WO2016129053A1 - Ordinateur de gestion pour dispositif de stockage - Google Patents

Ordinateur de gestion pour dispositif de stockage Download PDF

Info

Publication number
WO2016129053A1
WO2016129053A1 PCT/JP2015/053636 JP2015053636W WO2016129053A1 WO 2016129053 A1 WO2016129053 A1 WO 2016129053A1 JP 2015053636 W JP2015053636 W JP 2015053636W WO 2016129053 A1 WO2016129053 A1 WO 2016129053A1
Authority
WO
WIPO (PCT)
Prior art keywords
pool
logical volume
zero
unallocated
information
Prior art date
Application number
PCT/JP2015/053636
Other languages
English (en)
Japanese (ja)
Inventor
佑介 小泉
金子　聡
Original Assignee
株式会社日立製作所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社日立製作所 filed Critical 株式会社日立製作所
Priority to PCT/JP2015/053636 priority Critical patent/WO2016129053A1/fr
Publication of WO2016129053A1 publication Critical patent/WO2016129053A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures

Definitions

  • the present invention relates to a storage computer management computer.
  • Large storage devices are also called storage subsystems, and not only enable high-speed and large-capacity data storage, but also have advanced data management functions.
  • the storage device is equipped with a plurality of physical storage devices such as hard disk drives.
  • a logical volume which is a logical storage area, is configured using the storage areas in these storage devices.
  • the storage apparatus provides a logical volume to the host computer. The host computer reads / writes data from / to the logical volume.
  • the thin provisioning technology provides a host computer with a virtual logical volume (hereinafter referred to as a virtual volume) that replaces a conventional logical volume.
  • a conventional logical volume requires a physical storage area (real storage area) of a size specified at the time of volume creation.
  • the thin provisioning technology takes out a real storage area for storing the data from the pool, which is a set of one or more logical volumes, and creates a virtual volume when the data is actually written. assign.
  • the thin provisioning technology it is not necessary to allocate a real storage area to the virtual volume until data is actually written, so that the real storage area can be saved.
  • Patent Document 1 discloses the following matters.
  • the management computer determines whether or not there is a predetermined pool that requires expansion of the pool size, based on the usage state of each pool.
  • the management computer includes a volume addition method for adding an unused real volume to the predetermined pool, and a data movement method for moving virtual logical volume data to a pool other than the predetermined pool. At least one of the methods is selected based on a predetermined selection criterion, and the pool size of the predetermined pool is expanded according to the selected method.
  • a page is a unit of a real storage area allocated to a virtual volume. Page zero writing is performed periodically by a scheduled job, and a page for which zero writing has not been completed is zero-written at the time of page allocation.
  • Patent Document 1 does not consider how many pages that have not been written zero in the logical volume, and may cause a decrease in I / O performance when a new page is allocated.
  • a typical example of the present invention is a management computer that manages a storage device that provides a host with a virtual volume to which a new zero page is allocated from a logical volume belonging to a pool in accordance with data writing, and includes a processor, a memory,
  • the memory includes information on a pool to which each logical volume belongs, information on the current unallocated zero page amount of each logical volume, information on zero write performance of each logical volume, and information on frequency of access to each logical volume
  • the processor refers to the first information
  • the processor refers to the first information, information on the current unallocated zero page amount, information on zero write performance, and access frequency of the logical volume belonging to the first pool.
  • the overall configuration of the computer system is shown.
  • An example of the logical volume table which a management computer has is shown.
  • An example of the shortage area table which a management computer has is shown.
  • An example of the statistical table which a physical computer has is shown.
  • summary of embodiment is shown.
  • An example of calculating the shortage amount of unassigned zero pages and the shortage occurrence time is shown.
  • the flowchart of a logical volume extraction program is shown.
  • the flowchart of a logical volume search process is shown.
  • aaa table various types of information may be described using the expression “aaa table”, but the various types of information may be expressed using a data structure other than the table.
  • the “aaa table” can be called “aaa information”.
  • an information element composed of values in each column in the table is called an entry, and an entry of “aaa table” is called “aaa table entry” for the sake of explanation.
  • the program may be installed in each computer or storage device from the program source.
  • the program source may be, for example, a program distribution server or a storage medium.
  • FIG. 1 is a diagram showing an overall configuration of a computer system according to an embodiment.
  • the computer system includes a host computer 1, a management computer 8, and a storage device 22.
  • the host computer 1 and the storage device 22 are connected via a data network 33.
  • the management computer 8, the host computer 1, and the storage device 22 are connected via a management network 34.
  • the data network 33 is, for example, a SAN (Storage Area Network) or an IP (Internet Protocol) network, but may be a network other than these.
  • the management network 34 is, for example, a SAN (Storage Area Network) or an IP (Internet Protocol) network, but may be a network other than these.
  • the data network 33 and the management network 34 may be the same network.
  • the host computer 1 and the management computer 8 may be the same computer.
  • the host computer 1 includes a CPU (Central Processing Unit) 2, a memory 3, an input device 4, a display device 5, a host interface (I / F) 6, and a management I / F 7.
  • the CPU 2 is a control device, and the memory 3 is a storage device.
  • the host computer 1 may not include the input device 4 and the display device 5.
  • the CPU 2 executes a program stored in the memory 3 and operates according to the program.
