JP6291937B2 - Block storage, storage system, computer system, block storage control method, storage system control method and program - Google Patents

Description

  The present invention relates to storage, and more particularly to hierarchical storage.

  A technology is known that provides a high-performance storage while suppressing costs by arranging a plurality of physical devices having different performances in a hierarchy and arranging data according to access frequency. An example of such a technique is described in Patent Document 1.

  The storage system described in Patent Document 1 includes a plurality of storage devices with different performances, and associates logical volumes with these storage devices based on access frequency.

International Publication No. WO2013 / 103003A1

  In the storage system described in Patent Document 1 described above, there is a problem that data having a capacity exceeding the specifications of the storage device included in the storage system cannot be stored.

  The main object of the present invention is to solve the above-mentioned problems and to provide a technique for realizing a high-performance, low-cost and large-capacity storage.

The block storage according to the present invention that achieves the above object is as follows.
Objectification aggregation means for aggregating a predetermined number of blocks whose access frequency is lower than a threshold, and transmitting the aggregated blocks to an object storage;
Tier switching means for deleting the block transmitted to the object storage from the block storage.

The storage system according to the present invention that achieves the above object provides:
The block storage described above,
And an object storage for storing the data received from the block storage as an object.

The computer system according to the present invention that achieves the above-described object provides:
The above storage system;
A host computer capable of accessing the storage system.

A block storage control method according to the present invention that achieves the above object is as follows.
Aggregate a predetermined number of blocks in the block storage whose access frequency is lower than the threshold,
Sending the aggregated block to object storage;
The block transmitted to the object storage is deleted from the block storage.

A program according to the present invention that achieves the above object is as follows:
A process of aggregating a predetermined number of blocks of block storage whose access frequency is lower than the threshold;
Processing to send the aggregated block to object storage;
And causing the computer to execute a process of deleting the block transmitted to the object storage from the block storage.

  According to the present invention, it is possible to increase the storage capacity while suppressing a decrease in performance and an increase in cost.

1 is a block diagram illustrating a configuration of a storage system according to a first embodiment of the present invention. 5 is a diagram illustrating an example of contents of an LD mapping table 300. FIG. It is a figure which shows an example of the content of the hierarchy management table. 5 is a diagram illustrating an example of contents of a hierarchy 0 mapping table 320. FIG. It is a figure which shows an example of the content of the hierarchy 1 mapping table. It is a figure which shows an example of the content of the hierarchy 2 mapping table 322. 5 is a diagram illustrating an example of contents of an object management table 330. FIG. It is a figure which shows an example of the content of the object aggregation table. 5 is a diagram illustrating an example of contents of a policy management table 350. FIG. 5 is a flowchart showing the operation of an access frequency update unit 200. 5 is a flowchart showing an operation of rearranging data with low access frequency to the object storage 2; It is a flowchart showing operation | movement in case reading is performed to the address made into an object. It is a flowchart showing operation | movement at the time of writing in the address made into an object. 5 is a flowchart showing the operation of an object deletion unit 230. It is a flowchart showing operation | movement in case the object storage management part 240 stores an object. It is a flowchart showing operation | movement in case the object storage management part 240 deletes a target object. It is a flowchart showing operation | movement when the hierarchy switching part 220 receives a deletion instruction | indication. It is a flowchart showing operation | movement when the hierarchy switching part 220 receives the replacement instruction | indication. It is a block diagram showing the structure of the block storage which concerns on the 2nd Embodiment of this invention. And FIG. 11 is a block diagram illustrating an example of elements constituting a computer.

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

[First Embodiment]
FIG. 1 is a block diagram showing the configuration of the storage system according to the first embodiment of the present invention. The storage system SS according to the first embodiment of the present invention includes a block storage 1 and an object storage 2 connected to the block storage 1 via a WAN (Wide Area Network). The block storage 1 is a storage device that provides virtualized access in units of blocks.
The block storage 1 is connected to the virtual server 100 by FC (Fiber Channel), iSCSI (Internet Small Computer System Interface) or SAS (Serial Attached SCSI). The object storage 2 may be an object storage service provided by another vendor such as Amazon S3 (Amazon Simple Storage Service) (registered trademark).

