The judgment map may also be changed according to the operation form and operation status of the remote copy. [Explanation of symbols]
[0121] 100: computer system, 110a: primary storage system (first storage system), 110b: secondary storage system (second storage system), 130: management computer, JVOL1: primary journal volume, JVOL2: secondary journal volume, PVOL1: primary volume, SVOL1: secondary volume
Claims
1. a first storage system constituting a primary site; a second storage system constituting a secondary site; and A management computer; A computer system comprising: the first storage system comprises a primary volume and a primary journal volume; the second storage system comprises a secondary volume and a secondary journal volume; The management computer managing a remote copy in which the primary volume, the primary journal volume, the secondary journal volume, and the secondary volume are paired; and expanding or releasing the capacity of the primary journal volume and / or the secondary journal volume according to operation information of resources related to the remote copy; Computer system.
2. The management computer determining an increasing or decreasing trend in the utilization rate of each of the primary journal volume and the secondary journal volume according to the resource operation information; expanding the capacity in accordance with the determination of the increasing trend; releasing the capacity in accordance with the determination of the decreasing trend.
2. The computer system of claim 1.
3. the resource operation information includes a load of the primary journal volume, a load of the first storage system, a load of the second storage system, and a load of a network between the primary journal volume and the secondary journal volume; 2. The computer system of claim 1.
4. The management computer determining a load trend for each of the plurality of resources related to the remote copy; Setting a pattern for expanding or releasing the capacity for each combination of the plurality of trends; 2. The computer system of claim 1.
5. The management computer determining a trend value of the operation information based on the history of the operation information; determining a trend of the load by comparing a current value of the operation information with the trend value; 5. The computer system of claim 4.
6. The management computer Comparing a current value of the operation information with a trend value of the operation information; Based on the comparison result, the capacity is expanded or released.
6. The computer system of claim 5.
7. The management computer when an operation for changing the configuration of the secondary volume is performed, neither the expansion nor the release of the capacity is executed for the primary journal volume and / or the secondary journal volume; 2. The computer system of claim 1.
8. The management computer if a failure occurs in the network between the primary volume and the secondary volume, neither the expansion nor the release of capacity is executed for the primary journal volume and / or the secondary journal volume; 2. The computer system of claim 1.
9. The management computer and executing the capacity expansion when the usage rate of the journal volume is equal to or greater than a threshold value when determining an increasing trend of the usage rate.
2. The computer system of claim 1.
10. The management computer expanding the capacity of the journal volume when the capacity of the journal volume is equal to or smaller than a predetermined value; 10. The computer system of claim 9.
11. The management computer At the initial design time of the remote copy, a predetermined storage capacity is allocated to each of the primary journal volume and the secondary journal volume; After the remote copy is started, the allocation of capacity to the primary journal volume and / or the secondary journal volume is executed according to an operational state of the remote copy.
2. The computer system of claim 1.
12. A method for controlling a storage system, in which a management computer manages remote copying between a first storage system constituting a primary site and a second storage system constituting a secondary site, comprising: the management computer dynamically expands or releases and changes the capacity of the journal volume in accordance with operation information of resources related to the operation of the remote copy; A method for controlling a storage system.
Citation Information
Patent Citations
Storage system
JP2005018506A