  • a CPU is a processor that executes a program stored in a memory connected thereto.
  • the host I / F 6 is an interface with the data network 33 and transmits / receives data and control commands to / from the storage apparatus 22.
  • the management I / F 7 is an interface with the management network 34 and transmits / receives data and control commands to / from the management computer 8 and the storage device 22.
  • FC Fibre Channel
  • the management computer 8 is a management system for the storage apparatus 22, and includes a CPU 9, a memory 12, an input device 10, a display device 11, and a management I / F 21.
  • the management computer 8 may not include the input device 10 and the display device 11.
  • the memory 12 stores a logical volume retrieval program 13, a logical volume return program 14, a logical volume migration correction program 15, a configuration information acquisition program 16, a logical volume table 17, a shortage area table 18, and a statistics table 19.
  • the logical volume retrieval program 13 retrieves a logical volume that satisfies a predetermined condition, and adds it to a pool that lacks unallocated zero pages. Details of the processing of the logical volume extraction program 13 will be described later with reference to the flowchart of FIG.
  • a page is a unit of a real storage area allocated to a virtual volume.
  • the zero page is a page that is initialized by writing predetermined initialization data.
  • the initialization data is also called zero data.
  • the logical volume return program 14 returns a logical volume that has been extracted from another pool by the logical volume extraction program 13 and added to the pool that lacks unallocated zero pages to the migration source pool.
  • the shortage of unassigned zero pages is temporary, and the performance of the migration source pool can be returned to the state before the migration by returning the logical volume to the migration source pool.
  • the logical volume movement correction program 15 further adds another logical volume to the pool when it is presumed that the allocation of the zero-written incomplete page occurs in the pool to which the logical volume has been added by the logical volume extraction program 13. Try.
  • the logical volume migration correction program 15 determines that when a logical volume extraction source by the logical volume extraction program 13 is a pool, it is estimated that zero write incomplete pages are allocated in the extraction source pool. Attempt to add a logical volume to the pool.
  • the configuration information acquisition program 16 acquires the configuration information of the storage device 22.
  • the logical volume table 17 manages information related to the logical volume.
  • the shortage area table 18 manages a shortage of zero page capacity and a shortage occurrence time.
  • the statistical table 19 manages statistical information of the storage device 22.
  • the management I / F 20 is an interface with the management network 34, and transmits and receives data and control commands between the host computer 1 and the storage device 22.
  • the storage device 22 has a disk controller 23 and a disk device 31.
  • the disk controller 23 includes a CPU 24, a memory 25, a host I / F 28, a management I / F 29, and a disk I / F 30.
  • the memory 25 stores a configuration information management program 26 and a logical volume migration program 27.
  • the configuration information management program 26 manages the configuration information of the storage device 22.
  • the logical volume migration program 27 changes the allocation of the logical volume to the pool.
  • the host I / F 28 is an interface with the data network 33 and transmits / receives data and control commands to / from the host computer 1.
  • the management I / F 29 is an interface with the management network 34 and transmits and receives data and control commands between the host computer 1 and the management computer 8.
  • the disk I / F 30 transmits and receives data and control commands to and from the disk device 31.
  • the disk device 31 includes a plurality of types of physical storage media 32 such as an HDD (Hard Disk Drive) and an SSD (Solid State Drive).
  • the storage device 22 manages a set of real storage areas in a plurality of physical storage media 32 in the disk device 31 as a logical volume, and manages a set of logical volumes as a pool.
  • the storage device 22 allocates a real storage area (page) included in the logical volume in the pool to the virtual volume when writing new data.
  • the virtual volume is a Thin Provisioning volume whose capacity can be dynamically expanded.
  • FIG. 2 shows an example of the logical volume table 17 held by the management computer 8.
  • the logical volume table 17 is a table for managing logical volume information.
  • the logical volume ID column 41 indicates a logical volume ID for identifying a logical volume.
  • the pool ID column 42 indicates a pool ID for identifying the pool to which the logical volume belongs.
  • the capacity column 43 indicates the capacity value of the logical volume.
  • the used amount 44 column indicates the amount of storage area used in the logical volume.
  • the unallocated zero page amount column 45 indicates the total capacity of unallocated zero pages in the logical volume. Unassigned zero pages are pages that are initialized with predetermined data and are not assigned to any virtual volume.
  • the disk type column 46 indicates the type of physical storage medium that provides the logical volume.
  • the pool ID for the logical volume is indicated by “Unallocated”.
  • Information managed by the logical volume table 17 is acquired from the configuration information management program 26 of the storage apparatus 22 by the configuration information acquisition program 16 of the management computer 8.
  • FIG. 3 shows an example of the shortage area table 18 held by the management computer 8.
  • the shortage area table 18 manages information on how many zero pages need to be added to the pool by when the unallocated pages that have not yet been zero-written cannot be allocated from the pool to the virtual volume. It is a table to do.
  • the shortage area table 18 has a plurality of columns.
  • the deletion process ID column 51 indicates a deletion process ID for identifying the virtual volume deletion process.
  • the pool ID column 52 indicates a pool ID for identifying the pool to which the virtual volume to be deleted belongs.
