Storage device
Patent Information
- Application Number
- JP2023081804
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-17
- Publication Date
- 2024-11-28
- Estimated Expiration
- 2043-05-17
AI Technical Summary
【0006】 本開示の一態様によれば、ストレージ装置のボリュームデータのコピー性能を改善できる。
Smart Images

Figure 2024165529000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a storage device. [Background technology]
[0002] Background of the present disclosure is U.S. Patent Application Publication No. 2008 / 0077735. For example, U.S. Patent Application Publication No. 2008 / 0077735 discloses a cache disk storage including: "a plurality of disk drive subunits configured to store data, each having a maximum capacity, the sum of the disk drive subunits comprising a maximum capacity of a cache; and a cache controller configured to respond to a requested upgrade adjustment of an effective capacity of the cache disk storage, the cache controller defining a capacity limit applicable to a file system providing active recognized effective capacity across the plurality of disk drive subunits, the recognized effective capacity of each of the disk drive subunits being a substantial proportion of the maximum capacity of the disk drive subunits in the same manner as a relationship between an aggregate amount of active recognized effective capacity of data storage provided by the file system in the cache disk storage and a maximum capacity of the cache, the capacity limit being related to the requested upgrade adjustment, and the capacity limit controlling a rate of data transfer with respect to the file system." [Prior art documents] [Patent documents]
[0003] [Patent Document 1] US Patent Application Publication No. 2008 / 0077735 Summary of the Invention [Problem to be solved by the invention]
[0004] A storage device may contain a mixture of data from systems with various requirements. Designers are required to design each volume taking into account various requirements, and the performance design of the storage device is becoming more complex. In order not to affect the processing performance of host access in the storage device, designers often set the copy speed to a uniformly low speed without designing the performance of copying volume data such as backup, or set up a small amount of dedicated CLPR (Cache Logical Partition) for operation. As a result, the copy performance that can be achieved is not being achieved. [Means for solving the problem]
[0005] A storage apparatus according to one embodiment of the present disclosure includes a storage controller that processes I / O requests from a host, and a storage device that stores data for the host, wherein the storage controller stores a schedule indicating copy speed levels for each of consecutive time periods within a cycle for each of a plurality of volumes, collects and stores current performance information of the storage controller in memory, and determines the copy speed level for the next time period based on the copy speed level for the next time period indicated by the schedule and the relationship between the value indicated by the performance information and a threshold value. Effect of the Invention
[0006] According to one aspect of the present disclosure, the copy performance of volume data in a storage device can be improved. [Brief description of the drawings]
[0007] [Figure 1] 1 illustrates a schematic configuration example of a computer system according to an embodiment of the present specification. [Diagram 2] 2 illustrates an example of a hardware configuration of a storage device. [Diagram 3] 2 illustrates an example of a hardware configuration of a management device. [Figure 4] 2 illustrates an example configuration of software executed by a storage controller and a management device. [Diagram 5] 13 shows an example of the configuration of a bitmap management table. [Figure 6] 13 shows an example of the configuration of a usage function management table. [Figure 7] 1 shows an example of the configuration of a performance information management table. [Figure 8] 1 shows an example of the configuration of a speed table. [Figure 9] 13 shows an example of the configuration of an execution speed management table. [Figure 10] 13 shows an example of the configuration of a speed adjustment count and special day management table. [Figure 11] 13 shows an example of the configuration of a deviation rate management table. [Figure 12A] 13 shows a flowchart of an example of a real-time copy speed adjustment process. [Figure 12B] 13 shows a flowchart of an example of a real-time copy speed adjustment process. [Figure 12C] 13 shows a flowchart of an example of a real-time copy speed adjustment process. [Figure 12D] 13 shows a flowchart of an example of a real-time copy speed adjustment process. [Figure 13] 13 shows a flowchart of an example of a process for creating a backup execution plan for the next day. [Figure 14] 13 shows a flowchart of an example of a process for determining a special day and adjusting an ideal value. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0008] Hereinafter, the embodiments will be described with reference to the drawings. Note that the embodiments are merely examples for realizing the present invention, and do not limit the technical scope of the present invention, and all of the combinations of features described in the embodiments are not necessarily essential to the solution of the invention.
[0009] In the following description, various types of information may be described using the expression "xxx table", but the various types of information may be expressed in a data structure other than a table. To indicate that it does not depend on the data structure, the "xxx table" may be called "xxx information". Also, in the following description, numbers are used as identification information for elements, but other types of identification information (e.g., names, identifiers) may be used.
[0010] In addition, in the following description, when describing elements of the same type without distinguishing between them, common symbols (or reference signs) may be used, and when describing elements of the same type with distinction between them, reference signs (or element IDs) may be used.
[0011] The program is executed by a processor (e.g., a CPU (Central Processing Unit)) included in the storage controller to perform a defined process using a storage resource (e.g., a main memory) and / or a communication interface device as appropriate, so the subject of the process may be the storage controller or the processor. The storage controller may also include a hardware circuit that performs some or all of the process. The computer program may be installed from a program source. The program source may be, for example, a program distribution server or a computer-readable storage medium.
[0012] 1 shows a schematic configuration example of a computer system according to an embodiment of this specification. The computer system includes a storage device 10, hosts 21 and 22, a management device 30, an external storage device 40, a public storage device 50, and a remote copy (RC) destination storage device 60. The number of each type of device is arbitrary, and some types of devices may be omitted. The external storage device 40, the public storage device 50, and the remote copy destination storage device 60 are destination storage devices.
[0013] The host 21 accesses a volume provided by the storage device 10, for example, via a storage area network. The host 21 sends an I / O request (write request or read request) specifying an I / O destination to the storage device 10. The host 22 accesses a volume provided by an external storage device 40, for example, via a storage area network. The host 22 sends an I / O request specifying an I / O destination to the external storage device 40.
[0014] The storage device 10 copies data stored in a volume to multiple backup destinations as needed. Copying for data backup (also simply called backup) is synchronous or asynchronous. The data to be copied may be the entire data of the copy source volume or differential data resulting from updates.
[0015] Backup within the storage apparatus 10 or backup from the storage apparatus 10 to another storage apparatus is controlled by the copy control unit 130. The copy control unit 130 controls the copy speed based on information collected in the computer system. This will be described in detail later. Note that, although data copying for backup purposes will be mainly described below, at least some of the features of the present disclosure may be applied to data copying for purposes other than backup.
[0016] 1, data of some primary volumes of the storage device 10 is copied to secondary volumes of the storage device 10 or other storage devices. The data of the primary or secondary volumes is stored in the physical storage device of the storage device that manages the volume.