  The block storage 1 includes a hierarchical logical disk 50, an access frequency update unit 200, an objectification aggregation unit 210, a hierarchy switching unit 220, an object deletion unit 230, and a policy management table 350. The hierarchical logical disk 50 is a logical disk that is hierarchized by combining physical devices and object storages having different performances. The access frequency update unit 200 periodically updates data access from the virtual server 100 and instructs various operations. The objectification aggregation unit 210 aggregates blocks whose access frequency is lower than a predetermined threshold value, and transmits the blocks to an object storage management unit 240 (described later). The hierarchy replacing unit 220 replaces actual data. The object deletion unit 230 issues an instruction to delete the dirty object. The policy management table 350 stores settings for the hierarchical logical disk 50.

  The object storage 2 includes an object storage destination device 70 and an object storage management unit 240. The object storage destination device 70 is a physical device that stores objects. The object storage management unit 240 converts the actual data received from the objectification aggregation unit 210 into an object, and stores the objectized data in the object storage destination device 70.

  The virtual server 100 includes guest OSs (Operating Systems) 101-102. The virtual server 100 is sometimes called a host computer. The virtual server and the storage system SS may be collectively referred to as a computer system.

  The hierarchical logical disk 50 includes an LD (Logical Disk) mapping table 300, a hierarchy management table 310, a hierarchy 0 mapping table 320, a hierarchy 1 mapping table 321, a hierarchy 2 mapping table 322, an object management table 330, an object An aggregation table 340 and a hierarchical device 60 are included. The LD mapping table 300 stores the primary address of the hierarchical logical disk. The hierarchy management table 310 stores the hierarchy type. The tier 0 mapping table 320, the tier 1 mapping table 321 and the tier 2 mapping table 322 store secondary addresses. The object management table 330 stores management information converted into an object. The object aggregation table 340 stores addresses of blocks aggregated as objects. The hierarchical device 60 stores actual data written to the hierarchical logical disk 50. In the block storage 1 according to the first embodiment, the hierarchical device 60 is an SSD (Solid State Drive), a SAS, and an Object (object storage). In the following description, “SAS” represents a SAS hard disk drive.

  The access frequency update unit 200 has a function of referring to the policy management table 350 and periodically instructing operation according to the access frequency update interval, objectization interval, layer replacement interval, and object deletion interval of each hierarchical logical disk. Further, the access frequency update unit 200 has a function of updating the access frequency of the LD mapping table at a timing based on the access frequency update interval.

  The objectification aggregation unit 210 acquires the aggregation number by referring to the policy management table 350 in accordance with an operation instruction from the access frequency update unit 200. The objectification aggregation unit 210 has a function of referring to the access frequency of the LD mapping table 300 and extracting data to be converted into an object based on the number of aggregations acquired from the policy management table 350. The objectification aggregation unit 210 has a function of transmitting the extracted actual data to the object storage management unit 240. The objectification aggregation unit 210 has a function of receiving metadata from the object storage management unit 240. The objectization aggregation unit 210 has a function of storing management information of the aggregated data in the object management table 330 and the object aggregation table 340. In addition, the objectification aggregation unit 210 has a function of instructing the hierarchy replacement unit 220 to perform replacement when deleting objected real data.

  The hierarchy switching unit 220 has a function of moving actual data between the hierarchy devices 60 in accordance with an operation instruction from the access frequency update unit 200 and then updating the LD mapping table 300 and the hierarchy mapping table. Further, the hierarchy switching unit 220 has a function of updating the LD mapping table 300 and the hierarchy mapping table in accordance with the replacement instruction from the objectification aggregation unit 210. The hierarchical mapping table refers to the hierarchical 0 mapping table 320, the hierarchical 1 mapping table 321, and the hierarchical 2 mapping table 322. Each hierarchy mapping table indicates whether the hierarchy address of each hierarchy is in use (valid).

  The object deletion unit 230 has a function of issuing an object deletion instruction to the object storage management unit 240 in accordance with an operation instruction from the access frequency update unit 200. The object deletion unit 230 has a function of updating the object management table 330 and the object aggregation table 340.

The object storage management unit 240 has a function of generating an object by performing objectification by assigning metadata and an object ID to the actual data received from the objectification aggregation unit 210. The object storage management unit 240 has a function of storing objectized data in the object storage destination device 70 and returning metadata to the objectification aggregation unit 210. The object storage management unit 240 has a function of deleting a stored object in response to an object deletion instruction from the object deletion unit 230.