  • the unallocated zero page shortage column 53 indicates the amount of unallocated zero pages estimated to be insufficient in the pool in the future.
  • An unallocated zero page shortage occurrence time column 54 indicates a time when it is estimated that an unallocated zero page shortage will occur in the pool in the future.
  • the disk type column 55 indicates a disk type of an unallocated zero page that is insufficient.
  • FIG. 4 shows an example of the statistical table 19 held by the management computer 8.
  • the statistical table 19 is a table for managing statistical information of various I / Os in each logical volume.
  • the statistical table 19 has a plurality of columns.
  • the logical volume ID column 63 indicates a logical volume ID for identifying a logical volume.
  • the pool ID column 62 indicates a pool ID for identifying the pool to which the logical volume belongs.
  • the storage ID column 61 indicates a storage ID for identifying the storage apparatus to which the logical volume belongs.
  • the new page allocation frequency column 64 indicates the value of the new page allocation frequency for the logical volume calculated based on the statistics.
  • the logical volume extraction performance column 65 indicates the logical volume extraction performance value calculated based on the statistics.
  • the zero write performance column 66 indicates the value of zero write performance for the logical volume.
  • the information managed by the statistics table 19 is acquired from the configuration information management program 26 of the storage device 22 by the configuration information acquisition program 16 of the management computer 8.
  • the statistical information may be an average value, a maximum value, a minimum value, etc. for a predetermined period, or may be an average value, a maximum value, a minimum value, etc. for each date, day of the week, and time, and other values. It may be. Instead of the statistical information, the value of the latest specific time may be used.
  • FIG. 5 shows an outline of the present embodiment.
  • the VVOL 70 is a virtual volume.
  • POOL 71 is a pool.
  • the VOL 72 is a logical volume.
  • the VOL 72 is included in the POOL 71, and the VVOL 70 is assigned a real storage area from the POOL 71.
  • the area in the VOL 72 allocated to the VVOL 70 becomes an unallocated area.
  • the area in the VOL 72 that has been allocated to the VVOL 70 is an unallocated area that has not yet been zero-written.
  • the storage apparatus 22 When a page included in an unallocated area where zero writing has not been completed is newly allocated to a virtual volume, the storage apparatus 22 performs zero writing on the page before allocation. The storage device 22 executes zero writing to a page included in an unallocated area where zero writing has not been completed asynchronously or by other methods. A page for which zero writing has been completed becomes an unallocated zero page.
  • the storage apparatus 22 allocates after performing zero-write on the page at that time.
  • the allocation of the unallocated page that has not been zero-written places a higher load on the I / O processing of the CPU 24 and the disk device 31 of the storage device 22. Therefore, the I / O performance of the storage device 22 is degraded.
  • the following processing is performed in order to suppress the I / O performance degradation of the storage device 22 as described above.
  • it is estimated whether or not a shortage of unallocated zero pages occurs in the POOL 71, that is, whether or not an unallocated page (unallocated non-zero page) in which zero writing has not been completed occurs in the POOL 71.
  • the present embodiment adds a VOL 73, which is a logical volume including an unallocated zero page, to the POOL 71 from outside the POOL 71. Furthermore, this embodiment allocates a new page from the added unallocated zero page. This avoids allocation of unallocated pages that have not yet been zero-written.
  • This embodiment searches for a logical volume candidate to be migrated to the POOL 71 in a logical volume that satisfies a predetermined condition. Specifically, the present embodiment estimates the shortage amount and shortage occurrence time of unassigned zero pages, includes logical volumes that include unassigned zero pages that are larger than the shortage amount and can be moved to the POOL 71 by the shortage occurrence time. Search and move the corresponding logical volume to the POOL 71.
  • FIG. 6 shows a graph 80 of a calculation example of the shortage amount (column 53) and shortage occurrence time (column 54) of the unallocated zero page in the shortage area table 18.
  • the horizontal axis 81 represents time.
  • the vertical axis 82 represents the amount of storage area.
  • a solid line 88 represents a change over time in the total amount of unallocated zero pages in the pool in a situation where no new page allocation occurs.
  • the intercept v0 is the total amount of current unallocated zero pages, for example, the total amount of unallocated zero pages at the time when the virtual volume is deleted.
  • the total amount v0 of the current unallocated zero pages is the sum of the unallocated zero page amounts (column 45) of all the logical volumes belonging to the pool in the logical volume table 17.
  • the logical volume table 17 is updated as needed, and stores the latest data when the virtual volume is deleted.
  • the amount v1 is the total amount of all unallocated pages in the pool at the current time.
  • the amount v1 matches the total unused amount of all logical volumes in the pool.
  • the unused amount of the logical volume is a value obtained by subtracting the used amount (column 44) from the capacity (column 43) in the logical volume table 17.
  • Unallocated non-zero pages are sequentially zeroed, and the total amount of unallocated zero pages increases with time. At time t1, zero writing of all unallocated pages in the pool is completed.
  • the slope of the total amount of unallocated zero pages can be calculated based on the zero write performance (column 66) recorded in the statistical table 19. For example, the slope is a weighted average of zero write performance with the unallocated non-zero page amount at the current time of the logical volume in the pool as a weight.
  • the unallocated non-zero page amount of each logical volume at the current time is a value obtained by subtracting the used amount (column 44) and the unallocated zero page amount (column 45) from the capacity (column 43) in the logical volume table 17. .