[0017] 1, a secondary volume 42 for a primary volume 121 stored in the storage device 10 is stored in an external storage device 40. The external storage device 40 is an on-premise storage device, and communicates with the storage device 10, for example, via a storage area network. Data in the primary volume 121 of the storage device 10 is copied to the secondary volume 42 of the external storage device 40 via a virtual volume in the storage device 10.
[0018] A secondary volume 52 for a primary volume 122 stored in the storage device 10 is stored in a public storage device 50. The public storage device 50 is, for example, a storage device in a public cloud. Communication between the storage device 10 and the public storage device 50 can utilize, for example, a public IP network.
[0019] A secondary volume 124 obtained by copying a primary volume 123 stored in the storage device 10 is stored in the storage device 10. A secondary volume 62 obtained by copying the secondary volume 124 as a primary volume is stored in a remote copy destination storage device 60. The data in the secondary volume 62 can be used, for example, for recovering data in the primary volume 123 in the event of a disaster.
[0020] The storage device 10 and the remote copy destination storage device 60 communicate with each other via a network path including extenders 71 and 72. The extenders 71 and 72 enable long distance transmission on the network and convert between the protocol of the storage devices 10 and 60 and the protocol between the extenders 71 and 72. For example, a dedicated line may be used between the extenders, and a storage area network may be used between the extenders and the storage devices.
[0021] The management device 30 communicates with the storage device 10, the external storage device 40, the remote copy destination storage device 60, the hosts 21 and 22, and the network extenders 71 and 72 via a management network. The management device 30 collects various types of information from the other devices and stores the information in a database 350. In the example shown in Fig. 1, the performance management unit 320 collects performance information (load information) from each of the storage device 10, the external storage device 40, the remote copy destination storage device 60, and the hosts 21 and 22.
[0022] If possible, the management device 30 may communicate with the public storage device 50 or a device managing it in order to obtain performance information (load information) of the public storage device 50.
[0023] The replication management unit 330 collects remote copy-related performance information (load information) from the remote copy source storage apparatus 10 and the remote copy destination storage apparatus 60. The network (NW) bandwidth management unit 340 collects bandwidth information (communication load information) of the network used by the storage apparatus 10 for communication. In the example of Fig. 1, bandwidth information is collected from the extenders 71 and 72. The network bandwidth management unit 340 may additionally collect bandwidth information from the storage apparatus 10 and / or a network device between the storage apparatus 10 and the public storage apparatus 50.
[0024] It should be noted that only a part of the collected information may be collected, or performance information of other devices may be collected in addition. For example, information of the hosts 21 and 22 may not be collected, and information may be collected only from the host 21 that communicates with the storage device 10. Bandwidth information of other networks may be collected in addition to or instead of the bandwidth information of the network. For example, the network bandwidth management unit 340 may collect bandwidth information from the storage device 10, the external storage device 40, and / or a network device of the network between them, such as a switch. It should be noted that collection of performance information from the storage device 10 may be omitted. The storage device 10 holds and manages its own information.
[0025] Fig. 2 shows an example of the hardware configuration of the storage device 10. The storage device 10 includes drives 120, which are one or more physical storage devices, and a storage controller 110 connected to the drives 120. The storage controller 110 includes interfaces (I / F) 113-116, a memory 112, and a processor 111 connected thereto. In Fig. 2, each of a plurality of processors and memories is indicated by a reference symbol as an example.
[0026] The interface 114 is a communication interface device that mediates data exchange between the host and the storage controller 110. The host is connected to the interface 114, for example, via a storage area network. The interface 115 is an interface device for communicating with the external storage apparatus 40 and the remote copy destination storage apparatus 60. The interface 115 is connected to other storage apparatuses via a storage area network.
[0027] The interface 116 is a communication interface device for communicating with the management device 30 and the public storage device 50. The interface 116 is connected to, for example, a LAN. The interface 116 communicates with the management device 30 in the LAN or via the LAN and a public line. The interface 116 communicates with the public storage device 50 via the LAN and a public line.
[0028] The interface 113 is a communication interface device that mediates data exchange between the multiple drives 120 and the storage controller 110. The multiple drives 120 are connected to the interface 113, for example, via a switch (not shown).
[0029] The memory 112 stores programs executed by the processor 111 and data used by the processor 111. The memory 112 is a RAM (Random Access Memory), an NVRAM (Non Volatile RAM), etc. Programs and data stored in a non-transitory storage medium (not shown) are read into the memory 112.
[0030] The processor 111 executes a program stored in the memory 112. For example, a set of the memory 112 and the processor 111 is duplicated. The processor 111 is configured using a CPU, an MPU (Micro Processing Unit), a GPU (Graphics Processing Unit), etc. The processor 111 reads out and executes a program stored in the memory 112, thereby realizing a predetermined function.
[0031] The drive 120 is, for example, an HDD (Hard Disc Drive), and mainly stores user data. The drive 120 may be a drive made of semiconductor memory such as a flash memory. A plurality of drives 120 may constitute a RAID group. One or more logical storage areas (volumes) are created in the drive 120 based on the RAID group. The volume is associated with a physical storage area that the drive 120 has. Note that the same explanation as for the storage device 10 may also be applied to the hardware configurations of the other storage devices 40, 50, and 60.
[0032] 3 shows an example of the hardware configuration of the management device 30. The management device 30 includes a processor 302, a main memory device 303, an auxiliary memory device 304, an input device 305, an output device 306, and a communication interface 307. Each part of the management device 30 is connected to each other so as to be able to communicate with each other via a communication means such as a bus (not shown). Note that the management device 30 may be configured such that all or part of the configuration is realized by a virtual resource such as a cloud server.
[0033] The processor 302 is configured using a CPU, an MPU, a GPU, etc. A predetermined function is realized by the processor 302 reading and executing a program stored in the main memory device 303. The main memory device 303 is a device that stores programs and data, and is a ROM (Read Only Memory), a RAM (Random Access Memory), an NVRAM (Non Volatile RAM), etc.
[0034] The auxiliary storage device 304 is, for example, a solid state drive (SSD), an NVRAM, or an HDD. The auxiliary storage device 304 includes a non-transitory storage medium that stores programs and data. The programs and data stored in the auxiliary storage device 304 are read into the main storage device 303 as needed.
[0035] The input device 305 is an interface that accepts input of information, and is, for example, a keyboard, a mouse, a touch panel, a card reader, a microphone, etc. The output device 306 is an interface that outputs various information, and is, for example, a display device, a printer, an audio output device such as a speaker, etc. The communication interface 307 is a device that allows the management device 30 to communicate with other devices. Some of the components shown in FIG. 3 may be omitted, and other components may be added.
[0036] 4 shows an example of the configuration of software executed by the storage controller 110 and the management device 30. The software executed by the management device 30 includes a performance management unit 320, a replication management unit 330, and a network bandwidth management unit 340.