  The LD mapping table 300 is an address table that is referred to when accessing the hierarchical logical disk 50 from the virtual server 100. FIG. 2 is a diagram showing an example of the contents of the LD mapping table 300. In the LD mapping table 300 shown in FIG. 2, a logical address, a hierarchical address, a hierarchical number, an OBJ (Object) flag, and an access frequency are recorded in association with each other. The access frequency represents the frequency of access to the address. As for the access frequency, 0 is the lowest, and the access frequency is higher as the numerical value is larger. The access frequency may be obtained, for example, by the access frequency update unit 200 summing up the frequency of access from the virtual server 100 to the address at regular intervals. When the value of the OBJ flag is “1”, it indicates that the data is made into an object. When the value is “0”, it means that the data is not made into an object (actual data). Represent. For example, in the second row of the LD mapping table 300 shown in FIG. 2, the logical address is “00000010”, the hierarchical address is “00000010”, the hierarchical number is “1”, the OBJ flag is “0”, and the access frequency is “3”. It is. This indicates that the actual data is stored at the 00000010 address of the hierarchy 1 and the access frequency is 3 (medium frequency access).

  The hierarchy management table 310 is a table representing physical devices assigned to the hierarchy. FIG. 3 is a diagram showing an example of the contents of the hierarchy management table 310. In the hierarchy management table 310 shown in FIG. 3, a hierarchy number and a hierarchy device are recorded in association with each other. Here, the values “SSD” and “SAS” of the hierarchical device indicate that these physical devices are allocated to this hierarchical level. Further, the value “OBJ” of the hierarchical device represents that the object storage is allocated to this hierarchical level. For example, in the first row of the hierarchy management table 310 shown in FIG. 3, the hierarchy number is “0” and the hierarchy device is “SSD”. This indicates that the SSD is assigned to the hierarchy 0.

  The tier 0 mapping table 320 is an address table that is referred to when accessing the tier 0 tier device 60 from the LD mapping table 300. FIG. 4 is a diagram showing an example of the contents of the hierarchy 0 mapping table 320. In the layer 0 mapping table 320 shown in FIG. 4, the layer 0 layer address and the valid flag are recorded in association with each other. For example, in the first row of the hierarchy 0 mapping table 320 shown in FIG. 4, the hierarchy address is “00000000” and the valid flag is “0”. This indicates that the 00000000 address of layer 0 is a free area.

  The tier 1 mapping table 321 is an address table that is referred to when accessing the tier 1 tier device 60 from the LD mapping table 300. FIG. 5 is a diagram showing an example of the contents of the hierarchy 1 mapping table 321. In the layer 1 mapping table 321 shown in FIG. 5, the layer 1 layer address and the valid flag are recorded in association with each other. For example, in the first row of the hierarchy 1 mapping table 321 shown in FIG. 5, the hierarchy address is “00000000” and the valid flag is “1”. This indicates that the 00000000 address of layer 1 is in use and points to the address 00000000 of the device of layer number 1.

  The tier 2 mapping table 322 is an address table that is referred to when accessing the tier 2 tier device 60 from the LD mapping table 300. FIG. 6 is a diagram showing an example of the contents of the tier 2 mapping table 322. In the hierarchy 2 mapping table 322 shown in FIG. 6, the hierarchy address of the hierarchy 2 and the valid flag are recorded in association with each other. For example, in the first row of the hierarchy 2 mapping table 322 shown in FIG. 6, the hierarchy address is “80000000” and the valid flag is “1”. This indicates that the 80000000 address in layer 2 is in use and points to the address 80000000 in the object management table.

  The object management table 330 is an address table that is referred to when accessing an object address. FIG. 7 is a diagram showing an example of the contents of the object management table 330. In the object management table 330 shown in FIG. 7, an address (object address), a pointer to the object aggregation table 340 (object aggregation address), and an aggregation number are recorded in association with each other. For example, in the first row of the object management table 330 shown in FIG. 7, the object address is “80000000”, the object aggregation address is “90000000”, and the aggregation number is “1024”. This represents that 1024 addresses (blocks) are aggregated in the object aggregation address 90000000.