  • the alternate long and short dash line 89 and the alternate long and two short dashes line 90 represent examples of changes in the page amount that may be newly allocated from the current time.
  • the intercept is zero.
  • the inclination is calculated from the value of the new page allocation frequency (column 64) recorded in the statistical table 19. Specifically, it is the sum of products of new page allocation frequency and page size for all volumes in the pool.
  • the alternate long and short dash line 89 indicates an example in which new page allocation exceeding the unallocated zero page amount of the pool occurs.
  • An alternate long and two short dashes line 90 shows an example in which only new page allocation equal to or less than the unallocated zero page amount occurs.
  • Time t0 is a shortage occurrence time (column 54) in the shortage area table 18. Furthermore, it can be seen that there may be a shortage of unassigned zero pages of amount v2.
  • the amount v2 is an insufficient amount of unassigned zero pages (column 53) in the insufficient region table 18. In this case, it is preferable to search and move a logical volume that includes unallocated zero pages of the amount v2 or more and can be moved to the pool by time t0.
  • the management computer 8 compares the total amount graph 88 of unassigned zero pages in the pool with the graph 89 or 90 of the page amount that may be newly allocated from the current time. It can be estimated whether allocation of non-zero pages will occur. When there is a possibility that allocation of pages that have not yet been zero-written occurs, the management computer 8 calculates the shortage occurrence time and shortage amount of unassigned zero pages and stores them in the shortage area table 18.
  • the management computer 8 replaces the zero write performance indicated by the statistical table 19 with the minimum zero write performance. It may be used to calculate a shortage amount of zero pages and a shortage occurrence time.
  • a plurality of types of logical volumes having different disk types may be included in one pool, and a plurality of types of real storage areas having different characteristics may be allocated to one virtual volume created from the pool.
  • the shortage area table 18 is created for each disk type (tier), and logical volume migration is executed for each disk type (tier). Is done.
  • FIG. 7 is a flowchart showing processing of the logical volume extraction program 13 according to the embodiment.
  • the logical volume retrieval program 13 detects an increase in non-allocated non-zero pages accompanying the logical volume migration in the pool of the storage device 22 managed by the management computer 8 (S1).
  • the logical volume extraction program 13 detects an increase in unallocated non-zero pages in the logical volume table 17 updated by the configuration information acquisition program 16.
  • the amount of non-allocated non-zero pages is a value obtained by subtracting the usage amount and the amount of unallocated zero pages from the capacity. For example, when a virtual volume is deleted, the usage amount decreases and the amount of unallocated non-zero pages increases.
  • the logical volume extraction program 13 will allocate an unallocated page (unallocated non-zero page) in which zero writing has not been completed due to a shortage of unallocated zero pages in the future. It is estimated whether or not (S2).
  • the logical volume extraction program 13 refers to the logical volume table 17 and the statistics table 19 and calculates the time change 88 of the unallocated zero page in the pool shown in FIG.
  • unallocated zero page time variation 88 indicates the time variation of the total amount of unallocated zero pages in the pool in a situation where no new page allocation occurs. The calculation method is as described with reference to FIG.
  • the logical volume extraction program 13 refers to the logical volume table 17 and the statistical table 19 and calculates the change 89 or 90 of the page amount that may be newly allocated as shown in FIG.
  • the logical volume extraction program 13 compares the time change 88 of the unallocated zero page amount with the change 89 or 90 of the page amount that may be newly allocated, and determines whether there will be a shortage of unassigned zero pages in the future. Estimate whether or not.
  • the management computer 8 may accept a user setting of an allowable capacity value for allocation of pages that have not yet been written to zero. Thereby, a logical volume search according to a user request is realized.
  • the capacity allowable value is determined in advance, when the shortage of unallocated zero pages is equal to or smaller than the capacity allowable value, the estimation result of step S2 is NO.
  • the shortage amount of the unallocated zero page is the amount v2 described with reference to FIG. 6 and is a value stored in the column 53 of the shortage region table 18.
  • step S2 if the estimation result is NO, that is, if there is no possibility of allocation of unallocated pages that have not yet been zero-written, the logical volume extraction program 13 ends this processing.
  • step S2 if the estimation result is YES, that is, if there is a possibility that unallocated non-zero pages will be allocated, the logical volume extraction program 13 shows the unallocated zero page shortage v2 shown in FIG. And the shortage occurrence time t0 is calculated. The logical volume extraction program 13 stores the calculated shortage amount v2 and the shortage occurrence time t0 in the columns 53 and 54 of the shortage region table 18 (S3).
  • the shortage occurrence time t0 includes a time change 88 of an unassigned zero page amount in a situation where a new page assignment does not occur, a change 89 of a page amount that may be newly assigned, The time at the intersection of
  • the insufficient amount v2 is the maximum amount of non-allocated non-zero pages that may cause allocation.
  • the shortage amount v2 may be newly assigned to the unassigned zero page amount 88 in a situation where no new page assignment occurs at the time when the maximum value v1 of the unassigned zero page amount in a situation where no new page assignment occurs. This is a difference from the page amount 89.
  • the logical volume retrieval program 13 also stores necessary information in other columns of the shortage area table 18. Specifically, the logical volume extraction program 13 stores the deletion processing ID of the virtual volume deletion processing that caused the shortage of unallocated zero pages in the column 51. The logical volume extraction program 13 stores the pool ID of the pool in which the shortage of unallocated zero pages occurs in the column 52.