[0037] The performance management unit 320 collects and manages performance information 322 of the external storage device 40 and performance information 323 of the hosts 21, 22. The performance management unit 320 periodically collects information from the external storage device 40 and the hosts 21, 22, and includes it in external storage performance information 322 and host performance information 323. These are stored in a database 350. The external storage performance information 322 can include the same items as the performance information 142 of the storage controller 110, and the host performance information 323 indicates performance information of the host, and can include processor utilization rate, memory utilization rate, etc. Details of the items of the performance information 142 will be described later.
[0038] The replication management unit 330 collects and manages remote copy-related performance information 332 of the remote copy source storage unit 10 and the remote copy destination storage unit 60. The replication management unit 330 periodically collects information from the remote copy source storage unit 10 and the remote copy destination storage unit 60, and includes the information in the replication performance information 332. The replication performance information 332 is stored in a database 350.
[0039] The external storage performance information 322 and the replication performance information 332 may include, for example, the processor operating rate of the storage controller, the cache write waiting rate, the port data transfer amount, etc. The cache write waiting rate is the ratio of the area of data waiting to be destaged to the drive to the cache area. When multiple storage controllers or multiple ports are implemented, their average value or a value of a pre-specified device may be used. If possible, the remote copy destination storage device 60 may be treated as external storage, and information on the remote copy destination storage device 60 may be included in the external storage performance information 322 of the performance management unit 320.
[0040] The network bandwidth management unit 340 collects performance information of the network used by the storage device 10, includes it in network bandwidth information 342, and manages it. The network bandwidth information 342 can be acquired from network devices such as the extenders 71 and 72 and switches. The network bandwidth information 342 is stored in a database 350.
[0041] The firmware 150 executed by the storage controller 110 includes a copy control unit 130 and a performance information acquisition unit 140. The performance information acquisition unit 140 collects current performance information 142 (load information) of the storage controller 110. The performance information 142 can include, for example, a processor operating rate, a cache write waiting rate, a backup job utilization rate, etc. The backup job utilization rate is the ratio of the number of running processes to the maximum number of backup processes that the processor can execute.
[0042] The performance information acquisition unit 140 may further collect bandwidth information of the network used by the storage controller 110. For example, the information may include the port data transfer rates of the interfaces 113 to 116. The performance information 142 may be transferred to the management device 30 and stored in the database 350, for example. The collection cycle of the current performance information of different functional units may or may not be the same.
[0043] The copy control unit 130 includes a speed setting unit 132, a bitmap management unit 133, and a copy speed control unit 135. The bitmap management unit 133 manages the bitmap in volume copying. The speed setting unit 132 sets the copy speed for each volume. Note that multiple volumes may constitute one copy group, and a common copy speed may be set for the volumes in the copy group. The copy speed control unit 135 controls the copy speed for each volume based on the information of the computer system collected by the management device 30 and the performance information acquisition unit 140. The operation of the copy control unit 130 will be described in detail later.
[0044] The following describes an example of information held and managed by the storage controller 110. The information is stored in the memory 112. The management information of the storage controller 110 may also be stored in the drive 120 or other non-volatile storage device.
[0045] Fig. 5 shows an example of the configuration of a bitmap management table 810. The bitmap management table 810 manages the data difference between the source volume and the destination volume. In the example of Fig. 5, an address in the bitmap volume is represented by 8 bits. Column 811 indicates the first 4 bits, and row 812 indicates the last 4 bits. If the data at a certain address differs between the source volume and the destination volume, the value of that address in the bitmap management table is 1. If the data is the same and there is no difference, the value of that address is 0.
[0046] 6 shows an example of the configuration of a function in use management table 820. The function in use management table 820 manages flags for specifying the range of performance information to be referenced in copy control. The function in use management table 820 includes a flag column 822 indicating flags, and a function in use column 821 indicating functions to be used for copy control, which are each signified by the flag. The copy control unit 130 controls data copying using performance information collected by the functions indicated in the function in use column 821.
[0047] For example, flag 000 means performance information collected only by the firmware of the storage controller 110, i.e., the performance information acquisition unit 140. Flags 001, 010, and 100 respectively mean performance information from the performance management unit 320 only, performance information from the replication management unit 330 only, and performance information from the network bandwidth management unit 340 only. In the following explanation, "T" may represent the performance management unit 320, "R" may represent the replication management unit 330, and "N" may represent the network bandwidth management unit 340.
[0048] 7 shows an example of the configuration of a performance information management table 830. The performance information management table 830 manages information on performance items referenced in copy control. It includes a data storage destination column 831, a data item name column 832, an ideal value column 833 for days other than peculiar days, an ideal value column 834 for peculiar days, a flag column 835, and an instantaneous count column 836.
[0049] The initial values of the ideal value (other than a special day) and ideal value (special day) of each record in the performance information management table 830 are set by the user. In one example, the initial values of the ideal value column 833 for days other than a special day and the ideal value column 834 for a special day of each record are the same. The values of the ideal value column 833 for days other than a special day and the ideal value column 834 for a special day are updated by the copy control unit 130. Details of the update will be described later. In FIG. 7, the ideal value is represented in 16 bits.
[0050] The ideal values for the performance items obtained by the performance management unit 320, the replication management unit 330, and the network bandwidth management unit 340 may be ideal values set by the user in the performance management unit 320, the replication management unit 330, and the network bandwidth management unit 340, respectively, and stored in the performance information management table 830.
[0051] The data storage destination column 831 indicates the address of the memory 112 where the data is stored. The data item name column 832 indicates the item of performance information referenced in copy control. The ideal value column 833 for days other than special days indicates the ideal value of the data item other than special days. The ideal value column 834 for special days indicates the ideal value of the data item on special days. The ideal value is a threshold value referenced to determine the copy speed level. The meaning of the flag indicated in the flag column 835 is the same as that of the flag column 822 in the function usage management table 820. The real-time count column 836 indicates the real-time count number for one month, and the initial value is 0. The real-time count indicates the number of times the copy speed level has been lowered in real time, and is updated during copy processing. The real-time count will be described in detail later.
[0052] Each record in the performance information management table 830 indicates information on a data item in a data item name column 832 in performance information collected by a function indicated by the flag in a flag column 835. For example, the value in the flag column 835 of record 837 is 000, and therefore the subject is information collected by the performance information acquisition unit 140 of the storage controller 110. The data item name column 832 indicates a cache write latency ratio. Therefore, record 837 indicates information on the cache write latency ratio of the storage controller 110. Details of the peculiar days and ideal values will be described later.