  The object aggregation table 340 is a table that stores addresses (blocks) aggregated as objects and object IDs (Identification) that are referred to when accessing the object storage. FIG. 8 is a diagram showing an example of the contents of the object aggregation table 340. In the object aggregation table 340 illustrated in FIG. 8, an object ID, a logical address, and a valid flag are recorded in association with each other. For example, in the first line of the object aggregation table 340 shown in FIG. 8, the object ID is “aad00exfdsdddb”, the logical address list is “00001000” to “000001ff0”, and the valid flag is “1”. This indicates that actual data of logical addresses 00001000 to 00001ff0 is stored in the object whose object ID is aad00exfdsdddb.

  The policy management table 350 is a table that stores the aggregation number for each hierarchical logical disk 50. FIG. 9 is a diagram showing an example of the contents of the policy management table 350. In the policy management table 350 shown in FIG. 9, the hierarchical logical disk number, the number of aggregations, the access frequency update interval, the objectization interval, the hierarchy replacement interval, and the object deletion interval are recorded in association with each other. For example, in the first line of the policy management table 350 shown in FIG. 9, the hierarchical logical disk number is “0”, the aggregation number is “1024”, the access frequency update interval is “1800 seconds”, and the objectization interval is “7200 seconds”. The layer replacement interval is “3600 seconds” and the object deletion interval is “14400 seconds”. This is because the logical disk with hierarchical logical disk number 0 aggregates 1024 addresses (blocks) into an object, updates the access frequency every 30 minutes, makes the data object every 2 hours, and changes the data hierarchy to 1 This indicates that the deletion of the object is performed every 4 hours every time.

  Next, the operation of the storage system SS according to the first embodiment of the present invention will be described in detail.

  FIG. 10 is a flowchart showing the operation of the access frequency update unit 200. Hereinafter, the operation when the access frequency update unit 200 updates the access frequency of the hierarchical logical disk 50 will be described with reference to FIG.

  The following processing may be periodically executed based on the access frequency update interval of the policy management table 350. The access frequency update unit 200 decrements (subtracts) the access frequency of the LD mapping table 300 of the target hierarchical logical disk 50 at once (step S10). Then, the process ends.

  Next, the data rearrangement operation for data with low access frequency using the object storage 2 by the objectification aggregation unit 210 will be described. The data rearrangement operation is divided into a first data rearrangement operation, a second data rearrangement operation, and a third data rearrangement operation.

  The first data relocation operation is periodically executed and includes the following processing. The objectification aggregation unit 210 refers to the policy management table 350, the LD mapping table 300, and the object aggregation table 340. Then, the objectification aggregation unit 210 transmits data to be converted into an object storage management unit 240. Then, the objectification aggregation unit 210 adds the object ID returned from the object storage management unit 240 to the object aggregation table 340. Then, the objectification aggregation unit 210 instructs the hierarchy exchanging unit 220 to delete actual data to be rearranged in the object storage.

  The second data rearrangement operation includes a process in which the objectification aggregating unit 210 returns actual data from the object storage 2 when a read occurs at an objectized logical address.

  The third data rearrangement operation includes a process in which the objectification aggregation unit 210 sets the data in the object storage 2 as invalid data when a write occurs to the objectized logical address.

  FIG. 11 is a flowchart illustrating an operation (first data relocation operation) in which the objectification aggregation unit 210 periodically relocate data with low access frequency to the object storage 2. Hereinafter, the first data rearrangement operation will be described with reference to FIG.

  The first data relocation operation may be executed by being periodically called from the access frequency update unit 200 based on the objectization interval of the policy management table 350. First, the objectification aggregation unit 210 acquires the aggregation number of the hierarchical logical disk 50 by referring to the policy management table 350 (step S20). Next, the objectification aggregation unit 210 refers to the LD mapping table 300 and acquires all entries having an access frequency of 0 and an OBJ flag of 0 (step S21). In addition to the above-described conditions, the objectification aggregation unit 210 may acquire only the entry having the lowest hierarchy number whose hierarchy device is other than OBJ by referring to the hierarchy management table 310.

  Next, the objectification aggregation unit 210 refers to the object aggregation table 340. Then, the objectification aggregation unit 210 determines whether the logical address of the entry of the LD mapping table 300 acquired in step S21 is included in the entry whose validity flag is 1 in the object aggregation table 340 (step S22). If it is included (“YES” in step S22), the objectification aggregation unit 210 updates the hierarchical address of the LD mapping table 300 and deletes it from the object of reobjectification (step S27). Then, the process proceeds to step S26. If not included ("NO" in step S22), the objectification aggregation unit 210 determines whether the number of entries exceeds the aggregation number (step S23). When it exceeds (“YES” in step S23), the objectification aggregation unit 210 transmits the aggregated actual data (block) to the object storage management unit 240 by REST (Representational State Transfer) (step S24). If the number of entries does not exceed the aggregation number (“NO” in step S23), the process ends.

  Next, the objectification aggregation unit 210 adds the object ID returned from the object storage management unit 240 and the aggregated logical address to the object aggregation table 340 (step S25). Next, the objectification aggregation unit 210 instructs the hierarchy switching unit 220 to delete the actual data (block) of the aggregation target logical address (step S26). Then, the process ends.

  FIG. 12 is a flowchart showing an operation (second data relocation operation) when reading is performed on an object address. Hereinafter, the second data rearrangement operation will be described with reference to FIG.

  When reading is performed on the objectized address (block), the objectification aggregation unit 210 retrieves the corresponding data by specifying the object ID to the object storage management unit 240 (step S30). The extracted data is a block represented by an address from which reading is performed. Next, the objectification aggregation unit 210 refers to the hierarchy management table 310 and acquires a hierarchy number whose hierarchy device is other than OBJ and which is the lowest (step S31). Next, the objectification aggregation unit 210 acquires one entry whose validity flag is 0 from the hierarchy mapping table corresponding to the hierarchy number acquired in step S31 (step S32). Next, the objectification aggregation unit 210 designates the actual data extracted in step S30 and the entry acquired in step S32, and instructs the hierarchy exchanging unit 220 to replace (step S33). Then, the process ends.

  FIG. 13 is a flowchart showing an operation (third data relocation operation) when writing is performed to an object address. Hereinafter, the third data rearrangement operation will be described with reference to FIG.

  When writing to the objectized address (block) is performed, the objectification aggregation unit 210 takes out the corresponding data by designating the object ID to the object storage management unit 240 (step S40). The data to be taken out is a block in which writing is performed and a block aggregated into an object together with this block. Next, the objectification aggregation unit 210 refers to the hierarchy management table 310 and acquires a hierarchy number whose hierarchy device is other than OBJ and which is the lowest (step S41). Next, the objectification aggregation unit 210 acquires the number of aggregations whose effective flag is 0 from the hierarchy mapping table corresponding to the hierarchy number acquired in step S41 (step S42). Next, the objectification aggregation unit 210 designates the actual data extracted in step S40 and the entry acquired in step S42, and instructs the hierarchy exchanging unit 220 to replace (step S43). Next, the objectification aggregation unit 210 updates the valid flag of the entry of the object in the object aggregation table 340 to 0 (step S44). As a result, the object becomes invalid data to be deleted by the object deleting unit 230. Then, the process ends.

  FIG. 14 is a flowchart showing the operation of the object deletion unit 230. Hereinafter, with reference to FIG. 14, an operation when the object deleting unit 230 deletes unnecessary data from the object storage 2 will be described.

  The following processing may be executed by periodically calling from the access frequency update unit 200 based on the object deletion interval of the policy management table 350. First, the object deletion unit 230 refers to the object management table 330 and the object aggregation table 340, and searches the object aggregation table 340 for an entry whose valid flag is 0 (step S50). If there is no target entry (“NO” in step S51), the process ends. If there is a target entry (“YES” in step S51), the object deletion unit 230 instructs the object storage management unit 240 to specify the object ID for deletion (step S52). Next, the object deletion unit 230 updates the object aggregation address of the object management table 330 to 0 (invalid) (step S53). Then, the process ends.

  Next, the operation of the object storage management unit 240 will be described. The operation of the object storage management unit 240 is divided into a first operation called from the objectification aggregation unit 210 and a second operation called from the object deletion unit 230.

  The first operation of the object storage management unit 240 includes the following processing. When called from the objectification aggregation unit 210, the object storage management unit 240 converts the actual data received from the objectification aggregation unit 210 into an object. Then, the object storage management unit 240 distributes the objectized data and stores it in the object storage destination device 70. Then, the object storage management unit 240 returns the result to the objectification aggregation unit 210.

The second operation of the object storage management unit 240 includes the following processing.
When called from the object deletion unit 230, the object storage management unit 240 deletes the target object.

  FIG. 15 is a flowchart showing an operation (first operation) when the object storage management unit 240 is called from the objectification aggregation unit 210 and stores an object. Hereinafter, the first operation of the object storage management unit 240 will be described with reference to FIG.

  When called from the objectification aggregation unit 210, first, the object storage management unit 240 converts the data received from the objectification aggregation unit 210 into an object. Then, the object storage management unit 240 distributes the objectized data and stores it in the object storage destination device 70 (step S60). Next, the object storage management unit 240 returns the objectized metadata to the objectification aggregation unit 210 (step S61). Then, the process ends.

  FIG. 16 is a flowchart showing an operation (second operation) when the object storage management unit 240 is called from the object deletion unit 230 and deletes the target object. Hereinafter, the second operation of the object storage management unit 240 will be described with reference to FIG.

When called from the object deletion unit 230, first, the object storage management unit 240 deletes the object indicated by the object ID received from the object deletion unit 230 from the object storage destination device 70 (step S70). Next, the object storage management unit 240 returns the deletion result to the object deletion unit 230 (step S71). And a process is complete | finished Next, operation | movement of the hierarchy switching part 220 is demonstrated. The operation of the hierarchy exchanging unit 220 is divided into a first operation when a deletion instruction is received from the objectification aggregation unit 210 and a second operation when a replacement instruction is received from the objectification aggregation unit 210.

  FIG. 17 is a flowchart showing an operation (first operation) when the hierarchy switching unit 220 receives a deletion instruction from the objectification aggregation unit 210. Hereinafter, the first operation of the hierarchy switching unit 220 will be described with reference to FIG.

  When called from the objectification aggregation unit 210, first, the hierarchy switching unit 220 sets the hierarchy address included in the entry of the LD mapping table 300 instructed by the objectification aggregation unit 210 to an address referring to the object management table 330. Update (step S80). Next, the hierarchy switching unit 220 updates the valid flag of the entry in the hierarchy mapping table referenced by the hierarchy address before update to 0 (invalid) (step S81). Then, the process ends.

  FIG. 18 is a flowchart showing an operation (second operation) when the hierarchy exchanging unit 220 receives an exchange instruction from the objectification aggregation unit 210. Hereinafter, with reference to FIG. 18, the second operation of the hierarchy exchanging unit 220 will be described.

  When called from the objectification aggregation unit 210, first, the hierarchy switching unit 220 writes actual data to the hierarchy device designated by the objectification aggregation unit 210. Then, the hierarchy switching unit 220 updates the entry of the corresponding hierarchy mapping table (step S90). Next, the hierarchy switching unit 220 updates the hierarchy address, the hierarchy number, and the OBJ flag included in the corresponding entry of the LD mapping table 300 with the information of the written actual data (step S91). Then, the process ends.

  As described above, the storage system SS according to the first embodiment of the present invention aggregates data whose access frequency is low in the block storage 1 and moves it to the object storage 2. As a result, the storage system SS according to the first embodiment of the present invention can store data having a capacity exceeding the specifications of the block storage 2. In addition, it is possible to minimize a decrease in performance and an increase in cost by moving data with low access frequency to an object storage with a low unit capacity. Furthermore, by consolidating data (blocks) and moving them to the object storage 2, the frequency of data objectification decreases, so that it is possible to prevent a decrease in processing performance related to objectization. Furthermore, by using an inexpensive object storage service provided by an external vendor or the like, the capacity of storable data can be flexibly increased.

[Second Embodiment]
Next, a second embodiment of the present invention will be described.

  FIG. 19 is a block diagram showing the configuration of the block storage according to the second embodiment of the present invention. The block storage 2 according to the second exemplary embodiment of the present invention includes an objectification aggregation unit and a hierarchy switching unit.

The objectification aggregation unit aggregates a predetermined number of blocks whose access frequency is lower than the threshold value, and transmits the aggregated blocks to the object storage. As described above, the tier replacement unit deletes the block transmitted to the object storage from the block storage 2. As described above, the storage system SS according to the second embodiment of the present invention uses the data whose access frequency is low in the block storage 1. Are moved to the object storage 2. As a result, the block storage 2 according to the second embodiment of the present invention has the same effects as the storage system SS according to the first embodiment.

  Although the object storage 2 in the first embodiment has been described as an object storage service provided by another vendor, other object storage services may be used. For example, the user of the storage system may install and operate the block storage 1 and the object storage 2 and connect them via a LAN (Local Area Network).

  FIG. 20 is a block diagram illustrating an example of elements constituting the computer. 20 includes a CPU (Central Processing Unit) 910, a RAM (Random Access Memory) 920, a ROM (Read Only Memory) 930, a hard disk drive 940, and a communication interface 950.

  The components of the block storage 1 and the object storage 2 described above may be realized by executing a program in the CPU 910 of the computer 900. Specifically, the access frequency update unit 200, the objectification aggregation unit 210, the hierarchy switching unit 220, the object deletion unit 230, and the object storage management unit 240, which are the components described in FIG. 1 and FIG. The CPU 910 may be realized by reading a program from the ROM 930 or the hard disk drive 940 and executing the read program by the CPU 910 as shown in the flowcharts of FIGS. In such a case, the present invention described by using the above-described embodiment as an example can be applied to a computer-readable storage medium (for example, hard disk drive 940 or the like) that stores a code representing the computer program or a code representing the computer program. It can be understood that it is constituted by a removable magnetic disk medium, optical disk medium, memory card, etc. (not shown).

  Alternatively, the access frequency update unit 200, the objectification aggregation unit 210, the hierarchy switching unit 220, the object deletion unit 230, and the object storage management unit 240 may be realized by dedicated hardware. Further, the block storage 1 and the object storage 2 may be dedicated hardware including these components.

  The present invention can be applied to the use of a disk array device.

SS Storage System 1 Block Storage 2 Object Storage 50 Hierarchical Logical Disk 60 Hierarchical Device 70 Object Storage Destination Device 100 Virtual Server 200 Access Frequency Update Unit 210 Objectization Aggregation Unit 220 Hierarchy Replacement Unit 230 Object Deletion Unit 240 Object Storage Management Unit 300 LD Mapping Table 310 Hierarchy management table 320 Hierarchy 0 mapping table 321 Hierarchy 1 mapping table 322 Hierarchy 2 mapping table 330 Object management table 340 Object aggregation table 350 Policy management table 900 Computer 910 CPU
920 RAM
930 ROM
940 Hard disk drive 950 Communication interface

Claims (10)

Aggregate a predetermined number of blocks whose access frequency is lower than the threshold, and send the aggregated block to the object storage that stores the object generated by converting the received data into an object that gives metadata and object ID Objectification aggregation means to
A block storage comprising: a hierarchy exchanging means for deleting the block transmitted to the object storage from the own block storage.   The block storage according to claim 1, wherein the objectification aggregation unit acquires the block from the object storage in response to reading of the block transmitted to the object storage.   The objectization aggregation means acquires the block and the block aggregated together with the block from the object storage in response to the writing to the block transmitted to the object storage. Block storage. Block storage according to any one of claims 1 to 3,
A storage system comprising: object storage that stores the data received from the block storage as the object. A storage system according to claim 4;
A computer system comprising a host computer accessible to the storage system. Aggregate a predetermined number of blocks in the block storage whose access frequency is lower than the threshold,
The aggregated block is transmitted to an object storage that stores an object generated by converting the received data into an object that gives metadata and an object ID .
A block storage control method for deleting a block transmitted to the object storage from the block storage.   The block storage control method according to claim 6, wherein the block is acquired from the object storage in response to reading of the block transmitted to the object storage.   The block storage control method according to claim 6 or 7, wherein the block and the block aggregated together with the block are acquired from the object storage in response to writing to the block transmitted to the object storage. The block storage performs the processing according to any one of claims 6 to 8,
Object storage, the storage system control method for storing data received from said block storage by said object of. A process of aggregating a predetermined number of blocks in the block storage whose access frequency is lower than the threshold;
A process of transmitting the aggregated object to an object storage for storing an object generated by converting the received data into an object that gives metadata and an object ID .
A program that causes a computer to execute a process of deleting a block transmitted to the object storage from the block storage.

Images