  • the logical volume extraction program 13 stores the disk type in which the shortage of unallocated zero pages occurs in the column 55.
  • the disk type of the tier where shortage of unallocated zero pages occurs is stored.
  • the disk type of the pool is stored.
  • the logical volume retrieval program 13 executes a logical volume search process (S4).
  • the logical volume retrieval program 13 searches for a logical volume that can be moved to make up the shortage amount v2 of the unallocated zero page by the shortage occurrence time t0 as a candidate for the logical volume to be moved.
  • the logical volume extraction program 13 refers to the search result and determines whether or not the corresponding logical volume exists (S5).
  • step S5 determines whether the corresponding logical volume does not exist. If the determination result in step S5 is NO, that is, if the corresponding logical volume does not exist, the logical volume extraction program 13 ends the process. If the determination result in step S5 is YES, that is, if the corresponding logical volume exists, the logical volume retrieval program 13 instructs the volume migration program 27 of the storage device 22 to migrate the logical volume.
  • FIG. 8 shows a flowchart of the logical volume search process S4 according to this embodiment.
  • the logical volume retrieval program 13 refers to the shortage area table 18 to obtain from the column 53 the shortage of unallocated zero pages that occur in the pool, and further obtains the disk type from the column 55.
  • the logical volume extraction program 13 refers to the logical volume table 17 and searches for one logical volume that satisfies the following three conditions as a logical volume candidate to be moved (S7).
  • the unallocated zero page amount indicated by the column 45 is equal to or greater than the insufficient amount of unallocated zero pages.
  • the disk type indicated by the column 46 is the same as the disk type of the insufficient unallocated zero page.
  • the pool ID indicated by the column 42 is “Unallocated”.
  • condition (1) by searching for a single logical volume that satisfies the shortage, it is possible to efficiently perform logical volume migration processing to compensate for the shortage.
  • searching for a logical volume of the same disk type as the insufficient disk type as indicated by the condition (2) it is possible to compensate for the unallocated zero page having the insufficient characteristics.
  • the above condition (3) indicates a logical volume that does not belong to the pool.
  • the logical volume extraction program 13 may include, in the search target in step S7, a logical volume that includes only unallocated zero pages that are smaller than the shortage of unallocated zero pages in the pool, according to user settings. .
  • the management computer 8 receives an input of an allowable difference amount (capacity allowance value) between the unallocated zero page amount of the logical volume to be included in the search target and the shortage amount of the unallocated zero page from the user.
  • the logical volume extraction program 13 searches for a logical volume having a total amount of unallocated zero pages equal to or larger than the amount obtained by subtracting the capacity allowable value from the shortage of unallocated zero pages in the pool. .
  • a pool with insufficient unallocated zero pages is configured with a plurality of storage area hierarchies with different characteristics, that is, a plurality of types of logical volumes with different disk types are included in the pool.
  • a plurality of types of real storage areas with different characteristics may be allocated to a virtual volume created from a pool.
  • the disk type of the logical volume to be searched (column 46 of the logical volume table 17) may be different from the disk type of unassigned zero page that is insufficient (column 55 of the insufficient area table 18).
  • the performance of the disk type to be searched is higher than the performance of the insufficient disk type. As a result, the shortage of the storage area can be compensated and a decrease in the performance of the pool can be avoided.
  • the logical volume retrieval program 13 may select logical volumes of the same disk type with priority over logical volumes of different disk types. That is, when there is no logical volume of the same disk type that satisfies the condition, the logical volume extraction program 13 searches for a disk type logical volume with higher performance.
  • the logical volume retrieval program 13 searches for a plurality of logical volumes that can compensate for the shortage. As a result, the shortage of unallocated zero pages in the pool can be compensated by the total amount of unallocated zero pages included in the plurality of logical volumes.
  • the logical volume extraction program 13 determines whether there is a logical volume that satisfies the above three conditions (S8). If the decision result in the step S8 is YES, that is, if the corresponding logical volume exists, the logical volume retrieval program 13 stores the logical volume search result in the memory 12, and ends the logical volume search process S4. .
  • step S8 determines whether there is no logical volume that satisfies the above three conditions. If the determination result of step S8 is NO, that is, if there is no logical volume that satisfies the above three conditions, the logical volume retrieval program 13 refers to the logical volume table 17 and the unallocated zero page is different from the insufficient pool. In a logical volume belonging to another pool, a logical volume satisfying the conditions (1) and (2) is searched (S9).
  • the logical volume retrieval program 13 has an unallocated zero page amount equal to or larger than the insufficient amount of unallocated zero pages in the pool, and is the same as the disk type of unallocated zero pages whose disk type is insufficient, Search for logical volumes belonging to another pool.
  • step S7 The change of conditions (1) and (2) is the same as step S7.
  • some logical volumes may not belong to any pool, and other logical volumes may belong to other pools.
  • the logical volume retrieval program 13 determines whether or not there are logical volumes that satisfy the conditions (1) and (2) and belong to another pool (S10). If the determination result in step S10 is NO, that is, if there is no logical volume that satisfies the condition, the logical volume retrieval program 13 stores the logical volume search result in the memory 12, and ends the logical volume search process S4. .