[0053] Since the value of the flag column 835 of the record 838 is 001, the subject is information collected by the performance management unit 320 of the management device 30. The data item name column 832 indicates the processor operating rate. As described above, the performance management unit 320 collects and manages external storage performance information 322 and host performance information 323. In the example of FIG. 7, for ease of explanation, an ideal value common to the external storage device 40 and the hosts 21, 22 is defined, but a value different from that of the hosts 21, 22 may be set for the external storage device 40, and common or different ideal values may also be set between the hosts 21, 22.
[0054] Since the value of the flag column 835 of the record 839 is 010, the subject is information collected by the replication management unit 330 of the management device 30. The data item name column 832 indicates the port data transfer amount. Therefore, the ideal value of the data transfer amount per unit time of a specific port for remote copying of the storage device 10 is indicated in the record 839. The port data transfer amount may be, for example, the port data transfer amount of the remote copy destination storage device 60, or the average value of the data transfer amount per unit time of multiple ports.
[0055] Since the value of the flag column 835 of the record 840 is 000, the subject is information collected by the performance information acquisition unit 140 of the storage controller 110. The data item name column 832 indicates the copy process multiplicity. In other words, it indicates the ideal value of the copy process multiplicity of the storage controller 110. As the copy process multiplicity increases, the copy speed increases.
[0056] 8 shows an example of the configuration of a speed table 845. The speed table 845 shows copy speeds that can be set by the copy control unit 130. The copy speed is set for each volume. Note that multiple volumes may form a copy group, and a common copy speed may be set for these volumes.
[0057] The speed table 845 includes a speed column 846 and an identifier column 847. The example in FIG. 8 shows three speed levels, namely, low speed, medium speed, and high speed. The identifier is expressed by 8 bits, with low speed defined as -1, medium speed defined as 0, and high speed defined as +1. Note that the speed table 845 shows copy speed levels, which are different from the actual copy speed. The copy speed control unit 135 determines the actual copy speed according to the specified copy speed level. The copy speed changes depending on, for example, the degree of copy processing multiplicity, and is controlled more precisely than these speed levels.
[0058] Fig. 9 shows an example of the configuration of an execution speed management table 850. The execution speed management table 850 shows a schedule of copy speed levels set in advance for each volume in a predetermined cycle (time period) and a history of copy speed levels that have actually been selected. In one example, the predetermined cycle coincides with the cycle for acquiring performance information of the firmware 150. The storage controller 110 holds an execution speed management table 850 with the configuration shown in Fig. 9 for each volume or each copy group.
[0059] The execution speed management table 850 manages the schedule and performance information of the copy speed level for all subsequent times, for example, from the 1st to the last day of the month. The length of the time period for each subsequent time is fixed, for example, 3 minutes. Note that the backup copy is executed, for example, during a limited time period of one day. For example, the schedule of the copy speed level for one day, which is the backup cycle, is determined after the backup of the previous day is completed, and is registered in the execution speed management table 850.
[0060] The execution speed management table 850 includes a date column 851, an iteration column 852, and a speed column 853. The date column 851 indicates the date of the month and is represented by 5 bits. The iteration column 852 indicates the iteration on each day and is represented by 16 bits. Before the start of the copy process, the initial value of the speed column 853 specifies the copy speed level for each iteration (time period). During the execution of the copy process, the value of the speed column 853 is updated as needed to the copy speed level actually used in each iteration. As shown in the speed table 845, the copy speed levels are represented by -1, 0, and +1 for low, medium, and high speed, respectively. In the example described below, the initial value command specifies a specific copy speed or specifies that the copy speed level is the same as that of the previous iteration.
[0061] As will be described later, the actual copy speed level of each volume during each time period in a day is determined based on the set speed level indicated in the speed column 853 and the acquired performance information of the computer system. The value in the speed column 853 is updated based on the speed level actually used.
[0062] Fig. 10 shows an example of the configuration of a speed adjustment count and special day management table 860. The speed adjustment count and special day management table 860 holds the number of times the speed of the volume was changed from the 1st to the last day of the month in order to determine a special day. In the example of Fig. 10, the speed adjustment count and special day management table 860 includes a date column 861, an immediate count column 862, a change count column 863, and a special day flag column 864.
[0063] The immediate count column 862 and the change count column 863 indicate the immediate count number and change count number for one day, respectively. These are updated as needed during the copy process for each day within the month. The immediate count and change count will be described later. The value of the special day flag column 864 for the month is set before the start of the month. Details will be described later in the monthly flow. In Figure 10, the value of the flag indicating a special day is 1. A speed adjustment count and special day management table 860 is prepared for each month and each volume.
[0064] 11 shows an example of the configuration of a deviation rate management table 870. The deviation rate management table 870 holds a user-set value indicating how much deviation from an ideal value is to be judged as HIGH or LOW. The deviation rate management table 870 includes a HIGH column 871 and a LOW column 872. The HIGH column 871 indicates the threshold value of the deviation rate from the ideal value that is judged as HIGH, and the LOW column 872 indicates the threshold value of the deviation rate from the ideal value that is judged as LOW.
[0065] The processing of the copy control unit 130 will be described below. In the example described below, the copy control unit 130 determines special days and normal days other than special days based on the results of backup processing in a month, as will be described later with reference to FIG. 14. Furthermore, it determines the initial values of the copy speed level schedule for each of the special days and normal days. On the first day of a special day and the first day of a normal day in a month, the corresponding copy speed level schedule is used. This makes it possible to set an appropriate copy speed level even for special days that are different from normal days.
[0066] As will be described with reference to Fig. 13, the copy control unit 130 determines the copy speed level schedule for the second and subsequent special days based on the performance on the previous special day. Also, the copy speed level schedule for the second and subsequent normal days is determined based on the performance on the previous normal day. For example, if the special day is Tuesday, the copy speed level schedule for the second special day is determined based on the performance on Tuesday of the previous week. The copy speed level schedule for Wednesday, which is a normal day, is determined based on the performance on Monday of that week. This makes it possible to determine a more appropriate copy speed level schedule for each day.
[0067] 12A to 12D, the copy control unit 130 determines the copy speed level for each time period on the day based on a preset schedule and collected performance information of the computer system. This makes it possible to determine a more appropriate copy speed for the backup process on each day.
[0068] First, an example of real-time copy speed adjustment processing will be described with reference to Figures 12A to 12D. In this example, the real-time speed adjustment processing adjusts the copy speed to be equal to or lower than the ideal value at regular intervals, for example, at the same interval as performance information acquisition. This reduces the impact of the copy processing on other processing, including host IO processing.
[0069] In the example described below, the real-time copy speed adjustment process is executed at a predetermined interval (e.g., every 3 minutes) during a specific period of one day. The copy speed level of each volume during each time period is stored in the speed column 853 of the execution speed management table 850. As described above, the command specifically indicates low speed, medium speed, or high speed.
