JP2008299559A - Storage system and data migration method in storage system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique capable of preventing degradation of access performance during data transfer at the time of data transfer in a storage system. <P>SOLUTION: In this storage system, a control part 100 performs processing (P1) of data transfer, for example, from a low-speed volume (V1) 51 to a high-speed volume (V2) 52 of a storage area 202, transfers object data to a first area (CA2, V3) dynamically ensured on a shared memory 120 (cache memory 114) according to this, and performs processing (P2) allowing response to access from a host 300 or the like. According to this, access performance or the like during transfer can be ensured. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a storage apparatus having a function such as RAID control, and a storage system (information processing system) configured to include the storage apparatus, and more particularly to techniques such as data / information lifecycle management and data migration.

  In a storage system, data life cycle management (DLCM), information life cycle management (ILCM), and the like are known. These are technologies that manage and operate a series of life cycles from creation / storage of data / information to storage / storing / deletion. Data / information can be adjusted according to changes in importance and utility value. Information is stored in an appropriate storage device / storage area.

  Further, in a storage system, there is a data migration technique that migrates (rearranges) a storage area / storage device that stores data in accordance with data access performance or the like. For example, some storage devices move data between a high-speed volume and a low-speed volume when access increases or decreases.

As an example of technology relating to life cycle management and data migration, Japanese Patent Application Laid-Open No. 2006-215554 (Patent Document 1) describes an archive management method for a storage apparatus. Japanese Patent Laid-Open No. 9-69031 (Patent Document 2) describes logical volume relocation of a storage apparatus.
JP 2006-215594 A JP-A-9-69031

  In connection with the DLCM and data migration, when data on a volume is migrated between storage areas / storage devices, access performance to the data is lowered during the data migration. In particular, during the data migration time, it is impossible to respond to data input / output access (data read / write request) from the host to the data. In addition, since it was a volume unit, it was migrated including unnecessary data, so it took extra time to migrate.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a technology capable of preventing a decrease in access performance or the like during data migration when performing DLCM or data migration in a storage system. It is in.

  Of the inventions disclosed in the present application, the outline of typical ones will be briefly described as follows. In order to achieve the above object, the present invention provides a storage device unit that provides a storage area (for example, a volume) by a storage device group such as an HDD, and a control unit (controller) that controls storage of data in the storage device unit. , A storage system including the storage apparatus, and a data migration method in the storage system, characterized by having the following configuration.

  In the storage system of the present invention, the storage device has a first volume and a second volume in which data is stored in a storage area, and the control unit includes a shared memory or cache memory in the control unit, and a host device Including an external communication interface processing unit and a storage device processing interface processing unit. Then, the control unit dynamically secures on the shared memory in accordance with the first process for controlling the data migration from the first volume to the second volume and the first process according to a predetermined condition. By transferring the target data corresponding to the data to be migrated in the first volume to the first area so that the access response is possible and reflecting it to the second volume, at least access to the target data during the migration And a second process for ensuring performance.

  The present invention relates to data (information) life cycle management and data migration, and when data on a volume is migrated (moved / copied) between storage areas / storage devices, the data is being migrated (from the migration source to the migration destination volume). In order to ensure the access performance and the like for the data even in the transition time until the data is migrated, the following configuration is provided. In the system and method of the present invention, a virtual volume or cache is used as a first area in a memory (for example, a shared memory of a storage device) in order to prevent a decrease in data access performance or the like during a migration time in the storage system. An area is dynamically secured, and target data corresponding to the migration data is transferred onto the first area so that access to the data can be made responsive.

  The first area is used to ensure access during the migration time of data (migration data) processed as data migration, and the target data is stored in the first area according to policy settings and the like. The The first area (virtual volume or cache area) depends on the data migration trigger, migration data capacity, migration time (migration start and end), memory status, policy settings (threshold settings, etc.), etc. Secured dynamically in the system. That is, the first area is managed from creation (reservation) to deletion (release) with the start and end of data migration, and the capacity (size), remaining time, and the like also vary. The first area is secured with the start of migration, and is erased when the first area or data on it becomes unnecessary after the migration is completed.

  With the above-described configuration, at least access performance to the data during the data migration time is ensured. In addition, it is possible to secure data access performance associated with data migration while suppressing the load on the system.

  In this system and method, data migration processing between storage areas (volumes) (referred to as first processing), and accompanying processing for securing access performance and the like by dynamic allocation of the first area (second processing) And the like.

  The first process has a migration source first volume and a migration destination second volume, and migrates the target data from the first volume to the second volume. The first process is performed as a dynamic storage reallocation process based on policy settings or the like. For example, the control unit determines migration data (such as a file) or a trigger for migration based on the degree of access to the data, and migrates migration data between volumes. Only necessary data is transferred to the second volume based on the threshold comparison judgment regarding the number of times of access, the order, and the like for the file having access in the first volume.

  In the second process, in addition to the dynamic allocation of the first area, particularly the virtual volume (third volume) associated with the first and second volumes, the first volume to the third volume Transfer of target data to (transfer / copy), response to access such as data input / output from the host in the third volume (data reference or update), from the third volume to the migration destination second volume The target data (with update possibility) is reflected (moved / copied), the old data is deleted, and the like. The second process is performed as an automatic process associated with the first process based on policy setting or the like. For example, as the transfer data and timing are determined based on the degree of access to the data, the control unit transfers the data to be transferred to the first area (such as the access area in the file) or the first area. The capacity and the like are determined, and the target data is transferred to the first area. Corresponding to the migration data, only necessary data is transferred to the first area (third volume). The data (target data) to be transferred to the first area corresponds to the data (migration data) in the first volume of the migration source, for example, a file that is data handled by the host, or the Access area data, etc.

  In the first area, the virtual volume (third volume) is external to the storage, particularly the shared memory (access from the host) of the control unit, with respect to the volumes (first and second volumes) secured on the storage side. Is allocated on the memory) and can be correlated. The cache area is an area (virtual volume cache area) provided for arranging one or more virtual volumes, for example. That is, a virtual volume cache area is primarily provided for a shared memory or the like, and a virtual volume is secondarily provided. Then, the target data is arranged in the virtual volume by transfer. On the other hand, when erasing (releasing) these data, the data is erased in the order of target data, virtual volume, and cache area.

  For example, data migration (first processing) is performed between storage devices / volumes having different access processing speeds. For example, the high-speed volume is shifted to the low-speed volume in accordance with a decrease in the number of accesses to data. For example, the low-speed volume is shifted to the high-speed volume in accordance with an increase in the number of accesses to data. In addition, as the storage resource arrangement, for example, a configuration in which a plurality of types of volumes to which data migration is performed is provided inside the storage device, or a plurality of types of storage resources are divided into an internal storage device and an external storage device. There are configurations with volumes.

  Also, in particular, the unit of transfer to the first area is different from the unit of data migration (files, etc.), and is a unit such as a block of the access area in the file, thereby reducing excess data transfer and processing. Increase efficiency.

  Among the inventions disclosed in the present application, effects obtained by typical ones will be briefly described as follows. According to the present invention, when DLCM or data migration is performed in a storage system, it is possible to prevent a decrease in access performance or the like during data migration.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.

  In this embodiment, when data (file) on a low-speed volume is transferred to a high-speed volume in connection with data transfer in the storage system, the host for the data (access area data in the file) even during the data transfer In order to ensure the access performance and the like from the In this system, a virtual volume cache area and a virtual volume are dynamically allocated as the first area in the shared memory (cache memory) of the control unit of the storage apparatus, and the migration data is transferred to the first area (virtual volume). The target data (access area data) corresponding to the (file) is transferred so that it can respond to the data input / output access from the host. The first area is used for securing access during the migration time of data (migration data) processed as data migration. The first area is dynamically secured according to the migration data capacity, data migration time (migration start and end), and is erased when there is no access for a predetermined time or more after the migration is completed. With the configuration as described above, access performance such as data input / output access from the host to the migration data during the data migration time is ensured. In addition, while ensuring a load on the system, in particular, a double load between the data migration process and the host access response process in the control unit of the storage apparatus, it is possible to secure data access performance and the like accompanying the data migration.

(Embodiment 1)
The storage system and method (data migration method) according to the first embodiment of the present invention will be described with reference to FIGS. 1 is a schematic configuration, FIG. 2 is a system configuration example, FIG. 3 is a hardware configuration example, FIG. 4 is a storage structure example, FIG. 5 is a memory detail, FIG. 6 is a migration and transfer detail, 7 to 9 are processing flows, FIGS. 10 to 16 are management information table examples, and FIGS. 17 and 18 are display screen examples.

<Overview>
FIG. 1 shows a schematic configuration relating to features in the present embodiment. The storage system includes a host (host device) 300, a control unit (storage controller) 100, and a storage area (storage) 202. The control unit 100 is communicatively connected to the host 300. The control unit 100 controls storage of data in the storage area 202. Further, the control unit 100 and the storage area 202 may be combined and regarded as a storage device or a storage system.

  The host 300 is an information processing apparatus such as a PC or a server that is used by a user (client) or accessed from a user apparatus. The host 300 inputs / outputs data to / from the storage area 202 by accessing the control unit 100.

  The control unit 100 includes a processor 101 and a control program 102 on a memory (not shown), and realizes a function as the control unit 100 by executing the control program 102 by the processor 101. The control unit 100 includes a shared memory 120. The shared memory 120 includes a control information memory 113 and, in particular, a cache memory (CM) 114 for temporarily storing input / output data for the storage area 202. . The control information memory 113 stores control information used by the control unit 100, and particularly stores a management information table 150 for characteristic processing of the present system.

  The storage area 202 is composed of storage resources such as HDD groups, and may exist in one storage device, or may exist separately in a plurality of storage devices. In the present system, the storage area 202 has a low-speed volume area (VL) 211 and a high-speed volume area (VH) 212. The high-speed volume area (VH) 212 is an area composed of a volume whose processing speed of data input / output access is high compared to the low-speed volume area (VL) 211. In each volume area (VL, VH), one or more volumes (storage volume or logical volume), a low speed volume (V1) 51 and a high speed volume (V2) 52, respectively, are secured. The control unit 100 can read and write data by accessing a volume group in the storage area 202.

  The entire area of the cache memory 114 is managed by being divided into two cache areas, a business cache area (referred to as CA1) 41 and a virtual volume cache area (referred to as CA2). The business cache area (CA1) 41 is a normal (as in the prior art) cache control area used for business processing and the like. On the other hand, the virtual volume cache area (CA2) 42 is an area that is set and secured exclusively for processing characteristic of the present system.

  In the virtual volume cache area (CA2) 42, one or more virtual volumes (V3) 53 are further set and secured. The virtual volume 53 is a volume on the shared memory 120 on the control unit 100 side corresponding to the volumes (V1, V2) on the storage area 202 side.

  Further, for example, in the low speed volume (V1) 51, a file 60 (for example, F1) handled by the host 300 is stored. The file 60 is associated with a logical unit (LU) handled by the control unit 100. The contents of the file 60 are updated when the control unit 100 reads / writes data according to the data input / output access from the host 300.

  As the data migration process (first process), as indicated by P1, the control unit 100 determines from the low-speed volume (V1) 51 in the low-speed volume area (VL) 211 to the high-speed volume area ( Processing for migrating the target data (file 60) to the high-speed volume (V2) 52 of (VH) 212 is performed. The unit of the migration data is the file 60.

  Further, as a characteristic process (second process), as indicated by P2, the control unit 100, particularly in the file (F1) 60 of the low-speed volume (V1) 51, is associated with the first process (P1). A process of transferring the data in the access area (update area) as the target data to the virtual volume (V3) 53 dynamically secured on the cache memory 114 is performed. The unit of the target data is a data block group. The data transferred to the virtual volume (V3) 53 is ready to respond to data input / output access from the host 300.

  After the migration of the file (F1) 60 to the high-speed volume (V2) 52 is completed, the update contents in the data on the virtual volume (V3) (data updated by access from the host 300) are changed to the high-speed volume (V2). 52 is reflected (transferred).

  In the storage system of the prior art example, when the first process (P1) is performed, the second process (P2) as described above is not performed. The data migration unit was a volume unit (not a file unit). In the storage device (control unit), a load is applied twice between the data migration processing between volumes and the access processing of the host to the volume, which affects the host access performance and the like. In addition, since it was a volume unit, it was migrated including unnecessary data, so it took extra time to migrate.

<System (1)>
In FIG. 2, as a first system configuration example according to the first embodiment, a configuration example assuming a DLCM in the storage apparatus 1 is shown. A configuration example of the storage area 202 is a case where volume areas (VL, VH) having different processing speeds are configured in one storage apparatus 1 (first storage apparatus 1A).

  The storage device unit 200 of the storage device 1 has a processing speed (such as a disk access speed of the HDD 30) such as a low-speed storage device unit (PL) 221, a medium-speed storage device unit (PM) 223, and a high-speed storage device unit (PH) 222. It has three different types of storage devices. The high-speed storage unit (PH) 222 is configured by a relatively high-speed storage device (physical device: PDEV) group, for example, an FC (fiber channel) -HDD 30 or the like. The low-speed storage device (PL) 221 is configured by a relatively low-speed storage device group, for example, a SATA (serial ATA) -HDD 30 or the like. Similarly, the medium-speed storage unit (PM) 223 is configured by a storage device group having an intermediate processing speed. The high speed storage unit (PH) 222 provides a high speed volume area (VH) 212, and the low speed storage unit (PL) 221 provides a low speed volume area (VL) 211.

  Then, data migration is performed from the low speed volume (V1) 51 in the low speed volume area (VL) 211 to the high speed volume (V2) 52 in the high speed volume area (VH) 212.

<Hardware configuration>
FIG. 3 shows an example of the overall configuration of the storage system and an example of the hardware configuration of the storage apparatus 1. The first storage device 1A is connected to the host 300. The second storage device 1B is externally connected to the first storage device 1A via the communication network 500. The communication network 500 is, for example, a storage area network using fiber channel switches. Similar to the first storage device 1A, the second storage device 1B is a device including the control unit 100, the storage device unit 200, and the like, and may have different functions and storage capacities. Further, the second storage device 1B may be directly connected to the first storage device 1A. In addition, the management terminal 400 is connected to the first storage device 1A via an SVP (service processor) 190.

  The host 300 includes an application program (AP) 301 group for business processing or the like, an OS, a storage use program 302 for using the storage device 1, an HBA (host bus adapter) 303 for connecting to the storage device 1, and the like. Have. For storage (input / output) of data used by the AP 301, the host 300 accesses the control unit 100 of the storage apparatus 1 by the storage use program 302 and inputs / outputs data to / from the file 60 (LU). The control unit 100 accesses (input / output) data in the shared memory 120 or the storage area 202 in response to an access from the host 300 (for example, a data read / write request / command), and returns a processing result to the host 300. To do.

  The first storage device 1A includes a control unit 100, a SW (connection control unit) 201, a storage device unit 200 (storage or disk array), and the like.

  The control unit 100 includes, for example, a CHA (channel adapter) 111, a DKA (disk adapter) 112, a SW (connection control unit) 115, a shared memory 120, an SVP 190, and the like. The shared memory 120 includes a control information memory 113 and a cache memory 114. The CHA 111 is a communication interface processing unit with the outside such as the host 300 and the second storage device 1B. The DKA 112 is a communication interface processing unit with the storage device unit 200 via the SW 201 (for example, a fiber channel switch). The SW 115 is, for example, a crossbar switch that controls interconnection between the processing units. The shared memory 120 is a memory that is shared and accessed within the control unit 100. The control information memory 113 stores control information handled by the control unit 100, management information (including the management information table 150), and the like. The cache memory 114 temporarily stores input / output data for the storage device unit 200.

  The control program 102 is executed by a processing unit in the control unit 100, for example, the CHA 111. The control program 102 includes a data migration (first process) control program, a virtualization control (second process) control program, and the like. The start and end of the execution of the process by the control program for the second process is accompanied by the start and end of the process by the control program for data migration (however, as will be described later, it is not limited to perfect match). Note that the control unit 100 is not limited to a form in which a plurality of functional processing units are interconnected as described above, and a form in which the whole is controlled by a CPU, for example, is also possible.

  In the control unit 100, for example, system configuration information, access log information 161, and the like are stored in the control information memory 113. The management information table 150 also includes setting information related to the first and second processes. The access log information 161 is information indicating the actual number of times of access and the order based on the record of the access processing result, and is managed and stored as in the past.

  The storage device unit 200 is composed of, for example, a group of HDDs 30 and is interconnected by the SW 201. A plurality of HDDs 30 can set a RAID group 35 corresponding to a predetermined RAID level. A volume or the like is set on the RAID group 35 and is associated with the storage area 202. Note that the storage device (physical device) is not limited to the HDD 30, and other storage devices such as a magnetic tape device can be used, and devices having different processing speeds can be used together.

  The management terminal (management server) 400 is an information processing apparatus that is used by a user (such as a system administrator) and performs processing such as management and setting for the storage system (including the first storage apparatus 1A). In particular, the management terminal 400 includes a Web browser, a management / setting program 401, and the like, and settings for the characteristic processing can be performed on a display screen (Web page) through a graphical user interface.

  The SVP 190 is an information processing apparatus that performs maintenance / management processing of the first storage apparatus 1A, and is connected to the inside or outside of the control unit 100. The SVP 190 is connected to each processing unit such as the CHA 111 of the control unit 100. It should be noted that the management terminal 400 and the SVP 190 may be arranged and connected in other places, or they may be integrated. Further, the host 300 may have a form in which the functions of the management terminal 400 and the SVP 190 are installed.

  The management / setting program 401 includes a setting program for data migration (first processing) and a setting program for virtualization control (second processing). The management / setting program 401 allows storage configuration settings, data migration settings, virtualization control settings and instructions, policy settings related to dynamic allocation of the virtual volume cache area (CA2) 42 and virtual volume 53, storage operation status, Various management and settings such as cache memory status and access performance are possible.

  The normal data input / output access process in the first storage device 1A is, for example, as follows. In response to a data write request from the host 300, the control unit 100 temporarily stores data received by the CHA 111 in the CM 114, writes control information in the control information memory 113, and responds to the host 300. On the other hand, the DKA 112 writes the data to the volume of the storage device unit 200 via the SW 201 based on the reference of the control information in the control information memory 113. Further, the control unit 100 refers to the control information in the control information memory 113 by the CHA 111 in response to a data read request from the host 300, and if there is target data in the CM 114, reads it and responds to the host 300. If not, the DKA 112 reads the target data from the volume of the storage device unit 200 via the SW 201 and stores it in the CM 114, and transmits and responds to the host 300 via the CHA 111.

<Memory structure>
FIG. 4 shows a storage storage structure corresponding to the system configuration of FIG. The host 300 (HBA 303) is connected to the CHA 111 port (target port 91) of the first storage device 1A. Further, the port of the CHA 111 (external port 92) of the first storage device 1A and the port of the second storage device 1B (target port 93) are connected.

  A logical unit 70 (for example, LU # 0, # 1) and a command device 71 for command control (“CMD”) are connected to the port (target port 91) of the first storage device 1A. A logical device 72 (for example, LDEV # 0, # 1, # 2) is associated with the logical unit 70 (LU # 0, # 1). In the first storage device 1A, a logical device 72 (for example, LDEV # 2) corresponds to a physical device (PDEV) of the storage device unit 200, for example, a RAID group 35 by the FC-HDD 30 as a high-speed storage device unit. Attached.

  When the storage resource in the external second storage device 1B is associated with the logical device 72 (for example, LDEV # 0, # 1) in the first storage device 1A, the virtual device ( VDEV) 73 is associated. For example, LDEV # 0 of LU # 0 in the first storage device 1A is associated with LDEV # 0 of LU # 0 in the second storage device 1B by VDEV # 0. The PDEV in the second storage device 1B is associated with, for example, a RAID group 35 by the SATA-HDD 30 as a low-speed storage device.

  Each volume (V1, V2) in the storage area 202 is associated with the LU 70. The file 60 is associated with the LU 70.

<Migration / transfer processing>
FIG. 5 shows a detailed processing example of data migration (first processing) and virtualization control (second processing). The file (F1) 60 is divided into an access area (X) and a non-access area (Y) related to the presence / absence of access (update) from the host 300. The access area (X) is an area (or its data) that is a target of data input / output access (a) from the host 300 or the like.

  In the second process (P2), the control unit 100 stores the data in the access area (X) of the file (F1) 60 in the low speed volume (V1) 51 in the virtual volume (V3) as indicated by S1. Forward to. In the first process (P1), the control unit 100 moves the data in the non-access area (Y) of the file (F1) 60 to the high-speed volume (V2) 52, as indicated by S2. In the second process (P2), the control unit 100 transfers the data (update data) of the virtual volume (V3) 53 to the high-speed volume (V2) 52 after the migration of S2, as indicated by S3. Reflect (transfer).

  The host 300 can perform data input / output access (b) to the data arranged in V3 of CA2. Therefore, access performance is ensured even during data migration. Specifically, when the CHA 111 receives an access request from the host 300, the CHA 111 refers to the shared memory 120, and the access target data exists in the virtual volume (V3) 53 in the CA2 of the cache memory 114. The processing result is returned to the host 300.

<Memory allocation>
FIG. 6 shows the allocation of the virtual volume cache area (CA2) 42 and LU 70 (virtual volume (V3) 53) in the cache memory 114. CA2 has a capacity (CA2 capacity) secured by a fixed value or a ratio value with respect to the capacity of the entire area of the cache memory 114. The cache memory 114, CA2, and virtual volume 53 have a hierarchical configuration, and the virtual volume 53 is secured as a secondary area for CA2. When CA2 and V3 are secured, a suitable capacity is automatically secured according to the capacity of the low speed volume (V1) 51 and the setting of the ratio value.

  The storage area 202, the shared memory 120, and the like are each configured in units of blocks (segments). The LU 70, the file (F1) 60, etc. are composed of data in units of a plurality of blocks (segments). In the virtual volume (V3) 53, the data of the access area (X) in the file (F1) 60 is stored in units of blocks. Note that storage in units of 60 files (including non-access areas (Y)) can be performed for CA2 or V3.

  The LU 70 that is the virtual volume (V3) 53, for example, “LU011” (LUN (LU number) is “011”) is assigned from the address (“0001”) of the first block in the entire area of CA2.

<Overall processing flow>
Next, FIG. 7 shows a flow of characteristic processing (data migration and virtualization control) in this system. In this example, in the first process (P1), as a trigger for starting, the number of times of access to the file (F1) 60 arranged in the low speed volume (V1) 51 is set as a threshold (set by the user). When it exceeds, the data (F1) is transferred to the high-speed volume (V2) 52. Along with this, the second process (P2) is started, and access to the data of the low-speed volume (V1) is secured by the data transferred on the virtual volume (V3) 53, and the high-speed processing is started. The data migration to the volume (V2) 52 is completed.

  In S100 of FIG. 7 (S represents a processing step), it is assumed that the host 300 first accesses the low-speed volume (V1) 51, particularly the file (F1) 60 therein. Hereinafter, the processing subject is mainly the control unit 100 (control program 102).

  In S101, the control unit 100 performs a predetermined threshold value confirmation process in accessing the file (F1) 60 of the low speed volume (V1) 51. In S101A, it is determined whether or not the number of accesses to the data exceeds a threshold value. If not (N), normal access or the like is continued. If it exceeds (Y), the process proceeds to S102. This threshold value confirmation process (S101) is a process for determining the trigger (timing) of execution of data migration and virtualization control by comparison with a threshold value. The threshold value confirmation processing in S101 is realized by using the access log information 161 and comparing with the threshold value.

  In S102, a virtual volume cache area (CA2) 42 and a virtual volume (V3) 53 are created (secured) (FIG. 8).

  In S103, processing is performed to transfer the data in the access area (X) of the file (F1) 60 accessed (updated) of the low-speed volume (V1) 51 onto the virtual volume (V3) 53 of CA2 as target data. (S1 in FIG. 5). The transferred data is ready for data input / output processing.

  In S104, a process of moving the data in the non-access area (Y) of the data migration target file (F1) 60 to the high-speed volume (V2) 52 is performed (S2 in FIG. 5).

  In S105, the data transferred to the virtual volume (V3) 53 (update data corresponding to X), that is, data whose contents are updated by data input / output access from the host 300 or the like (may not be updated). Then, the process of reflecting (transferring) to the high-speed volume (V2) is performed (S3 in FIG. 5).

  In S106, the process of deleting the old data (X, Y) of the file (F1) 60 of the low speed volume (V1) 51 is performed.

  In S107, processing for deleting the update data on the virtual volume (V3) 53 is performed based on policy setting (usage time, etc.). In addition, the virtual volume (V3) 53 secured this time is deleted. Along with the deletion of V3, the virtual volume cache area (CA2) 42 is deleted. That is, as a part of the function of dynamically securing the first area (CA2, V3), a process of sequentially erasing the hierarchical configuration V3 and CA2 is performed. It should be noted that it is possible not to delete CA2 depending on the policy setting or the situation (for example, when a plurality of LUs 70 are continuously secured in CA2).

  In S108, as ancillary processing, defragmentation processing (storage area arrangement) for the low-speed volume (V1) 51 is performed.

<Process (S102)>
FIG. 8 shows details of the processing of S102 (dynamic securing of CA2 and V3).

  In S201, the control unit 100 creates (dynamically secures) a virtual volume cache area (CA2) 42 in the cache memory (CM) 114 area. For this purpose, a setting value (depending on a fixed value or a ratio value) relating to the CA2 capacity, which is selected and set by the user from the management terminal 400 or the like using a Web browser or the like, is recognized. That is, by this, the capacity to be allocated as CA2 is once determined.

  Next, in S202, the control unit 100 confirms whether or not there is a free capacity for allocating CA2 (the capacity recognized in S201) in the area of the CM 114. If the free space exists (Y), CA2 is allocated to the free space area of the CM 114 in S202A. If there is no free space (N), in S202B, if there is an auxiliary storage device prepared separately from the shared memory 120, it is expanded by address mapping in the memory area of the auxiliary storage device. As a result, CA2 is allocated over the memory area of the CM 114 and its auxiliary storage device. When there is no auxiliary storage device, for example, CA2 is created only in the CM 114, or CA2 is not created at that time.

  Next, in S203, the setting value related to the virtual volume (V3) 53, which is selected and set by the user from the management terminal 400 or the like using a Web browser or the like, is recognized, and V3 is allocated in CA2. The set value is a threshold value related to the capacity (size) of V3, the usage time (remaining time), and the like.

  The second process (P2), such as securing the first area (CA2, V3) and transferring the target data, is not performed in all cases along with the data migration (P1), but for policy setting (threshold), etc. It is made according to the judgment based on. Further, the CA2 and the like are secured dynamically according to the data to be arranged in the area with respect to the settings such as capacity and usage time. Therefore, it is avoided that the extra capacity is consumed and the CA1 and the CM 114 are compressed.

<Process (S103 to S105)>
FIG. 9 shows a detailed example of S103 to S105 corresponding to FIG. 5 (S1 to S3). The unit of data migration from V1 to V2 is the file (F1) 60, and the unit of transfer from V1 to V3 is the data in the access area (X) in the file (F1) 60.

  In S301, the control unit 100 (for example, the CHA 111) first converts the data of only the access area (X: especially X1) of the file (F1) 60 of the low-speed volume (V1) 53 into the virtual volume (V3) 53 of CA2. Copy to.

  In S302, the control unit 100 (for example, DKA 112), as background processing (parallel processing) with respect to the processing in S301 and the like, uses only the non-access area (Y) of the file (F1) 60 as the high-speed volume (V2). Copy to 52.

  The data block (X3) transferred and arranged on the virtual volume (V3) 53 is in a state where data input / output access from the host 300 is possible. Thereby, for example, when data writing occurs, the content of the data (X3) is updated accordingly.

  In S303, the control unit 100 (for example, the CHA 111) updates the data (X3) on the virtual volume (V3) 53 after the transfer of the data in the non-access area (Y) to the high-speed volume (V2) 52 in S302 is completed. Is copied (or moved) to the high-speed volume (V2) 52. As a result, the data of the file (F1) 60 of the low-speed volume (V1) 51 is transferred onto the high-speed volume (V2) 52.

  Note that all operations such as the transfer of data between the volumes are copy operations. Each copy source data is erased in accordance with policy setting or predetermined timing. Further, in the case of copying from V3 to V2 in S303, the data (X3) remains in a state where an access response can be made as it is, and is updated and reflected to V2 after a predetermined non-access time elapses according to policy settings, for example. . On the other hand, when moving from V3 to V2 in S303, the data (X3) does not remain on CA2. Further, for example, other detailed processing such as copying data (X and Y) of the entire file (F1) 60 in S302 is possible.

<Management information (1)>
Next, an example of the management information table 150 corresponding to the above configuration will be described. The storage location of the management information table 150 is a shared memory in the storage system (such as in the storage apparatus 1 or the host apparatus 300). In this example, the management information table 150 is arranged on the control information memory 113 of the shared memory 120 in the control unit 100 of the first storage device 1A.

  FIG. 10 shows a table 151 of the update file access tag setting, the access number threshold tag definition, and the access ranking threshold tag definition of the low-speed volume (V1) 51. This table 151 shows the setting for the threshold value confirmation process (S101) for the data area of each address of the update file (F1) 60 of V1.

  Based on this table 151, the control unit 100 relates to the transition of F1 from V1 to V2 and the process of transferring the data in the access area (X) of F1 to V3 accordingly, for each address of the file (F1) 60. The trigger (timing) of execution of the process is determined by comparing the number of accesses and the access ranking (overall ranking) at that time with respect to data with threshold values (access number threshold value, access ranking threshold value). The number of accesses, the ranking (count value), etc. at that time can be understood from the reference to the access log information 161.

  In the update file access tag setting (T1), items include an update file address number and an update file access tag. The address number of the update file indicates an address number of a block or the like in the update file (F1) 60. The update file access tag indicates a tag value defined in another table. For example, the threshold value “100” of the number of accesses indicated by the tag value “0” is applied to the data area of the update file (F1) with the address number “0111”.

  The access number threshold value tag definition indicates the threshold value of the access number that is distinguished by the tag value. With respect to the access number threshold tag, when access exceeding the defined access number (for example, 100) occurs, transfer of data in the access area (X) of the target file (F1) 60 is started. . Similarly, the access ranking threshold tag definition indicates an access ranking (overall rank) threshold that is distinguished by a tag value. For this access ranking threshold tag, similarly, the transfer of the target data is started when an access with a defined ranking (for example, 10th) or higher occurs.

<Management information (2)>
FIG. 11 shows a virtual volume cache area (CA2) management table 152 corresponding to FIG. The items of this table 152 include the segment number of the virtual volume cache area (CA2) 42, the segment number of the cache area (CM 114), the flag of the virtual volume (V3) 53, the LUN (LU number) of V3, and the CA2 capacity tag. . This table 152 is for managing CA1 and CA2 in the CM 114.

  The segment number of CA2 indicates an address of a segment (block) unit on the basis of CA2. The segment number of the CM 114 indicates an address in segment (block) units in the entire CM 114. The V3 flag indicates whether or not V3 is set, and is “1” when V3 is assigned to the segment of CA2, and “0” otherwise. When the flag is “0” and V3 is not assigned, the area can be assigned as CA1. The LUN of V3 indicates the number of the LU 70 associated with the V3. The CA2 capacity tag indicates the tag value as shown in FIG.

<Management information (3)>
FIG. 12 shows a table 153 of CA2 capacity tag settings and each capacity tag definition. A table 153 of tag setting (CA2 capacity tag setting) and tag definition (fixed value capacity tag definition, ratio value capacity tag definition) for adjusting the capacity in relation to CA2 creation is provided.

  In the CA2 capacity tag setting, the CA2 segment number is associated with the CA2 capacity tag for setting the capacity. The CA2 capacity tag can be selected between a fixed value and a ratio value according to each capacity tag definition. Setting with a fixed value secures the CA2 capacity with a fixed value (for example, in units of 10 MB) with respect to the entire capacity of the cache area (CM 114), and is effective when priority is given to securing the capacity of CA2. The ratio value setting secures the capacity of CA2 in a ratio (for example, in units of 10%) with respect to the entire capacity of the existing cache area (CM 114), and a capacity according to the status of business processing or the like is secured. .

<Management information (4)>
In FIG. 13, a virtual volume (V3) management table 154 is shown. In this table 154, items include the LUN of the virtual volume (V3), the segment number of the virtual volume cache area (CA2), the virtual volume (V3) non-access tag, and the virtual volume (V3) capacity tag. The V3 non-access tag is a tag indicating a non-access area (such as the non-access area (Y) of the file (F1) 60) in V3, and its tag setting and definition are shown in a table 155 in FIG. The V3 capacity tag is a tag indicating the allocated capacity of V3, and the tag setting and definition are shown in the table 156 of FIG. The tables of FIGS. 14 and 15 are tables for setting and defining tags for adjusting the capacity and usage time in connection with V3 creation.

  As for the LUN number of V3 and the segment number of CA2, as shown in FIG. 6, the CA2 segment (eg, numbers “0001”, “0002”,...) Is assigned to the V3 LU (eg, LUN “011”). .

<Management information (5)>
FIG. 14 shows a table 155 for V3 non-access tag setting and definition. In the V3 non-access tag setting table, a non-access tag is set for the LUN of V3. In the V3 non-access tag definition table, a non-access period (for example, one day unit) is defined for a tag. That is, the non-access tag of V3 defines the time (threshold value) in order to erase the virtual volume (V3) 53 that has not been accessed for a predetermined time or longer (corresponding to the remaining time). For example, when the tag value “0” is set, V3 that is not accessed for one day is deleted. In particular, an example of the erase determination part is shown on the lower side of FIG.

<Management information (6)>
In FIG. 15, a V3 capacity tag setting and definition table 156 is shown. In the V3 capacity tag setting table, a capacity tag is set for the LUN of V3. In the V3 capacity tag definition table, the capacity of V3 (for example, 10 MB unit) is defined for the tag. The capacity tag can be set as a capacity previously selected by the user in order to create V3.

  The lower side of FIG. 13 shows management information for erasure processing of the virtual volume 53 (LU 70) associated with the V3 management table 154, and an example of erasure processing determination. For example, as the management information, the LU 70 corresponding to the virtual volume (V3) has non-access days (actual value) (: A) and non-access days (threshold) (: B). The non-access days is the number of days that have passed since access was lost.

  For example, for the LU 70 whose LUN of V3 is “110”, segment numbers “0100” to “0150” are set (assigned) as continuous segments in order to set the capacity to 30 MB (capacity tag is “2”), for example. ) For the LU 70, the number of non-access days (B) is set to 11 days (non-access tag is “10”), for example. That is, the LU 70 is set to be deleted if there is no access for 11 days. If A ≧ B as described above as the erasure processing determination result, the LU 70 of the V3 (for example, LUN is “110”) is erased. If A <B, the LU 70 remains.

<Management information (7)>
FIG. 16 shows a low-speed volume (V1) update file management table 157 corresponding to the processing of FIG. In this table 157, the update file address number of the low-speed volume (V1) 51, the access area file flag on the virtual volume (V3) 53, the address number of the access area file on V3, and the high-speed volume (V2) 52 Non-access area file flag, and the address number of the non-access area file on V2.

  The update file address number of V1 indicates the address number of the segment (block) of the access area (X) and non-access area (Y) of the update file 60 (for example, F1). The access area file flag and address number on V3 indicate whether or not there is data (X3) in the access area (X) of the file 60 in V3 (for example, “1” exists), and the address number. The non-access area file flag and address number on V2 indicate the presence / absence of data in the non-access area (Y) of the file 60 in V2 (for example, “1” exists), and the address number.

<Display screen>
Next, FIGS. 17 and 18 are examples of display screens using the Web browser of the management terminal 400 operated by a system administrator or the like, in particular, in the case of threshold setting by the number of accesses and access ranking as an example of policy setting. Is shown.

  FIG. 17 shows a setting screen relating to the first processing (data migration processing from the low speed volume (V1) 51 to the high speed volume (V2) 52). For example, a migration source low speed volume list for selecting a migration source volume and a migration destination candidate high speed volume list for selecting a migration destination volume are shown. The migration source low-speed volume list has items such as a selection button, LUN, emulation type, total capacity, used capacity, RAID level, and storage device number. The migration destination candidate high-speed volume list has items such as a selection button, LUN, status (used (USED), unused (FREE), etc.), emulation type, capacity, RAID level, and storage device number. The user can freely select and set a volume (LU 70) to be processed.

  FIG. 18 shows a setting screen relating to the second processing (processing such as data transfer from the low-speed volume (V1) 51 to the virtual volume (V3) 53). For example, a low-speed volume list relating to the setting of V1, an access number threshold or ranking threshold selection button, an access number threshold setting target volume list, and an access ranking threshold setting target volume list are shown. In addition to the setting using the tag, the user can set in detail for each volume (LU 70).

  The low-speed volume list has items such as LUN, emulation type, capacity, RAID level, and storage device number. The type of threshold to be applied can be selected with a selection button. The access number threshold setting target volume list includes items such as LUN, emulation type, RAID level, and access threshold setting (IOPS: number of inputs / outputs per second). In the access threshold setting, for example, when “100” is set, the second process (P2) or the like is executed when the IOPS of access to the LU 70 exceeds 100. Further, the access ranking threshold setting target volume list includes items such as LUN, emulation type, RAID level, ranking threshold setting (lower limit value), and the like. In the ranking threshold setting, for example, when “10th” is set, the second process (P2) or the like is executed when the access ranking for the LU 70 becomes 10th or higher.

  As described above, according to the first embodiment, DLCM can be accelerated by data migration, and dynamic reservation of the first area (CA2, V3) is used, and data migration is in progress. Access performance such as data input / output access from the host 300 or the like can be ensured. Further, the storage apparatus 1 (control unit 100) can separate the load between the data migration processing of the file 60 between the volumes and the access processing of the host 300 or the like with respect to the file 60 of the volume, and can secure access performance and the like. Further, the data migration unit is not the volume but the file 60 unit, and the transfer unit to the first area is the data block unit of the access area (X) in the file 60, thereby reducing the extra data transfer and the migration time. The processing efficiency is improved.

(Embodiment 2)
Next, the storage system according to the second embodiment of the present invention will be described with reference to FIG. In FIG. 19, as a second system configuration example according to the second embodiment, a configuration assuming a DLCM using the second storage device 1B externally connected to the first storage device 1A is shown. This is a case where two types of storage resources having different processing speeds exist between the storage apparatuses 1. The hardware configuration of the storage device 1 is the same as described above.

  The first storage device 1A has only a high-speed storage device (PH) 222 and a high-speed volume area (VH) 212 thereby, and the second storage device 1B has a low-speed storage device (PL) 221 and Accordingly, only the low-speed volume area (VL) 211 is provided. For example, the configuration corresponds to FIG. The VL of the second storage device 1B has a low-speed volume (V1) 51 as a data migration source, and the VH of the first storage device 1A has a high-speed volume (V2) 52 as a data migration destination. . Further, a virtual volume cache area (CA2) 42, a virtual volume (V3) 53, and the like are secured on the cache memory 114 in the shared memory 120 of the control unit 100 of the first storage device 1A.

  Thus, even when data migration (P1) is performed between the first storage device 1A and the externally connected second storage device 1B, the second processing (P2) using CA2 and V3 is performed. As a result, the access performance of the host 300 or the like can be secured.

(Other embodiments)
The following are possible as other embodiments of the present invention.

  The first process (P1) is not limited to the case of data migration from the low-speed volume (V1) 51 corresponding to the increase in access to the high-speed volume (V2) 52, but the high-speed volume (V2) 52 corresponding to the decrease in access or the like. Similarly, a configuration in which the second process (P2) is performed in the case of data transfer from the low-speed volume (V1) 51 to the low-speed volume (V1) 51 is also possible.

  As the second process (P2), regarding the dynamic securing of the first area (CA2, V3), for example, only CA2 is provided for the shared memory 120 according to the policy setting, the situation, etc., and only V3 is provided. It is good also as providing, an area | region smaller than V3, etc. In addition, prior to the start of data migration, for example, CA2 may be secured. That is, CA2 having a certain capacity is estimated and secured in advance, and V3 is dynamically secured on the estimated CA2. Further, regarding the deletion of V3 and CA2, for example, it is not limited to being left based on the setting of the number of non-access days, but may be configured to be deleted immediately upon completion of data migration according to policy setting or the like.

  Further, for example, when determining the number of accesses for the second process (P2), a configuration may be adopted in which only access from the host 300 is targeted, not all unspecified access to data.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

1 is a diagram showing a schematic configuration of a storage system according to an embodiment of the present invention. It is a figure which shows the structure of the storage system which is Embodiment 1 of this invention. It is a figure which shows the structure of the storage system of Embodiment 1 of this invention, and the hardware constitutions of a storage apparatus. It is a figure which shows the memory structure in the storage system of Embodiment 1 of this invention. It is a figure which shows the detail of the 1st process in the storage system of Embodiment 1 of this invention, and a 2nd process. FIG. 3 is a diagram showing a configuration of cache memory allocation and the like in the storage system according to the first embodiment of the present invention. It is a figure which shows the flow of the whole process in the storage system of Embodiment 1 of this invention. It is a figure which shows the flow of a production | generation process (S102) of the 1st area | region (CA2, V3) in the storage system of Embodiment 1 of this invention. It is a figure which shows the flow of the detail (S103-S105) of the 1st process in the storage system of Embodiment 1 of this invention, and a 2nd process. It is a figure which shows the example (the 1) of the management information table in the storage system of Embodiment 1 of this invention. It is a figure which shows the example (the 2) of the management information table in the storage system of Embodiment 1 of this invention. It is a figure which shows the example (the 3) of the management information table in the storage system of Embodiment 1 of this invention. It is a figure which shows the example (the 4) of the management information table in the storage system of Embodiment 1 of this invention. It is a figure which shows the example (the 5) of the management information table in the storage system of Embodiment 1 of this invention. It is a figure which shows the example (the 6) of the management information table in the storage system of Embodiment 1 of this invention. It is a figure which shows the example (the 7) of the management information table in the storage system of Embodiment 1 of this invention. It is a figure which shows the example (the 1) of the display screen for the setting in the storage system of Embodiment 1 of this invention. It is a figure which shows the example (the 2) of the display screen for the setting in the storage system of Embodiment 1 of this invention. It is a figure which shows the structure of the storage system which is Embodiment 2 of this invention.

Explanation of symbols

  1 (1A, 1B) ... Storage device, 30 ... HDD, 35 ... RAID group, 41 ... Business cache area (CA1), 42 ... Virtual volume cache area (CA2), 51 ... Low speed volume (V1), 52 ... High speed volume (V2), 53 ... Virtual volume (V3), 60 ... File, 70 ... Logical unit (LU), 71 ... Command device, 72 ... Logical device (LDEV), 73 ... Virtual device (VDEV), 91-93 ... Port , 100 ... control unit (controller), 101 ... processor, 102 ... control program, 111 ... CHA (channel adapter), 112 ... DKA (disk adapter), 113 ... control information memory, 114 ... cache memory (CM), 115 ... Connection control unit (SW), 120 ... shared memory, 150 (15 157) Management information table 161 Access log information 190 SVP (service processor) 200 Storage unit 201 Connection controller SW 202 Storage area 211 Low-speed volume area (VL), 212 ... high speed volume area (VH), 213 ... medium speed volume area (VM), 221 ... low speed storage device section (PL), 222 ... high speed storage device section (PH), 223 ... medium speed storage device section (PM), 300 ... host (host device), 301 ... application program (AP), 302 ... storage use program, 303 ... HBA (host bus adapter), 400 ... management terminal, 401 ... management / setting program, 500 ... communication network.

Claims (10)

A storage system including a storage device connected to a host device,
The storage device includes a storage device unit that includes a plurality of storage devices and provides a storage area, and a control unit that controls storage of data in the storage area,
The storage area has a first volume and a second volume for storing data;
The control unit includes a shared memory or cache memory in the control unit, a communication interface processing unit with the outside including the host device, and a communication interface processing unit with the storage device unit,
The controller is
A first process for controlling data migration from the first volume to the second volume according to a predetermined condition;
In response to the first processing, the target data corresponding to the migrated data in the first volume is transferred to the first area dynamically secured on the shared memory or the cache memory, and an access response is made. A storage system characterized by performing at least a second process that ensures at least access performance to the target data during the migration by making it possible and reflecting it in the second volume. The storage system according to claim 1, wherein
A first storage device connected to the host device;
The first storage device includes the first volume having a relatively low access processing speed and the second volume having a relatively high speed.
In the first process, data migration is performed between the low-speed first volume and the high-speed second volume,
In the second processing, the first area is dynamically secured in the shared memory or cache memory of the control unit of the first storage device. The storage system according to claim 1, wherein
A first storage device connected to the host device, and a second storage device externally connected to the first storage device,
The first storage device has the second volume having a relatively high access processing speed, and the first storage device has a relatively low access processing speed in the first storage device. Have a volume of
The first area is dynamically secured in the shared memory or cache memory of the control unit of the first storage device,
In the first process, data migration is performed between the low-speed first volume and the high-speed second volume,
In the second processing, the first area is dynamically secured in the shared memory or cache memory of the control unit of the first storage device. The storage system according to claim 1, wherein
The control unit has a cache memory used for data cache control as the shared memory,
The control unit includes a first cache area used for normal data cache control including business processing by the host device, and the first area for the second process, with respect to the entire area of the cache memory. And separately manage the second cache area,
The capacity of the second cache area is secured by a fixed value or a ratio value with respect to the capacity of the cache memory. The storage system according to claim 1, wherein
The control unit has a cache memory used for data cache control as the shared memory,
The control unit includes a first cache area used for normal data cache control including business processing by the host device, and the first area for the second process, with respect to the entire area of the cache memory. And separately manage the second cache area,
In the second process, in the first area, one or more third volumes corresponding to the first volume are secured in the second cache area, and the target data is stored in the third volume. A storage system, wherein The storage system according to claim 1, wherein
The unit of data targeted by the first process is the file data in the first volume,
A storage system characterized in that a unit of data to be transferred to the first area in the second processing is data in an access area in the file. The storage system according to claim 1, wherein
The controller is
A process of copying first data, which is data in an access area of a file in the first volume, to the first area;
In parallel with the above process, a process of copying second data, which is data in a non-access area of a file in the first volume, to the second volume;
A process of copying the third data having update possibility corresponding to the first data in the first area to the second volume;
A process of erasing old data of the file in the first volume;
A process of erasing old data in the first area immediately after the process of copying the third data to the second volume or after the elapse of a predetermined non-access time. system. The storage system according to claim 1, wherein
The control unit executes the first process, the target data, and the execution of the second process based on a result of comparing the number of accesses or ranking of the data in the first volume with a threshold value. And determining the target data. The storage system according to claim 8, wherein
In the information processing apparatus connected to the storage apparatus, a policy setting including setting of the threshold value related to the first and second processes is performed by a user on a display screen. A data migration method in a storage system including a storage device connected to a host device,
The storage device includes a storage device unit that includes a plurality of storage devices and provides a storage area, and a control unit that controls storage of data in the storage area,
The storage area has a first volume and a second volume for storing data;
The control unit includes a shared memory or cache memory in the control unit, a communication interface processing unit with the outside including the host device, and a communication interface processing unit with the storage device unit,
The controller is
A first process for controlling data migration from the first volume to the second volume according to a predetermined condition;
In response to the first processing, the target data corresponding to the migrated data in the first volume is transferred to the first area dynamically secured on the shared memory or the cache memory, and an access response is made. And a second process for ensuring at least the access performance to the target data during the migration by making it possible and reflecting it in the second volume,
A processing step of dynamically securing the first area with the start of the transition;
A processing step of transferring the target data to the first area;
Processing to transfer and reflect the target data to the second volume;
And a processing step of erasing the target data or the first area after the migration.

Images