  • step S10 If the determination result in step S10 is YES, that is, if there is a logical volume that satisfies the condition, the logical volume extraction program 13 among the logical volumes that satisfy the condition, the unallocated zero page indicated by the column 54 of the insufficient area table 18 A search is made for a logical volume that can be moved by the shortage occurrence time (S11). This can reduce the possibility of unallocated non-zero page allocation in the pool.
  • the logical volume retrieval program 13 searches for a logical volume in which the data movement ends before the shortage occurrence time of the unallocated zero page.
  • the time required for the data movement corresponds to the time required to take out the logical volume from the pool.
  • the logical volume extraction program 13 refers to the statistics table 19 and acquires the value of the logical volume extraction performance of the corresponding logical volume from the column 65. Further, the logical volume retrieval program 13 refers to the logical volume table 17 and acquires the value of the usage amount of the corresponding logical volume from the column 44.
  • the logical volume extraction program 13 calculates the time taken to extract the logical volume from the pool by dividing the capacity value by the acquired logical volume extraction performance value. This time corresponds to the time required to move the logical volume from the migration source pool to the migration destination pool in this embodiment.
  • the logical volume extraction program 13 compares the calculated extraction time of each logical volume with the shortage occurrence time of the unallocated zero page indicated by the column 54 of the shortage area table 18 to determine whether there is a movable logical volume ( S12).
  • the user can also include an unmovable (retrievable) logical volume in the search condition in step S11 by the shortage occurrence time of the unallocated zero page in the shortage area table 18.
  • the management computer 8 accepts the setting by the user of the allowable delay time that allows the delay from the shortage occurrence time of the unallocated zero page.
  • the search for the logical volume according to the user request is realized.
  • the allowable delay time is zero or a value greater than zero.
  • the logical volume retrieval program 13 searches for a logical volume to which data movement is possible by a time obtained by adding a delay allowable time to the shortage occurrence time. When the delay allowable time is not set by the user, the predetermined delay time is zero.
  • step S12 determines whether there is no logical volume that satisfies the condition. If the determination result in step S12 is NO, that is, if there is no logical volume that satisfies the condition, the logical volume retrieval program 13 stores the logical volume search result in the memory 12, and ends the logical volume search processing 100.
  • step S12 determines whether there is a logical volume that satisfies the conditions. If the determination result in step S12 is YES, that is, if there is a logical volume that satisfies the conditions, the logical volume extraction program 13 can return a logical volume that can be returned before the occurrence of a shortage of zero pages in the migration source pool. Is searched (S13). That is, logical volumes that cannot be returned before the occurrence of a shortage of zero pages in the migration source pool are excluded from candidates for logical volumes to be migrated. Next, the logical volume retrieval program 13 stores the logical volume search result in the memory 12 and ends the logical volume search process S4.
  • step S13 specifically, the logical volume extraction program 13 estimates the time when a shortage of unallocated zero pages occurs in the migration source pool after the logical volume is migrated from the migration source pool. Further, the logical volume retrieval program 13 estimates the time for returning the logical volume to the migration source pool. The logical volume retrieval program 13 compares the two estimated times and determines whether or not the logical volume can be returned until a shortage of zero pages occurs in the migration source pool.
  • the time at which a zero page shortage occurs in the migration source pool can be calculated from the new page allocation frequency in the migration source pool before volume migration and the unallocated amount (unused amount) in the migration source pool after volume migration.
  • the logical volume extraction program 13 acquires the value of the new page allocation frequency for all the logical volumes in the pool from the statistics table 19 and calculates the sum of them. All logical volumes include logical volumes to be moved.
  • the logical volume retrieval program 13 refers to the logical volume table 17 and calculates a value obtained by subtracting the capacity of the logical volume to be moved from the total unallocated amount of all logical volumes in the pool.
  • the calculated value is the unallocated amount in the migration source pool after the volume migration.
  • the unallocated amount of one logical volume is a value obtained by subtracting the usage amount from the capacity.
  • the logical volume retrieval program 13 calculates a value obtained by dividing the total unallocated amount by the product of the total new page allocation frequency and the page size, and adds the value to the current time.
  • the calculated value is an estimated value of the zero page shortage occurrence time in the migration source pool.
  • the logical volume extraction program 13 may estimate the zero page shortage occurrence time of the migration source pool by the same method as the zero page shortage occurrence time of the migration destination pool.
  • a time change 88 of the total amount of unallocated zero pages in the pool is calculated for the remaining logical volumes after moving the logical volume.
  • the change 88 or 89 of the page amount that may be newly allocated from the current time is calculated for all the volumes in the pool including the logical volume to be moved.
  • the return time of the logical volume can be estimated from the time required for logical volume data movement in the migration source pool, the time required for zero writing in the migration destination pool, and the time required for data migration of the logical volume in the migration destination pool. A value obtained by adding these total times to the current time is the estimated return time.