[0070] When the previous time period is compared with the current time period (current number), and there is a difference in the speed column 853, a command is executed to change the speed level to the one stored in the speed column 853 for the current time period (current number). When the previous time period is compared with the current time period (current number), and there is no difference in the speed column 853, this means that the copy speed level is the same as the previous time period, and no change in the speed level, i.e., no command is executed. By executing a command to change the speed level only when there is a difference in the speed column 853, it is possible to avoid unstable operation due to frequent increases and decreases in the copy speed level.
[0071] In the example described below, the performance information 142 in the firmware 150 of the storage controller 110 is assumed to be the cache write wait rate, the processor operating rate, the port data transfer amount, and the backup job utilization rate.
[0072] First, the copy speed control unit 135 refers to the performance information 142 and the performance information management table 830, and compares the latest cache write latency rate value with an ideal value (S11). The ideal value is shown in the performance information management table 830. The ideal value is selected depending on whether the day in question is a special day or not. Information on special days is shown in the number of speed adjustments and the special day management table 860. The cache write latency rate is the maximum value, average value, etc. within the performance information acquisition interval. If the cache write latency rate value is greater than the ideal value, the flow proceeds to the flowchart of FIG. 12B via connector A.
[0073] The copy speed control unit 135 refers to the performance information 142 and the performance information management table 830, and compares the latest processor operating rate value with an ideal value (S12). The ideal value is selected depending on whether the relevant day is an unusual day or not. The processor operating rate is the maximum value, average value, etc., during the performance information acquisition interval. If the processor operating rate value is greater than the ideal value, the flow proceeds to the flowchart in FIG. 12B via connector A.
[0074] The copy speed control unit 135 refers to the performance information 142 and the performance information management table 830, and compares the latest port data transfer amount value with the ideal value (S13). The ideal value is selected depending on whether the relevant day is an unusual day or not. The port data transfer amount is the maximum value, average value, etc. during the performance information acquisition interval. If the port data transfer amount value is greater than the ideal value, the flow proceeds to the flowchart of FIG. 12B via connector A.
[0075] The copy speed control unit 135 refers to the performance information 142 and the performance information management table 830, and compares the latest backup job utilization rate value with an ideal value (S14). The ideal value is selected depending on whether the day in question is an unusual day or not. The backup job utilization rate is the maximum value, average value, etc., during the performance information acquisition interval. If the backup job utilization rate value is greater than the ideal value, the flow proceeds to the flowchart in FIG. 12B via connector A.
[0076] If all of the four values are equal to or less than the ideal value, the copy speed control unit 135 assigns 0 to the flags T, R, and N (S15). Here, the flags T, R, and N indicate the flags of the performance management unit 320, the replication management unit 330, and the network bandwidth management unit 340, respectively.
[0077] Next, the copy speed control unit 135 refers to the performance information management table 830 to determine whether or not there is cooperation with the performance management unit 320 (S16). If the flag column 835 indicates a value including the performance management unit 320, cooperation with the performance management unit 320 exists. If cooperation with the performance management unit 320 exists (S16: Yes), the copy speed control unit 135 assigns 1 to flag T, and further copies the information collected by the performance management unit 320 from the database 350 (S17). If cooperation with the performance management unit 320 does not exist (S16: No), step S17 is omitted.
[0078] Next, the copy speed control unit 135 refers to the performance information management table 830 to determine whether or not there is cooperation with the replication management unit 330 (S18). If the flag column 835 indicates a value including the replication management unit 330, cooperation with the replication management unit 330 exists. If cooperation with the replication management unit 330 exists (S18: Yes), the copy speed control unit 135 assigns 1 to flag R, and further copies the information collected by the replication management unit 330 from the database 350 (S19). If cooperation with the replication management unit 330 does not exist (S18: No), step S19 is omitted.
[0079] Next, the copy speed control unit 135 judges whether or not there is cooperation with the network band management unit 340 by referring to the performance information management table 830 (S20). If the flag column 835 indicates a value including the network band management unit 340, cooperation with the network band management unit 340 exists. If cooperation with the network band management unit 340 exists (S20: exists), the copy speed control unit 135 assigns 1 to flag N, and further copies the information collected by the network band management unit 340 from the database 350 (S21). If cooperation with the network band management unit 340 does not exist (S20: does not exist), step S21 is omitted.
[0080] Next, when flag T=1, the copy speed control unit 135 refers to the external storage performance information 322, the host performance information 323, and the performance information management table 830, and compares each of the latest values of the data items specified by the performance information management table 830 with the corresponding ideal values (S22). When the latest value of any one of the data items is greater than the ideal value, the flow proceeds to the flowchart of Fig. 12B via connector A. When the latest values of all the data items are less than the ideal value, the flow proceeds to step S23.
[0081] Next, if flag R=1, the copy speed control unit 135 refers to the replication performance information 332 and the performance information management table 830, and compares each of the latest values of the data items specified by the performance information management table 830 with the corresponding ideal values (S23). If the latest value of any one of the data items is greater than the ideal value, the flow proceeds to the flowchart of Figure 12B via connector A. If the latest values of all the data items are less than the ideal value, the flow proceeds to step S24.
[0082] Next, if flag N=1, the copy speed control unit 135 refers to the network bandwidth information 342 and the performance information management table 830, and compares each of the latest values of the data items specified by the performance information management table 830 with the corresponding ideal values (S24). If the latest value of any one of the data items is greater than the ideal value, the flow proceeds to the flowchart of Figure 12B via connector A. If the latest values of all the data items are less than the ideal value, the flow proceeds to the flowchart of Figure 12C via connector B.
[0083] Next, the flowchart in Fig. 12B will be described. The copy speed control unit 135 starts the processing of the flowchart in Fig. 12B via a connector A. First, the copy speed control unit 135 refers to the execution speed management table 850 and determines whether there is a volume whose set copy speed level in the current time period (current time) is medium (S25).
[0084] If a medium-speed volume exists (S25: YES), the copy speed control unit 135 sets a -1 bit for the current time period (target time period) for the target volume (S26). Bit information for each time period and each volume is recorded in a table (not shown) in the memory 112. The copy speed control unit 135 changes the set copy speed level of the target volume from medium to low. The copy speed control unit 135 increments the value of the real-time count column 862 of the speed adjustment count and special day management table 860 and the value of the real-time count column 836 of the corresponding data item in the performance information management table 830 (S27).
[0085] Next, the copy speed control unit 135 refers to the execution speed management table 850 and determines whether there is a volume whose set copy speed level in the current time period (current time) is high (S28). If a high-speed volume is present (S28: YES), the copy speed control unit 135 sets a 0 bit for the target volume in the current time period (target time period) (S29). The copy speed control unit 135 changes the set copy speed level of the target volume from high to medium. The copy speed control unit 135 increments the value of the real-time count column 862 of the speed adjustment count and special day management table 860 and the value of the real-time count column 836 of the corresponding data item in the performance information management table 830 (S30).
[0086] Next, the flowchart of Fig. 12C will be described. The copy speed control unit 135 starts the process of the flowchart of Fig. 12C via a connector B. The process shown in Fig. 12C calculates the deviation rate from the ideal value.
[0087] The copy speed control unit 135 assigns 0 to the variables Tave, Rave, and Nave (S31). Tave is the average value of the deviation rates from the ideal values of the data items of the performance management unit 320. Rave is the average value of the deviation rates from the ideal values of the data items of the replication management unit 330. Nave is the average value of the deviation rates from the ideal values of the data items of the network bandwidth management unit 340.
[0088] Next, the copy speed control unit 135 judges the value of flag T (S32). If the value of flag T is 1 (S32:1), the copy speed control unit 135 calculates the value of Tave (S33). For example, the calculation formula is (T data copies / ideal value)*100 / (number of data items in T). That is, the sum of the ratios of the values of all data items in the performance management unit 320 to their respective ideal values is calculated, and then divided by the number of data items. If the value of flag T is 0 (S32:0), step S33 is omitted.
[0089] Next, the copy speed control unit 135 judges the value of flag R (S34). If the value of flag R is 1 (S34:1), the copy speed control unit 135 calculates the value of Rave (S35). For example, the calculation formula is (R data copies / ideal value)*100 / (R number of data items). In other words, the sum of the ratios of the values of all data items in the replication management unit 330 to their respective ideal values is calculated, and then divided by the number of data items. If the value of flag R is 0 (S34:0), step S35 is omitted.
[0090] Next, the copy speed control unit 135 judges the value of flag N (S36). If the value of flag N is 1 (S36:1), the copy speed control unit 135 calculates the value of Nave (S37). For example, the calculation formula is (data copies of N / ideal value)*100 / (number of data items of N). That is, the sum of the ratios of the values of all data items in the network bandwidth management unit 340 to their ideal values is calculated, and then divided by the number of data items. If the value of flag N is 0 (S36:0), step S33 is omitted.
[0091] Next, the copy speed control unit 135 calculates the deviation rate C from the ideal value of the cache write latency rate of the storage controller 110 (S38). For example, the calculation formula is (cache write latency rate / ideal value)*100.
[0092] Next, the copy speed control unit 135 calculates the deviation rate M from the ideal value of the processor operating rate of the storage controller 110 (S39). For example, the calculation formula is (processor operating rate / ideal value)*100.
[0093] Next, the copy speed control unit 135 calculates (S40) the deviation rate M from the ideal value of the port data transfer amount of the storage controller 110. For example, the calculation formula is (port data transfer amount / ideal value)*100.
[0094] Next, the copy speed control unit 135 calculates the deviation rate M from the ideal value of the backup job utilization rate of the storage controller 110 (41). For example, the calculation formula is (backup job utilization rate / ideal value)*100.
[0095] Next, the copy speed control unit 135 calculates the average value Ave of the previously calculated Tave, Rave, Nave, C, M, P, and S (S42). Thereafter, the flow proceeds to the flowchart in FIG. 12D via a connector C.
[0096] Next, the flowchart of Fig. 12D will be described. The copy speed control unit 135 starts the process of the flowchart of Fig. 12D via a connector C. The process shown in Fig. 12D executes a judgment to increase the speed for the same time period on the next day (same time on the next day) by one step when the set copy speed level in the current time period (current time) has a margin with respect to the ideal value. Note that the next day is the next normal day if the current day is a normal day other than a special day, and the next special day if the current day is a special day.
[0097] The copy speed control unit 135 compares the value of the average deviation rate Ave calculated in step S42 of Fig. 12C with the value of the threshold value LOW (S45). The value of the threshold value LOW is registered in the deviation rate management table 870.
[0098] If the value of the average deviation rate Ave is equal to or greater than the threshold value LOW, the copy speed control unit 135 compares the value of the average deviation rate Ave with the value of the threshold value HIGH (S46). The value of the threshold value HIGH is registered in the deviation rate management table 870. If the value of the average deviation rate Ave is equal to or greater than the threshold value HIGH, this process ends.
[0099] In step S45, if the value of the average deviation rate Ave is less than the threshold value LOW, the copy speed control unit 135 judges whether or not there is a volume whose set copy speed level is medium (S47). If there is a volume whose set copy speed level is medium, a +1 bit is set for the target volume and the current time period (target time period), and the value of the speed adjustment count and change count field 863 of the peculiar day management table 860 is incremented (S48).
[0100] If there is no volume whose set copy speed level is medium in step S47, or if the value of the average deviation rate Ave is less than the threshold value HIGH in step S46, the copy speed control unit 135 determines whether there is a volume whose set copy speed level is low (S49).
[0101] If a volume with a low-speed set copy speed level exists, a 0 bit is set for that target volume and the current time period (target time period), the value in the change count field 863 of the speed adjustment count and special day management table 860 is incremented (S50), and this process ends. If no volume with a low-speed set copy speed level exists, this process ends.
[0102] In the process described with reference to FIGS. 12A to 12D, the copy speed control unit 135 records information on the copy speed level actually used in each time period for each volume in the execution speed management table 850.
[0103] 12A to 12D, the copy speed control unit 135 determines the copy speed level for the next time period based on the copy speed level for the next time period indicated by the execution speed management table 850 and the relationship between the value indicated by the performance information of the computer system and the ideal value. The copy speed control unit 135 determines whether the speed level needs to be maintained, increased, or decreased, and determines to maintain the copy speed level for the next time period if the speed level needs to be maintained or increased as shown in Fig. 12D, or determines to decrease the copy speed level for the next time period if the speed level needs to be decreased as shown in Fig. 12B.
[0104] When the system load is higher than the set level in relation to the ideal value, the copy speed level is immediately lowered. This makes it possible to reduce the impact on other processes. Even if the system load is lower than the set level in relation to the ideal value, the copy speed level is not immediately raised, but the relationship is simply recorded. This makes it possible to avoid unstable operation due to frequent increases and decreases in the copy speed level. The copy speed level may be immediately raised. Also, the copy speed level does not have to be immediately lowered.
[0105] Next, the process of creating a backup execution plan will be described. A backup execution plan is executed at a fixed, predetermined cycle. This process aggregates data for a predetermined period at the end of regular backup, and determines an execution plan for each volume for the next predetermined period. In the execution plan, the copy speed for each time period of each volume is determined, and registered in the execution speed management table 850. The process described with reference to Figs. 12A to 12D is repeatedly executed within the cycle of creating a backup execution plan. For example, a backup execution plan is executed every day (24 hours), and the real-time speed adjustment described with reference to Figs. 12A to 12D is executed every three minutes.
[0106] FIG. 13 shows a flowchart of an example of a process for creating a backup execution plan for the next day. The backup execution plan shows a copy speed level schedule for each volume for one day. The copy speed setting unit 132 refers to the results of the previous special day to create a copy speed level schedule for the special day. Also, refers to the results of the previous normal day to create a copy speed level schedule for normal days other than the special day. The copy speed setting unit 132 repeatedly acquires bit information for each volume for each performance information acquisition interval in the process described with reference to FIGS. 12A to 12D, and executes the following process. The copy speed level schedules for the first special day (first day) of each month and normal days will be described later with reference to FIG. 14.
[0107] The speed setting unit 132 determines whether a bit for the time period exists (S61). If a bit does not exist, the speed setting unit 132 determines whether a bit for the previous time period exists (S62). If a bit for the previous time period does not exist, this process ends. If a bit for the previous time period exists, the speed setting unit 132 sets the same copy speed level as the time period before that in the execution speed management table 850 for the target volume to the relevant time period of the next day in the execution speed management table 850. Note that the next day refers to the next normal day if the current day is a normal day other than a special day, and the next special day if the current day is a special day.
[0108] If a bit exists in step S61, the speed setting unit 132 judges whether the value indicated by the bit is -1 (S64). If the value indicated by the bit is -1, the speed setting unit 132 judges the value of the bit for the immediately previous time period (S65). If the value of the bit for the immediately previous time period is -1, this process ends. If the value of the bit for the immediately previous time period is different from -1, the speed setting unit 132 sets the speed level of the target volume for that time period on the next day in the execution speed management table 850 to low speed (S66).
[0109] In step S64, if the value indicated by the bit is different from -1, the speed setting unit 132 judges whether the value indicated by the bit is 0 (S67). If the value indicated by the bit is 0, the speed setting unit 132 judges the value of the bit for the immediately previous time period (S68). If the value of the bit for the immediately previous time period is 0, this process ends. If the value of the bit for the immediately previous time period is different from 0, the speed setting unit 132 sets the speed level of the target volume for that time period on the next day in the execution speed management table 850 to medium speed (S69).
[0110] In step S67, if the value indicated by the bit is different from 0, the speed setting unit 132 judges whether the value indicated by the bit for the previous time period is 1 (S70). If the value indicated by the bit for the previous time period is different from 1, the speed setting unit 132 sets the speed level of the target volume for that time period on the next day in the execution speed management table 850 to high speed (S69). If the value indicated by the bit for the previous time period is 1, this process ends.
[0111] By the above-mentioned process, the copy speed level schedule for the next day can be appropriately determined according to the state of the computer system based on the bit indicating the relationship between the ideal value and performance information in the previous day's copy process, and the relationship between that relationship and the copy speed level set for the volume. Furthermore, by managing the bits, it is possible to avoid unstable operation due to frequent increases and decreases in the copy speed level.
[0112] Next, the process of determining peculiar days and adjusting ideal values will be described with reference to the flowchart in Fig. 14. This process may be, for example, a monthly flow that is executed every month. The monthly flow, for example, aggregates data from the 1st to the last day of the previous month on the 1st of each month, determines peculiar days, and adjusts ideal values.
[0113] The speed setting unit 132 acquires information on the real-time count number from the 1st to the last day of the previous month from the speed adjustment count and special day management table 860. Here, the real-time count number for each day for all volumes is acquired. Furthermore, the speed setting unit 132 judges a predetermined number of days with the highest real-time count number and other days (S81). Here, the top five days in terms of real-time count number are separated from the other days. The real-time count is incremented when the copy speed level is lowered based on the relationship between the performance information and the ideal value. A large real-time count number means that the copy speed level is frequently lowered from the preset value. This allows the special day to be appropriately determined. Note that a day on which the real-time count number is greater than a threshold value may be selected as a special day.
[0114] The speed setting unit 132 determines that a day with an instantaneous count number of 6 or less is a normal day different from an unusual day (S82). Furthermore, the speed setting unit 132 sets and determines a copy speed schedule for the normal day. Specifically, the speed setting unit 132 determines the copy speed level result (history) of the day with the smallest change count number among the days determined to be normal days as the copy speed level schedule for the first day of normal days in the next month (S83). The change count number indicates the number of times a copy speed increase has been determined, and the initial value of the copy speed level schedule for a normal day of each month is determined based on the copy speed level of a normal day of the previous month with the smallest number of copy speed increase determinations.
[0115] A small change count number means that the performance value has been trending close to the ideal value. The copy speed level schedule for the normal day with the smallest change count number is stored in the record for the first day of the normal day of the next month in the execution speed management table 850.
[0116] The speed setting unit 132 determines a special day based on the periodicity of the day with the highest real-time count number, such as the day of the week or the end of the month (S84). For example, if three or more days of the same day of the week are included, the day of the week in the next month may be determined to be a special day. If the end of the month is included in the higher-ranked days, the end of the month may be determined to be a special day. A date that is not included in the periodicity of the day of the week or the end of the month may be determined to be a special day in the next month. A day with a real-time count number of 6 or less may be specified as a special day. Days other than the special days become final normal days. The special days are set in the speed adjustment count and special day management table 860 for the next month. In this way, a special day in the current month is selected according to the number of times the copy speed level is reduced, as indicated by the real-time count number.
[0117] Next, the speed setting unit 132 determines the copy speed level schedule for the special day (S85). For example, the speed setting unit 132 determines the copy speed level result (history) of the day with the largest real-time count number as the copy speed level schedule for the special day. The copy speed level schedule for the special day is stored in the record for the first day of the special day of the next month in the execution speed management table 850. The real-time count number indicates the number of times a copy speed reduction has been determined, and an initial value of the schedule of the copy speed level for the special day of the next month is determined based on the copy speed level on the day with the largest real-time count number.
[0118] The special days may be further classified. For example, copy speed level schedules are determined for special days on specific days of the week, special days at the end of the month, and special days on specific dates. The copy speed level schedule may use the results of the day with the largest real-time count among the same type of special days in the previous month. The copy speed level schedule is determined by the daily flow with reference to the previous results of the unified type of special day.
[0119] Next, the speed setting unit 132 extracts data items with real-time counts ranked 5th or lower in the performance information management table 830 for the previous month (S86). Next, the speed setting unit 132 calculates the average deviation rate between the ideal value and the actual value from the 1st (the first day of the month) to the last day of the month for all extracted items ranked 5th or lower. Furthermore, the calculated average deviation rate is compared with the threshold value LOW (S87).
[0120] If the deviation rate average is smaller than the threshold LOW, the speed setting unit 132 increases the ideal values of the top five items in the execution results from the 1st to the last day of the previous month by 10% (S88). In the above process, when the actual results are more than the ideal value many times for a specific item and the other items have a margin up to the ideal value, the ideal value of the item that is more than the ideal value many times is increased.
[0121] In this way, when the value of an item that has caused a decrease in the copy speed level less frequently than other items in multiple cycles indicated by the real-time count number is lower than a predetermined level with respect to the corresponding ideal value (threshold), the speed setting unit 132 increases the ideal value (threshold) of the item that has caused a decrease in the copy speed level more frequently than other items in multiple cycles. This makes it possible to more appropriately increase the copy speed level from the viewpoint of the performance of the entire items. Also, from the viewpoint of the overall performance, it is possible to automatically correct the ideal value to a more appropriate value. Note that the ideal values for both special days and normal days may be increased, and the increase amount is arbitrary. The correction of the ideal value may be performed in a cycle different from the determination of the special day.
[0122] As described above, the bits, immediate count, and change count indicate the history of decisions as to whether the copy speed level should be maintained, increased, or decreased, respectively. The bits indicate the copy speed level that should be set by the decision. The immediate count indicates the number of decisions to decrease and the number of actual decreases. The change count indicates the number of decisions to increase.
[0123] 12A to 12D, 13, and 14 are not limited to the above examples. The process cycle described with reference to FIG. 14 is the longest, and the process cycle described with reference to FIG. 12A to 12D is the shortest.
[0124] The present invention is not limited to the above-mentioned embodiment, and various modifications are included. For example, the above-mentioned embodiment has been described in detail to easily explain the present invention, and is not necessarily limited to those including all of the configurations described. Also, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Also, it is possible to add, delete, or replace a part of the configuration of each embodiment with another configuration.
[0125] In addition, the above-mentioned configurations, functions, processing units, etc. may be realized in part or in whole by hardware, for example, by designing them as integrated circuits. In addition, the above-mentioned configurations, functions, etc. may be realized in software by a processor interpreting and executing a program that realizes each function. Information such as the program, table, file, etc. that realizes each function can be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card or SD card.
[0126] In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and not all control lines and information lines in the product are necessarily shown. In reality, it can be considered that almost all components are connected to each other. [Explanation of symbols]
[0127] 10 Storage Devices 30 Management device 40 External storage device 50 Public storage facilities 60 Remote copy destination storage device 71, 72 Extender 130 Copy control section 132 Speed setting section 135 Copy speed control section 140 Performance information acquisition section 320 Performance Management Department 330 Replication Management Department 340 Network Bandwidth Management Department
Claims
1. A storage device, comprising: a storage controller that processes I / O requests from a host; a storage device for storing data of the host; The storage controller includes: storing a schedule indicating a copy speed level for each of the multiple volumes during each successive time period within the cycle; Collecting and storing current performance information of the storage controller in memory; determining a copy speed level for the next time period based on a copy speed level for the next time period indicated by the schedule and a relationship between a value indicated by the performance information and a threshold value; Storage device.
2. 2. The storage device according to claim 1, The storage controller includes: copying data of a first volume of the plurality of volumes to a destination storage device via a network; Acquire current performance information of the destination storage device; A storage device that determines a copy speed level for the next time period of the first volume based on the copy speed level for the next time period indicated by the schedule of the first volume, the relationship between the value indicated by the performance information of the storage controller and the threshold value, and the relationship between the value indicated by the performance information of the destination storage device and the threshold value.
3. 2. The storage device according to claim 1, The storage controller includes: copying data of a first volume of the plurality of volumes to a destination storage device via a network; Acquire current performance information of the destination storage device; Obtaining current performance information for said host; Obtaining current bandwidth information of the network; A storage device that determines a copy speed level for the next time period of the first volume based on the copy speed level for the next time period indicated by the schedule of the first volume, the relationship between the value indicated by the performance information of the storage controller and the threshold value, the relationship between the value indicated by the performance information of the destination storage device and the threshold value, the relationship between the value indicated by the performance information of the host and the threshold value, and the relationship between the value indicated by the bandwidth information and the threshold value.
4. 2. The storage device according to claim 1, The storage controller includes: determining whether it is necessary to maintain, increase, or decrease the copy speed level for the next time period indicated by the schedule based on a relationship between the value indicated by the performance information and the threshold value; determining a copy speed level to be maintained for said next time period in response to said maintenance or increase need; The storage device determines a reduction in the copy speed level for the next time period in response to the need for the reduction.
5. 5. The storage device according to claim 4, The storage controller includes: creating a history of the results of a determination as to whether the copy speed level needs to be maintained, increased, or decreased during the period; The storage device creates a schedule of copy speed levels for the next period based on the history.
6. 2. The storage device according to claim 1, said period being one day; the storage controller selects a special day in the current month according to the number of times the copy speed level was reduced on each day of the last month; The threshold value is defined for each of an unusual day and a normal day different from the unusual day.
7. 6. The storage device according to claim 5, said period being one day; The storage controller includes: Selecting a specific day in the current month according to the number of times the copy speed level was decreased on each day of the last month; determining an initial value of the schedule of the copy speed level on the special day of the current month based on the copy speed level on the special day of the previous month on which the number of determinations of a decrease in the copy speed was the largest; The storage device determines an initial value of the schedule of the copy speed level on a normal day different from the special day of the current month based on the copy speed level on a normal day of the previous month on which the number of determinations of an increase in the copy speed was the smallest.
8. 2. The storage device according to claim 1, The performance information indicates values of a plurality of items, The threshold value is set for each of the plurality of items, The storage controller increases the threshold of an item that has caused a reduction in the copy speed level more times than other items in multiple cycles when the value of the item that has caused a reduction in the copy speed level less times than other items in multiple cycles is lower than a predetermined level for the corresponding threshold.
9. A copy control method in a storage device, comprising: The storage device includes a storage controller. the storage controller stores a schedule indicating a copy speed level for each of a plurality of volumes during each successive time period within a period; The copy control method includes: The storage controller collects and stores current performance information of the storage controller in a memory; A copy control method in which the storage controller determines a copy speed level for the next time period based on the copy speed level for the next time period indicated by the schedule and the relationship between the value indicated by the performance information and a threshold value.
Citation Information
Patent Citations
Data dual writing method for disk controller
JP1992084215A
Disk cache controller
JP1994180671A
Backup system
JP2004206611A
Information processor, information processing program and copy processing method
JP2012053645A
Storage system, storage control device, storage control method, and storage control program
JP2014063289A