  • the time required for data migration in the migration source pool can be calculated from the extraction performance value of the logical volume and the usage amount of the logical volume.
  • the extraction performance value is acquired from the column 65 of the statistical table 19, and the usage amount is acquired from the column 44 of the logical volume table 17.
  • the time required for zero writing in the destination pool corresponds to t1 described with reference to FIG.
  • the time required for data movement of the logical volume in the migration destination pool can be calculated from the usage amount of the logical volume at time t1 when zero writing is completed in the migration destination and the logical performance extraction value of the logical volume.
  • the usage amount for example, the value of the shortage amount column 53 of the unallocated zero page in the shortage region table 18 is used. This assumes that the shortage of unallocated zero pages calculated before logical volume migration is actually allocated from the migration target logical volume.
  • the logical volume migration program 27 of the storage device 22 returns the migrated logical volume to the migration source pool according to the instruction from the logical volume return program 14.
  • the logical volume return program 14 detects the completion of zero writing to an unallocated area where zero writing has not been completed, which is caused by the virtual volume deletion processing, and moves the logical volume so that the moved logical volume is returned to the migration source pool. An instruction is issued to the program 27.
  • the logical volume return program 14 periodically determines whether or not an unallocated page that has not been zero-written occurs in the migration destination pool until the zero writing is completed. When it is determined that the allocation of an unallocated page that has not been completed with zero writing does not occur, the logical volume return program 14 returns the logical volume that has been moved to the migration source pool at the time before the completion of zero writing. The logical volume migration program 27 is instructed. As a result, an earlier timing logical volume can be returned to the migration source pool.
  • Whether or not allocation of an unallocated page that has not been completed with zero writing occurs is determined by the method described with reference to FIG. 6, and the latest value in the logical volume table 17 is referred to.
  • the intercept v0 and the slope of the solid line 88 are calculated from the values of the logical volume excluding the logical volume that has moved.
  • the logical volume extraction program 13 controls the storage device 22 to perform a high-performance disk type storage area (of the storage areas that have become unallocated pages that have not been zero-written due to the deletion of the virtual volume ( Zero writing may be performed with priority from the page.
  • the logical volume return program 14 may control the storage device 22 and return it to the migration source pool with priority from a high-performance disk type logical volume. As a result, the pool performance can be improved more quickly.
  • the logical volume migration correction program 15 monitors the amount of unallocated zero pages in the logical volume migration source or migration destination pool, and whether or not a future shortage of unallocated zero pages will occur based on the monitoring result. presume. When a future shortage of unallocated zero pages is estimated, it is determined to add a logical volume including unallocated zero pages to the pool.
  • the logical volume movement correction program 15 detects that the ratio of zero pages in the unallocated area has fallen below a threshold value. When the ratio falls below the threshold, the logical volume migration correction program 15 obtains the current value of zero write performance for the area and the current value of new page allocation frequency for the pool from the storage device 22 and needs to add a logical volume. It is determined whether or not there is. If it is necessary to add a logical volume, add a logical volume outside the pool.
  • the logical volume movement correction program 15 operates in the same manner as the logical volume retrieval program 13, but differs in the following points.
  • the logical volume migration correction program 15 does not detect the occurrence of deletion of the virtual volume as a trigger for starting the process, but the ratio of zero pages in the unallocated area in the migration source pool or migration destination pool of the logical volume falls below the threshold. Detect that.
  • the logical volume migration correction program 15 does not execute various calculation processes using the performance information recorded in the statistical table 19 but executes various calculation processes using the latest actual performance information of the storage device 22. .
  • the logical volume migration correction program 15 can add the necessary unallocated zero pages to the pool in response to a shortage of unallocated zero pages that occur during zero writing in the migration source pool or the migration destination pool.
  • the actual performance information instead of the statistical information, it is possible to determine the occurrence of a shortage of unallocated zero pages according to a short-term change in the situation. Note that the occurrence of a shortage of unallocated zero pages may be determined using statistical information.
  • this invention is not limited to the above-mentioned Example, Various modifications are included.
  • the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described.
  • a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment.
  • each of the above-described configurations, functions, processing units, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit.
  • Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor.
  • Information such as programs, tables, and files for realizing each function can be stored in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), or a recording medium such as an IC card or an SD card.
  • control lines and information lines indicate what is considered necessary for the explanation, and not all control lines and information lines on the product are necessarily shown. In practice, it may be considered that almost all the components are connected to each other.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

L'invention concerne un ordinateur de gestion qui gère un dispositif de stockage pour fournir un volume virtuel. L'ordinateur de gestion estime si un premier groupe sera à court de pages mises à zéro non attribuées dans le futur en se basant sur des informations concernant la quantité actuelle de pages mises à zéro non attribuées dans chaque volume logique appartenant au premier groupe, des informations concernant les performances d'écriture de zéro pour chaque volume logique, et des informations concernant la fréquence d'accès à chaque volume logique. S'il est estimé que le premier groupe sera à court de pages mises à zéro non attribuées, l'ordinateur de gestion décide d'ajouter, au premier groupe, un volume logique qui comprend des pages mises à zéro non attribuées et qui n'appartient pas au premier groupe.
PCT/JP2015/053636 2015-02-10 2015-02-10 Ordinateur de gestion pour dispositif de stockage WO2016129053A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2015/053636 WO2016129053A1 (fr) 2015-02-10 2015-02-10 Ordinateur de gestion pour dispositif de stockage

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2015/053636 WO2016129053A1 (fr) 2015-02-10 2015-02-10 Ordinateur de gestion pour dispositif de stockage

Publications (1)

Publication Number Publication Date
WO2016129053A1 true WO2016129053A1 (fr) 2016-08-18

Family

ID=56614382

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2015/053636 WO2016129053A1 (fr) 2015-02-10 2015-02-10 Ordinateur de gestion pour dispositif de stockage

Country Status (1)

Country Link
WO (1) WO2016129053A1 (fr)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010510597A (ja) * 2006-11-22 2010-04-02 インターナショナル・ビジネス・マシーンズ・コーポレーション データ・ストレージにおけるリソース再利用のための装置および方法
WO2011135617A1 (fr) * 2010-04-27 2011-11-03 株式会社日立製作所 Procédé et dispositif de mémoire permettant de limiter une capacité totale de volumes virtuels associés à une mémoire commune
JP2011242840A (ja) * 2010-05-14 2011-12-01 Hitachi Ltd ＴｈｉｎＰｒｏｖｉｓｉｏｎｉｎｇが適用されたストレージ装置
JP2012073825A (ja) * 2010-09-29 2012-04-12 Hitachi Ltd 計算機システム及び計算機システムの管理方法
JP2013161148A (ja) * 2012-02-02 2013-08-19 Fujitsu Ltd 仮想ストレージ装置，制御装置及び制御プログラム
US20130254500A1 (en) * 2012-03-21 2013-09-26 Hitachi, Ltd. Storage apparatus, computer system, and control method for storage apparatus

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010510597A (ja) * 2006-11-22 2010-04-02 インターナショナル・ビジネス・マシーンズ・コーポレーション データ・ストレージにおけるリソース再利用のための装置および方法
WO2011135617A1 (fr) * 2010-04-27 2011-11-03 株式会社日立製作所 Procédé et dispositif de mémoire permettant de limiter une capacité totale de volumes virtuels associés à une mémoire commune
JP2011242840A (ja) * 2010-05-14 2011-12-01 Hitachi Ltd ＴｈｉｎＰｒｏｖｉｓｉｏｎｉｎｇが適用されたストレージ装置
JP2012073825A (ja) * 2010-09-29 2012-04-12 Hitachi Ltd 計算機システム及び計算機システムの管理方法
JP2013161148A (ja) * 2012-02-02 2013-08-19 Fujitsu Ltd 仮想ストレージ装置，制御装置及び制御プログラム
US20130254500A1 (en) * 2012-03-21 2013-09-26 Hitachi, Ltd. Storage apparatus, computer system, and control method for storage apparatus

Similar Documents

Publication Publication Date Title
JP5685676B2 (ja) 計算機システム及びデータ管理方法
US8850152B2 (en) Method of data migration and information storage system
US8447946B2 (en) Storage apparatus and hierarchical data management method for storage apparatus
JP6051228B2 (ja) 計算機システム、ストレージ管理計算機及びストレージ管理方法
JP4684864B2 (ja) 記憶装置システム及び記憶制御方法
JP5309259B2 (ja) ストレージ装置及びその制御方法
US20130138908A1 (en) Storage system and pool capacity scale-down control method
CN106610903B (zh) 层级化存储器系统、存储器控制器和去重及存储器层级化方法
US8402214B2 (en) Dynamic page reallocation storage system management
JP5080611B2 (ja) ＴｈｉｎＰｒｏｖｉｓｉｏｎｉｎｇが適用されたストレージ装置
US9983806B2 (en) Storage controlling apparatus, information processing apparatus, and computer-readable recording medium having stored therein storage controlling program
CN103761053A (zh) 一种数据处理方法和装置
US20150331793A1 (en) Storage system
US20220066786A1 (en) Pre-scanned data for optimized boot
US20180267879A1 (en) Management computer, performance monitoring method, and computer system
US20170262220A1 (en) Storage control device, method of controlling data migration and non-transitory computer-readable storage medium
US10782922B2 (en) Storage device volume selection for improved space allocation
JP2009075814A (ja) ストレージサブシステム及び記憶制御方法
US10365846B2 (en) Storage controller, system and method using management information indicating data writing to logical blocks for deduplication and shortened logical volume deletion processing
US20160259725A1 (en) Storage apparatus, computer system, and method
US9552292B2 (en) Apparatus and method for allocating virtual memory addresses to continuous physical addresses
WO2016129053A1 (fr) Ordinateur de gestion pour dispositif de stockage
EP2642379A2 (fr) Appareil de traitement dýinformations, procédé et programme d'attribution de données
JP7050540B2 (ja) 論理記憶領域の配置制御装置、論理記憶領域の配置制御システム、論理記憶領域の配置制御方法、及び、論理記憶領域の配置制御プログラム
US9128819B2 (en) Storage system and management method therefor

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15881928

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15881928

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP