US20050044310A1 - Method of managing virtual volumes in a utility storage server system - Google Patents

Method of managing virtual volumes in a utility storage server system Download PDF

Info

Publication number
US20050044310A1
US20050044310A1 US10/959,569 US95956904A US2005044310A1 US 20050044310 A1 US20050044310 A1 US 20050044310A1 US 95956904 A US95956904 A US 95956904A US 2005044310 A1 US2005044310 A1 US 2005044310A1
Authority
US
United States
Prior art keywords
virtual volume
data
logical disk
pages
followed
Prior art date
Legal status (The legal status 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 status listed.)
Abandoned
Application number
US10/959,569
Inventor
Douglas Cameron
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
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 Individual filed Critical Individual
Priority to US10/959,569 priority Critical patent/US20050044310A1/en
Publication of US20050044310A1 publication Critical patent/US20050044310A1/en
Abandoned legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0628Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0662Virtualisation aspects
    • G06F3/0665Virtualisation aspects at area level, e.g. provisioning of virtual or logical volumes
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0602Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/0604Improving or facilitating administration, e.g. storage management
    • G06F3/0605Improving or facilitating administration, e.g. storage management by facilitating the interaction with a user or administrator
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0668Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/0671In-line storage system
    • G06F3/0683Plurality of storage devices
    • G06F3/0689Disk arrays, e.g. RAID, JBOD

Definitions

  • This invention relates to a utility storage server and more particularly to virtual volume management of the utility storage server.
  • a utility storage server may be defined as any carrier-class storage system that provisions physical storage to multiple users and/or applications. To meet the demands of multiple users and applications, a system administrator has to purchase enough physical storage for the users and the applications. Often the purchased physical storage is underutilized as the users and the applications generally fill their provisioned storage over time. Thus, what is needed is a method that allows the system administrator to increase asset utilization and defer expenses spent on the physical storage.
  • a method is provided to allow a system administrator of a utility storage server to provision virtual volumes several times larger than the amount of physical storage within the storage server.
  • a virtual volume is a virtual representation of multiple disks as a single large volume to a host or an application.
  • a virtual volume comprises an exception list containing the set of differences from a dummy base volume consisting of all zeros. This exception list can be made up of address tables that map virtual volume pages to logical disk pages. As storage demand grows, additional storage is allocated for the address tables and the data pages from separate pools of storage. If any of the pools runs low, more logical disk regions are allocated to that pool.
  • FIG. 1 illustrates a software architecture of a utility storage server in one embodiment of the invention.
  • FIG. 2 illustrates a representation of the mapping of pages in a virtual volume pages to pages in a logical disk of a node in one embodiment.
  • FIG. 3 illustrates the mapping of logical disk regions to physical disks in one embodiment of the invention.
  • FIG. 3A illustrates a virtual volume block address and its component segments in one embodiment of the invention.
  • FIGS. 4A, 4B , and 4 C illustrate a method for a virtual volume layer to respond to a write request to the virtual volume in one embodiment of the invention.
  • FIG. 5 illustrates a method for a system manager to respond to an event requesting an additional logical disk region from the virtual volume layer in one embodiment of the invention.
  • FIG. 6 illustrates a method for the virtual volume layer and the logical disk layer to respond to a read request from the host in one embodiment.
  • FIG. 1 illustrates a software architecture of a utility storage server 100 in one embodiment.
  • utility storage server 100 is shown to include a cluster of nodes 102 - 1 and 102 - 2 although the cluster may include additional nodes.
  • Node 102 - 1 includes a system manager 104 residing in the user level above the operating system and a data stack 106 - 1 residing in the kernel level of the operating system.
  • Data stack 106 - 1 includes a target driver 108 - 1 , a virtual volume layer 110 - 1 , a logical disk layer 112 - 1 , a physical disk layer 114 - 1 , and an initiator driver 116 - 1 .
  • Initiator driver 116 - 2 performs the actual reads and writes to the physical disk drive using, e.g., the SCSI protocol.
  • Physical disk layer 114 - 1 routes the read and write requests to the appropriate node with access to the physical disk drives on disk chassis 120 .
  • Logical disk layer 112 - 1 organizes “chunklets” of physical disks 202 - 1 to 202 - i ( FIG. 3 ) into logical disks (LDs) of specific RAID levels.
  • a chunklet is 256 megabytes of contiguous disk space.
  • System manager 104 allocates logical disk storage space, in units of logical disk regions (shown as a stripe through chunklets 204 - 1 to 204 - i from physical disks 202 - 1 to 202 - i in FIG. 3 ), to virtual volume layer 110 - 1 .
  • an LD region is 256 megabytes of logical disk storage space.
  • Virtual volume layer 110 - 1 divides up each LD region into pages for storing data.
  • a page has a size of 16 kilobytes and holds thirty-two 512 byte data blocks.
  • Virtual volume layer 110 - 1 maps pages in a virtual volume to pages in the logical disk and sends the read/write requests to the proper logical disk addresses within logical disk layer 112 - 1 .
  • Target driver 108 - 1 communicates the read and write requests to virtual volume layer 110 - 1 .
  • a host 118 sends read and write requests to the virtual volume via target driver 108 - 1 using, e.g., the SCSI protocol.
  • node 102 - 2 includes a data stack 106 - 2 residing in the kernel level.
  • Data stack 106 - 2 also includes a target driver 108 - 2 , a virtual volume layer 110 - 2 , a logical disk layer 112 - 2 , a physical disk layer 114 - 2 , and an initiator driver 116 - 2 .
  • Components of data stacks 106 - 1 and 106 - 2 communicate by a node-to-node link 122 .
  • System manager 104 resides only on one of the nodes of utility storage server 100 . If system manager 104 fails at one of the nodes, it can be restarted at another node. System manager 104 keeps a single system image of utility storage server 100 . System manager 104 also services events from the data stacks, delivers configuration commands to the data stacks, and records system configuration information in a table of contents (TOC) on a physical disk drive.
  • TOC table of contents
  • FIG. 2 illustrates the mapping of a virtual volume 208 to a logical disk (e.g., logical disk 207 in FIG. 3 ) in one embodiment of the invention.
  • Virtual volume 208 is an exception list made up of address tables that map pages of a virtual volume 208 to pages of logical disk 207 storage.
  • Virtual volume 208 also includes pools 412 and 414 of logical disk storage from which these exception list tables and logical disk data pages are allocated.
  • virtual volume 208 is implemented using a snapshot mechanism in which the exception list tracks changes to a dummy base volume consisting of only zeroes.
  • the address tables are divided into three levels. This is because virtual volume 208 is written or read in blocks each identified by a virtual volume block address.
  • the virtual volume block address includes a virtual volume page address consisting of parts 450 , 460 , and 470 ( FIG. 3A ). Parts 450 , 460 , and 470 lead to a page in virtual volume 208 , which is mapped to a corresponding page in logical disk 207 .
  • the virtual volume block address further includes a block offset 480 ( FIG. 3A ) that leads to a data block in the corresponding page in logical disk 207 .
  • a level 1 table (e.g., table 402 ) consists of entries that can be indexed by the first part 450 of the page address. Specifically, part 450 provides an offset from the start of the level 1 table. Each entry in the level 1 table stores a pointer to the start of a level 2 table that shares the first part of the page address.
  • Each of the level 2 tables (e.g., table 404 - 0 to 404 - 31 ) consists of entries that can be indexed by the second part 460 of the page addresses. Specifically, part 460 provides an offset from the start of a level 2 table. Each entry in the level 2 table stores a pointer to the start of a level 3 table that shares the first and the second part of the page address.
  • Each of the level 3 tables (e.g., tables 406 - 0 to 406 - 2047 in one embodiment) consists of entries that can be indexed by the third part 470 of the page addresses.
  • part 470 provides an offset from the start of a level 3 table.
  • Each entry in the level 3 table stores a pointer to a page in a logical disk.
  • a page in the virtual volume e.g., VV data page
  • a logical disk e.g., LD data page.
  • Part 480 of the page address identifies an offset of a data block (i.e., block offset) from the start of the LD data page.
  • Virtual volume layer 110 - 1 initially creates virtual volume 208 with only a blank level 1 table. As data is written to virtual volume 208 (described later), virtual volume layer 110 - 1 allocates LD data pages and adds the level 2 and level 3 tables that are necessary to manage these LD data pages.
  • FIGS. 4A, 4B , and 4 C illustrate a method 300 for virtual volume layer 110 - 1 to respond to a write request from host 118 (or an application) to virtual volume 208 in one embodiment.
  • step 302 virtual volume layer 110 - 1 receives a write request to a data block in virtual volume 208 .
  • the write request identifies the data block by the ID of virtual volume 208 and its virtual volume block address.
  • step 302 is followed by step 304 .
  • step 304 virtual volume layer 110 - 1 traverses tables 402 , 404 , and 406 to find the LD data page that corresponds to the virtual volume block address. Step 304 is followed by step 306 .
  • step 306 virtual volume layer 110 - 1 determines if the corresponding LD data page has been found. If so, step 306 is followed by step 308 . Otherwise step 306 is followed by step 312 .
  • the corresponding LD data page cannot be found if the VV data page to-be-written in virtual volume 208 has not been mapped to an LD data page by pointers and address tables as described above.
  • step 308 virtual volume layer 110 - 1 instructs logical disk layer 112 - 1 to issue a write to the corresponding LD data page.
  • virtual volume 110 - 1 identifies the block by a logical disk ID and an offset from the start of the logical disk.
  • the offset from the start of the logical disk is determined from the sum of (1) the offset of the LD region from the start of the logical disk, (2) the offset of the LD data page from the start of the LD region, and (3) the block offset of the data block from the start of the LD data page.
  • the write is replicated to the other nodes (e.g., node 102 - 2 ) for failover protection.
  • Step 308 is followed by step 310 .
  • step 310 virtual volume layer 110 - 1 returns a “pass” status to the host and ends method 300 .
  • step 312 virtual volume layer 110 - 1 determines if a level 2 (L 2 ) address table exists for the requested VV data page. If so, then step 312 is followed by step 326 . Otherwise step 312 is followed by step 314 .
  • L 2 level 2
  • step 314 virtual volume layer 110 - 1 determines if the number of available LD table pages in a pool 412 ( FIG. 2 ) for storing address tables is less than a threshold. If so, then step 314 is followed by step 316 . Otherwise step 314 is followed by step 322 .
  • step 316 virtual volume layer 110 - 1 issues an event to system manager 104 .
  • system manager 104 may allocate an available LD region to virtual volume layer 110 - 1 .
  • System manager 104 may allocate the LD region in a method 500 described later in reference to FIG. 5 .
  • Virtual volume layer 110 - 1 divides the allocated LD region into LD table pages and increments pool 412 with these new pages.
  • Step 316 is followed by step 318 .
  • step 318 virtual volume layer 110 - 1 determines if an LD table page is available to be used as an L 2 table. If so, then step 318 is followed by step 322 . Otherwise step 318 is followed by step 319 A.
  • step 319 A virtual volume layer 110 - 1 sleeps for a predetermined amount of time. Step 319 A is followed by step 319 B.
  • step 319 B virtual volume 110 - 1 determines if a timeout has been reached. If so, then step 319 B is followed by step 320 . Otherwise step 319 B is followed by step 318 .
  • step 320 virtual volume 110 - 1 returns a “fail” status to the host and ends method 300 .
  • step 322 virtual volume 110 - 1 creates an L 2 table from an available LD table page in pool 412 . Step 322 is followed by step 324 .
  • step 324 virtual volume 110 - 1 updates the level one (L 1 ) table with a pointer to the newly created L 2 table. Specifically, virtual volume 110 - 1 writes the pointer in the L 1 table entry having an offset identified by the first part of the virtual volume page address. Step 324 is followed by step 326 .
  • step 326 ( FIG. 4B ), virtual volume layer 110 - 1 determines if a level 3 (L 3 ) table exists for the requested VV data page. If so, then step 326 is followed by step 340 . Otherwise step 326 is followed by step 328 .
  • L 3 level 3
  • step 328 virtual volume layer 110 - 1 determines if the number of available LD table pages in pool 412 ( FIG. 2 ) for storing address tables is less than a threshold. If so, then step 328 is followed by step 330 . Otherwise step 328 is followed by step 336 .
  • step 330 virtual volume layer 110 - 1 issues an event to system manager 104 .
  • system manager 104 may allocate an available LD region to virtual volume layer 110 - 1 .
  • System manager 104 may allocate the LD region in a method 500 described later in reference to FIG. 5 .
  • Virtual volume layer 110 - 1 divides the LD region into LD table pages and increments pool 412 with these new pages.
  • Step 330 is followed by step 332 .
  • step 332 virtual volume layer 110 - 1 determines if an LD table page is available to be used as an L 3 table. If so, then step 332 is followed by step 336 . Otherwise step 332 is followed by step 333 A.
  • step 333 A virtual volume layer 110 - 1 sleeps for a predetermined amount of time. Step 333 A is followed by step 333 B.
  • step 333 B virtual volume 110 - 1 determines if a timeout has been reached. If so, then step 333 B is followed by step 334 . Otherwise step 333 B is followed by step 332 .
  • step 334 virtual volume 110 - 1 returns a “fail” status to the host and ends method 300 .
  • step 336 virtual volume 110 - 1 creates an L 3 table from an available page in pool 412 .
  • step 336 is followed by step 338 .
  • step 338 virtual volume 110 - 1 updates the corresponding L 2 table with a pointer to the newly created L 3 table. Specifically, virtual volume 110 - 1 writes the pointer in the L 2 table entry having an offset identified by the second part of the virtual volume page address. Step 338 is followed by step 340 .
  • step 340 ( FIG. 4C ), virtual volume layer 110 - 1 determines if an LD data page in the logical disk exists for the virtual volume block address received. If so, then step 340 is followed by step 352 . Otherwise step 340 is followed by step 342 .
  • step 342 virtual volume layer 110 - 1 determines if the number of available LD data pages in a pool 414 ( FIG. 2 ) for storing data is less than a threshold. If so, then step 342 is followed by step 344 . Otherwise step 342 is followed by step 350 .
  • step 344 virtual volume layer 110 - 1 issues an event to system manager 104 .
  • system manager 104 may allocate an available LD region to pool 414 .
  • System manager 104 may allocate the LD region in method 500 described later in reference to FIG. 5 .
  • Virtual volume layer 110 - 1 divides the LD region into LD data pages and increments pool 414 with these new pages. Step 344 is followed by step 346 .
  • step 346 virtual volume layer 110 - 1 determines if an LD data page is available to be used. If so, then step 346 is followed by step 350 . Otherwise step 346 is followed by step 347 A.
  • step 347 A virtual volume layer 110 - 1 sleeps for a predetermined amount of time. Step 347 A is followed by step 347 B.
  • step 347 B virtual volume layer 110 - 1 determines if a timeout has been reached. If so, then step 347 B is followed by step 348 . Otherwise step 347 B is followed by step 346 .
  • step 348 virtual volume layer 110 - 1 returns a “fail” status to the host and ends method 300 .
  • step 350 virtual volume layer 110 - 1 allocates an LD data page to virtual volume 208 from pool 414 to store data. Step 350 is followed by step 351 .
  • step 351 virtual volume layer 110 - 1 updates the corresponding L 3 table with a pointer to the new LD data page in virtual volume 208 . Specifically, virtual volume 110 - 1 writes the pointer in the L 3 table entry having an offset identified by the third part of the virtual volume page address. Step 351 is followed by step 352 .
  • step 352 virtual volume layer 110 - 1 writes the data into the new LD data page at an offset identified by the block offset of the virtual volume block address.
  • virtual volume layer 110 - 1 identifies the block by a logical disk ID and an offset from the start of the logical disk.
  • the offset from the start of the logical disk is determined from the sum of (1) the offset of the LD region from the starting address of the logical disk, (2) the offset of the LD data page from the starting address of the LD region, and (3) the block offset of the data block from the starting address of the LD data page.
  • Step 352 is followed by step 354 .
  • step 354 virtual volume 110 - 1 returns a “pass” status to the host and ends method 300 .
  • FIG. 5 illustrates a method 500 for system manager 104 to respond to the event from a virtual volume layer (e.g., virtual volume layer 110 - 1 ) in one embodiment.
  • a virtual volume layer e.g., virtual volume layer 110 - 1
  • system manager 104 validates the virtual volume ID and retrieves a data allocation control structure (DC) for the virtual volume identified by the virtual volume ID.
  • DC also known common provisioning group (CPG)
  • CPG common provisioning group
  • DC also sets the maximum physical allocation for each virtual volume, the maximum physical allocation of all the virtual volumes owned (i.e., controlled) by the DC, and warning points for the physical allocation of each virtual volume and the DC itself.
  • DC also sets the RAID characteristics of the logical disks and the set of nodes in a cluster from which the physical disk drive chunklets are allocated when creating additional logical disks.
  • step 504 system manager 104 determines if the physical allocation of the identified virtual volume (e.g., virtual volume 208 ) is over the maximum physical allocation specified by the DC.
  • the maximum physical allocation can have a default value or be set by the user. If the size of the virtual volume is over the maximum physical allocation, step 504 is followed by step 528 . If not, step 504 is followed by step 508 .
  • step 508 system manager 104 determines if one or more LD regions are available in existing logical disks (e.g., logical disk 207 ). If one or more LD regions are available, step 508 is followed by step 514 . If not, step 508 is followed by step 510 .
  • step 510 system manager 104 determines if the total physical allocation of the virtual volumes owned by the DC is over the maximum physical allocation specified by the DC. If the size of the DC is over the maximum physical allocation, step 510 is followed by step 528 . If not, step 510 is followed by step 512 .
  • step 512 system manager 104 instructs logical disk layer 112 - 1 to create one or more new logical disks from chunklets in the physical disks. Step 512 is followed by step 513 A.
  • step 513 A system manager 104 determines if the size of the DC is over a warning point specified by the DC.
  • the warning point provides an early warning to the user that the physical limit is approaching.
  • the warning point can have a default value or be set by the user. If the size of the DC is over the warning point, step 513 A is followed by step 513 B. If not, step 513 A is followed by step 514 .
  • step 513 B system manager 104 issues an allocation warning alert to the administrator of system 100 .
  • step 513 B is followed by step 514 .
  • step 514 system manager 104 allocates (e.g., assigns) one or more available LD regions in the logical disk (whether existing or newly created) to be mapped to virtual volume pages.
  • Each LD region is identified by a logical disk ID and an offset from the start of the logical disk.
  • Step 514 is followed by step 516 .
  • step 516 system manager 104 determines if the physical allocation of the virtual volume is over a warning point specified by the DC.
  • the warning point provides an early warning to the user that the physical limit is approaching.
  • the warning point can have a default value or be set by the user. If the size of the virtual volume is over the warning point, step 516 is followed by step 518 . If not, step 516 is followed by step 520 .
  • step 518 system manager 104 issues an allocation warning alert to the administrator of system 100 .
  • step 518 is followed by step 520 .
  • step 520 system manager 104 updates the table of contents (TOC).
  • TOC stores the organization of the virtual volumes, the logical disks, and the chunklets of server 100 on one or more physical disk drives. Step 520 is followed by step 522 .
  • step 522 system manager 104 delivers the one or more LD regions to virtual volume layer 110 - 1 .
  • virtual volume layer 110 - 1 divides the one or more LD regions into LD table or data pages and increments pool 412 or 414 with these new pages.
  • step 522 is followed by step 532 .
  • step 528 system manager 104 issues an allocation failure alert to the administrator of system 100 .
  • step 528 is followed by step 532 .
  • step 532 system manager 104 ends method 500 .
  • FIG. 6 illustrates a method 700 for a virtual volume layer (e.g., virtual volume layer 110 - 1 ) and a LD layer (e.g., logical disk layer 112 - 1 ) to respond to a read request from host 118 in one embodiment.
  • a virtual volume layer e.g., virtual volume layer 110 - 1
  • a LD layer e.g., logical disk layer 112 - 1
  • virtual volume layer 110 - 1 receives from host 118 a read request of a data block in a virtual volume (e.g., virtual volume 208 ).
  • the read request identifies the VV data block by a virtual volume ID and a virtual volume block address.
  • step 704 volume layer 110 - 1 traverses tables 402 , 404 , and 406 to find a LD data page corresponding to the virtual volume block address. As described above, if such an LD data page exists, the virtual volume block address can be mapped by a pointer identifying a logical disk ID and an offset from the start of the logical disk. Step 704 is followed by step 706 .
  • step 706 volume layer 110 - 1 determines if it has found the logical disk ID and the offset from the start of the logical disk. If so, step 706 is followed by step 710 . Otherwise step 706 is followed by step 708 .
  • virtual volume layer 110 - 1 returns all zeros to host 118 in response to the read request.
  • the virtual volume also referred to as a “Thin Provisioned Virtual Volume” or “TPVV”
  • TPVV Thin Provisioned Virtual Volume
  • the virtual volume is implemented using snapshot technology as a list of differences (exceptions) from a dummy base volume having all zeros.
  • TPVV Thin Provisioned Virtual Volume
  • virtual volume layer 110 - 1 will look to the dummy base volume for the requested data block having the specified virtual volume block address.
  • virtual volume layer 110 - 1 will find only zeros and thus return only zero data to host 118 .
  • step 710 virtual volume layer 110 - 1 issues a read command to the data block identified by the logical disk ID and the identified offset to logical disk layer 112 - 1 .
  • step 708 is followed by step 712 .
  • step 712 logical disk layer 112 - 1 performs a normal read to the data block identified by the logical disk ID and the identified offset. Step 712 is followed by step 714 .
  • step 714 logical disk layer 112 - 1 ends method 600 .

Abstract

A method is provided to allow a system administrator of a utility storage server to provision virtual volumes several times larger than the amount of physical storage within the storage server. A virtual volume is a virtual representation of multiple disks as a single large volume to a host or an application. In one embodiment, a virtual volume comprises an exception list containing the set of differences from dummy base volume consisting of all zeros. This exception list can be made up of address tables that map virtual volume pages to logical disk pages. As storage demand grows, additional storage is allocated for the address tables and the data pages from separate pools of storage. If any of the pools runs low, more logical disk regions are allocated to that pool.

Description

    CROSS REFERENCE TO RELATED APPLICATIONS
  • This is a Continuation of U.S. patent application Ser. No. 10/840,956 filed on May 7, 2004, which claims the benefit of U.S. Provisional Application No. 60/470,018 filed on May 12, 2003, which are hereby incorporated by reference.
  • FIELD OF INVENTION
  • This invention relates to a utility storage server and more particularly to virtual volume management of the utility storage server.
  • DESCRIPTION OF RELATED ART
  • A utility storage server may be defined as any carrier-class storage system that provisions physical storage to multiple users and/or applications. To meet the demands of multiple users and applications, a system administrator has to purchase enough physical storage for the users and the applications. Often the purchased physical storage is underutilized as the users and the applications generally fill their provisioned storage over time. Thus, what is needed is a method that allows the system administrator to increase asset utilization and defer expenses spent on the physical storage.
  • SUMMARY
  • In one embodiment of the invention, a method is provided to allow a system administrator of a utility storage server to provision virtual volumes several times larger than the amount of physical storage within the storage server. A virtual volume is a virtual representation of multiple disks as a single large volume to a host or an application. In one embodiment, a virtual volume comprises an exception list containing the set of differences from a dummy base volume consisting of all zeros. This exception list can be made up of address tables that map virtual volume pages to logical disk pages. As storage demand grows, additional storage is allocated for the address tables and the data pages from separate pools of storage. If any of the pools runs low, more logical disk regions are allocated to that pool.
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • FIG. 1 illustrates a software architecture of a utility storage server in one embodiment of the invention.
  • FIG. 2 illustrates a representation of the mapping of pages in a virtual volume pages to pages in a logical disk of a node in one embodiment.
  • FIG. 3 illustrates the mapping of logical disk regions to physical disks in one embodiment of the invention.
  • FIG. 3A illustrates a virtual volume block address and its component segments in one embodiment of the invention.
  • FIGS. 4A, 4B, and 4C illustrate a method for a virtual volume layer to respond to a write request to the virtual volume in one embodiment of the invention.
  • FIG. 5 illustrates a method for a system manager to respond to an event requesting an additional logical disk region from the virtual volume layer in one embodiment of the invention.
  • FIG. 6 illustrates a method for the virtual volume layer and the logical disk layer to respond to a read request from the host in one embodiment.
  • DETAILED DESCRIPTION
  • For a description of a utility storage server, please see U.S. Pat. No. 6,658,478, entitled “Data Storage System,” attorney docket no. M-8494 US, and U.S. patent application Ser. No. 09/883,681, entitled “Node Controller for a Data Storage System,” attorney docket no. M-8496 US, which are incorporated by reference in their entirety.
  • FIG. 1 illustrates a software architecture of a utility storage server 100 in one embodiment. For simplicity, utility storage server 100 is shown to include a cluster of nodes 102-1 and 102-2 although the cluster may include additional nodes.
  • Node 102-1 includes a system manager 104 residing in the user level above the operating system and a data stack 106-1 residing in the kernel level of the operating system. Data stack 106-1 includes a target driver 108-1, a virtual volume layer 110-1, a logical disk layer 112-1, a physical disk layer 114-1, and an initiator driver 116-1.
  • Initiator driver 116-2 performs the actual reads and writes to the physical disk drive using, e.g., the SCSI protocol. Physical disk layer 114-1 routes the read and write requests to the appropriate node with access to the physical disk drives on disk chassis 120.
  • Logical disk layer 112-1 organizes “chunklets” of physical disks 202-1 to 202-i (FIG. 3) into logical disks (LDs) of specific RAID levels. In one embodiment, a chunklet is 256 megabytes of contiguous disk space. System manager 104 allocates logical disk storage space, in units of logical disk regions (shown as a stripe through chunklets 204-1 to 204-i from physical disks 202-1 to 202-i in FIG. 3), to virtual volume layer 110-1. In one embodiment, an LD region is 256 megabytes of logical disk storage space. Virtual volume layer 110-1 divides up each LD region into pages for storing data. In one embodiment, a page has a size of 16 kilobytes and holds thirty-two 512 byte data blocks. Virtual volume layer 110-1 maps pages in a virtual volume to pages in the logical disk and sends the read/write requests to the proper logical disk addresses within logical disk layer 112-1.
  • Target driver 108-1 communicates the read and write requests to virtual volume layer 110-1. A host 118 sends read and write requests to the virtual volume via target driver 108-1 using, e.g., the SCSI protocol.
  • Similarly, node 102-2 includes a data stack 106-2 residing in the kernel level. Data stack 106-2 also includes a target driver 108-2, a virtual volume layer 110-2, a logical disk layer 112-2, a physical disk layer 114-2, and an initiator driver 116-2. Components of data stacks 106-1 and 106-2 communicate by a node-to-node link 122.
  • System manager 104 resides only on one of the nodes of utility storage server 100. If system manager 104 fails at one of the nodes, it can be restarted at another node. System manager 104 keeps a single system image of utility storage server 100. System manager 104 also services events from the data stacks, delivers configuration commands to the data stacks, and records system configuration information in a table of contents (TOC) on a physical disk drive.
  • FIG. 2 illustrates the mapping of a virtual volume 208 to a logical disk (e.g., logical disk 207 in FIG. 3) in one embodiment of the invention. Virtual volume 208 is an exception list made up of address tables that map pages of a virtual volume 208 to pages of logical disk 207 storage. Virtual volume 208 also includes pools 412 and 414 of logical disk storage from which these exception list tables and logical disk data pages are allocated. In one embodiment, virtual volume 208 is implemented using a snapshot mechanism in which the exception list tracks changes to a dummy base volume consisting of only zeroes.
  • The address tables are divided into three levels. This is because virtual volume 208 is written or read in blocks each identified by a virtual volume block address. The virtual volume block address includes a virtual volume page address consisting of parts 450, 460, and 470 (FIG. 3A). Parts 450, 460, and 470 lead to a page in virtual volume 208, which is mapped to a corresponding page in logical disk 207. The virtual volume block address further includes a block offset 480 (FIG. 3A) that leads to a data block in the corresponding page in logical disk 207.
  • A level 1 table (e.g., table 402) consists of entries that can be indexed by the first part 450 of the page address. Specifically, part 450 provides an offset from the start of the level 1 table. Each entry in the level 1 table stores a pointer to the start of a level 2 table that shares the first part of the page address.
  • Each of the level 2 tables (e.g., table 404-0 to 404-31) consists of entries that can be indexed by the second part 460 of the page addresses. Specifically, part 460 provides an offset from the start of a level 2 table. Each entry in the level 2 table stores a pointer to the start of a level 3 table that shares the first and the second part of the page address.
  • Each of the level 3 tables (e.g., tables 406-0 to 406-2047 in one embodiment) consists of entries that can be indexed by the third part 470 of the page addresses. Specifically, part 470 provides an offset from the start of a level 3 table. Each entry in the level 3 table stores a pointer to a page in a logical disk. Accordingly, a page in the virtual volume (e.g., VV data page) is mapped to a page in a logical disk (e.g., LD data page). Part 480 of the page address identifies an offset of a data block (i.e., block offset) from the start of the LD data page.
  • Virtual volume layer 110-1 initially creates virtual volume 208 with only a blank level 1 table. As data is written to virtual volume 208 (described later), virtual volume layer 110-1 allocates LD data pages and adds the level 2 and level 3 tables that are necessary to manage these LD data pages.
  • FIGS. 4A, 4B, and 4C illustrate a method 300 for virtual volume layer 110-1 to respond to a write request from host 118 (or an application) to virtual volume 208 in one embodiment.
  • In step 302, virtual volume layer 110-1 receives a write request to a data block in virtual volume 208. The write request identifies the data block by the ID of virtual volume 208 and its virtual volume block address. Step 302 is followed by step 304.
  • In step 304, virtual volume layer 110-1 traverses tables 402, 404, and 406 to find the LD data page that corresponds to the virtual volume block address. Step 304 is followed by step 306.
  • In step 306, virtual volume layer 110-1 determines if the corresponding LD data page has been found. If so, step 306 is followed by step 308. Otherwise step 306 is followed by step 312. The corresponding LD data page cannot be found if the VV data page to-be-written in virtual volume 208 has not been mapped to an LD data page by pointers and address tables as described above.
  • In step 308, virtual volume layer 110-1 instructs logical disk layer 112-1 to issue a write to the corresponding LD data page. Specifically, virtual volume 110-1 identifies the block by a logical disk ID and an offset from the start of the logical disk. The offset from the start of the logical disk is determined from the sum of (1) the offset of the LD region from the start of the logical disk, (2) the offset of the LD data page from the start of the LD region, and (3) the block offset of the data block from the start of the LD data page. In one embodiment, the write is replicated to the other nodes (e.g., node 102-2) for failover protection. Step 308 is followed by step 310.
  • In step 310, virtual volume layer 110-1 returns a “pass” status to the host and ends method 300.
  • In step 312, virtual volume layer 110-1 determines if a level 2 (L2) address table exists for the requested VV data page. If so, then step 312 is followed by step 326. Otherwise step 312 is followed by step 314.
  • In step 314, virtual volume layer 110-1 determines if the number of available LD table pages in a pool 412 (FIG. 2) for storing address tables is less than a threshold. If so, then step 314 is followed by step 316. Otherwise step 314 is followed by step 322.
  • In step 316, virtual volume layer 110-1 issues an event to system manager 104. In response to the event, system manager 104 may allocate an available LD region to virtual volume layer 110-1. System manager 104 may allocate the LD region in a method 500 described later in reference to FIG. 5. Virtual volume layer 110-1 divides the allocated LD region into LD table pages and increments pool 412 with these new pages. Step 316 is followed by step 318.
  • In step 318, virtual volume layer 110-1 determines if an LD table page is available to be used as an L2 table. If so, then step 318 is followed by step 322. Otherwise step 318 is followed by step 319A.
  • In step 319A, virtual volume layer 110-1 sleeps for a predetermined amount of time. Step 319A is followed by step 319B.
  • In step 319B, virtual volume 110-1 determines if a timeout has been reached. If so, then step 319B is followed by step 320. Otherwise step 319B is followed by step 318.
  • In step 320, virtual volume 110-1 returns a “fail” status to the host and ends method 300.
  • In step 322, virtual volume 110-1 creates an L2 table from an available LD table page in pool 412. Step 322 is followed by step 324.
  • In step 324, virtual volume 110-1 updates the level one (L1) table with a pointer to the newly created L2 table. Specifically, virtual volume 110-1 writes the pointer in the L1 table entry having an offset identified by the first part of the virtual volume page address. Step 324 is followed by step 326.
  • In step 326 (FIG. 4B), virtual volume layer 110-1 determines if a level 3 (L3) table exists for the requested VV data page. If so, then step 326 is followed by step 340. Otherwise step 326 is followed by step 328.
  • In step 328, virtual volume layer 110-1 determines if the number of available LD table pages in pool 412 (FIG. 2) for storing address tables is less than a threshold. If so, then step 328 is followed by step 330. Otherwise step 328 is followed by step 336.
  • In step 330, virtual volume layer 110-1 issues an event to system manager 104. In response to the event, system manager 104 may allocate an available LD region to virtual volume layer 110-1. System manager 104 may allocate the LD region in a method 500 described later in reference to FIG. 5. Virtual volume layer 110-1 divides the LD region into LD table pages and increments pool 412 with these new pages. Step 330 is followed by step 332.
  • In step 332, virtual volume layer 110-1 determines if an LD table page is available to be used as an L3 table. If so, then step 332 is followed by step 336. Otherwise step 332 is followed by step 333A.
  • In step 333A, virtual volume layer 110-1 sleeps for a predetermined amount of time. Step 333A is followed by step 333B.
  • In step 333B, virtual volume 110-1 determines if a timeout has been reached. If so, then step 333B is followed by step 334. Otherwise step 333B is followed by step 332.
  • In step 334, virtual volume 110-1 returns a “fail” status to the host and ends method 300.
  • In step 336, virtual volume 110-1 creates an L3 table from an available page in pool 412. Step 336 is followed by step 338.
  • In step 338, virtual volume 110-1 updates the corresponding L2 table with a pointer to the newly created L3 table. Specifically, virtual volume 110-1 writes the pointer in the L2 table entry having an offset identified by the second part of the virtual volume page address. Step 338 is followed by step 340.
  • In step 340 (FIG. 4C), virtual volume layer 110-1 determines if an LD data page in the logical disk exists for the virtual volume block address received. If so, then step 340 is followed by step 352. Otherwise step 340 is followed by step 342.
  • In step 342, virtual volume layer 110-1 determines if the number of available LD data pages in a pool 414 (FIG. 2) for storing data is less than a threshold. If so, then step 342 is followed by step 344. Otherwise step 342 is followed by step 350.
  • In step 344, virtual volume layer 110-1 issues an event to system manager 104. In response to the event, system manager 104 may allocate an available LD region to pool 414. System manager 104 may allocate the LD region in method 500 described later in reference to FIG. 5. Virtual volume layer 110-1 divides the LD region into LD data pages and increments pool 414 with these new pages. Step 344 is followed by step 346.
  • In step 346, virtual volume layer 110-1 determines if an LD data page is available to be used. If so, then step 346 is followed by step 350. Otherwise step 346 is followed by step 347A.
  • In step 347A, virtual volume layer 110-1 sleeps for a predetermined amount of time. Step 347A is followed by step 347B.
  • In step 347B, virtual volume layer 110-1 determines if a timeout has been reached. If so, then step 347B is followed by step 348. Otherwise step 347B is followed by step 346.
  • In step 348, virtual volume layer 110-1 returns a “fail” status to the host and ends method 300.
  • In step 350, virtual volume layer 110-1 allocates an LD data page to virtual volume 208 from pool 414 to store data. Step 350 is followed by step 351.
  • In step 351, virtual volume layer 110-1 updates the corresponding L3 table with a pointer to the new LD data page in virtual volume 208. Specifically, virtual volume 110-1 writes the pointer in the L3 table entry having an offset identified by the third part of the virtual volume page address. Step 351 is followed by step 352.
  • In step 352, virtual volume layer 110-1 writes the data into the new LD data page at an offset identified by the block offset of the virtual volume block address. Specifically, virtual volume layer 110-1 identifies the block by a logical disk ID and an offset from the start of the logical disk. The offset from the start of the logical disk is determined from the sum of (1) the offset of the LD region from the starting address of the logical disk, (2) the offset of the LD data page from the starting address of the LD region, and (3) the block offset of the data block from the starting address of the LD data page. Step 352 is followed by step 354.
  • In step 354, virtual volume 110-1 returns a “pass” status to the host and ends method 300.
  • FIG. 5 illustrates a method 500 for system manager 104 to respond to the event from a virtual volume layer (e.g., virtual volume layer 110-1) in one embodiment.
  • In step 502, system manager 104 validates the virtual volume ID and retrieves a data allocation control structure (DC) for the virtual volume identified by the virtual volume ID. DC, also known common provisioning group (CPG), is a part of system manager 104 that sets the maximum physical allocation for each virtual volume, the maximum physical allocation of all the virtual volumes owned (i.e., controlled) by the DC, and warning points for the physical allocation of each virtual volume and the DC itself. DC also sets the RAID characteristics of the logical disks and the set of nodes in a cluster from which the physical disk drive chunklets are allocated when creating additional logical disks.
  • In step 504, system manager 104 determines if the physical allocation of the identified virtual volume (e.g., virtual volume 208) is over the maximum physical allocation specified by the DC. The maximum physical allocation can have a default value or be set by the user. If the size of the virtual volume is over the maximum physical allocation, step 504 is followed by step 528. If not, step 504 is followed by step 508.
  • In step 508, system manager 104 determines if one or more LD regions are available in existing logical disks (e.g., logical disk 207). If one or more LD regions are available, step 508 is followed by step 514. If not, step 508 is followed by step 510.
  • In step 510, system manager 104 determines if the total physical allocation of the virtual volumes owned by the DC is over the maximum physical allocation specified by the DC. If the size of the DC is over the maximum physical allocation, step 510 is followed by step 528. If not, step 510 is followed by step 512.
  • In step 512, system manager 104 instructs logical disk layer 112-1 to create one or more new logical disks from chunklets in the physical disks. Step 512 is followed by step 513A.
  • In step 513A, system manager 104 determines if the size of the DC is over a warning point specified by the DC. The warning point provides an early warning to the user that the physical limit is approaching. The warning point can have a default value or be set by the user. If the size of the DC is over the warning point, step 513A is followed by step 513B. If not, step 513A is followed by step 514.
  • In step 513B, system manager 104 issues an allocation warning alert to the administrator of system 100. Step 513B is followed by step 514.
  • In step 514, system manager 104 allocates (e.g., assigns) one or more available LD regions in the logical disk (whether existing or newly created) to be mapped to virtual volume pages. Each LD region is identified by a logical disk ID and an offset from the start of the logical disk. Step 514 is followed by step 516.
  • In step 516, system manager 104 determines if the physical allocation of the virtual volume is over a warning point specified by the DC. The warning point provides an early warning to the user that the physical limit is approaching. The warning point can have a default value or be set by the user. If the size of the virtual volume is over the warning point, step 516 is followed by step 518. If not, step 516 is followed by step 520.
  • In step 518, system manager 104 issues an allocation warning alert to the administrator of system 100. Step 518 is followed by step 520.
  • In step 520, system manager 104 updates the table of contents (TOC). TOC stores the organization of the virtual volumes, the logical disks, and the chunklets of server 100 on one or more physical disk drives. Step 520 is followed by step 522.
  • In step 522, system manager 104 delivers the one or more LD regions to virtual volume layer 110-1. As described above, virtual volume layer 110-1 divides the one or more LD regions into LD table or data pages and increments pool 412 or 414 with these new pages. Step 522 is followed by step 532.
  • In step 528, system manager 104 issues an allocation failure alert to the administrator of system 100. Step 528 is followed by step 532.
  • In step 532, system manager 104 ends method 500.
  • FIG. 6 illustrates a method 700 for a virtual volume layer (e.g., virtual volume layer 110-1) and a LD layer (e.g., logical disk layer 112-1) to respond to a read request from host 118 in one embodiment.
  • In step 702, virtual volume layer 110-1 receives from host 118 a read request of a data block in a virtual volume (e.g., virtual volume 208). The read request identifies the VV data block by a virtual volume ID and a virtual volume block address.
  • In step 704, volume layer 110-1 traverses tables 402, 404, and 406 to find a LD data page corresponding to the virtual volume block address. As described above, if such an LD data page exists, the virtual volume block address can be mapped by a pointer identifying a logical disk ID and an offset from the start of the logical disk. Step 704 is followed by step 706.
  • In step 706, volume layer 110-1 determines if it has found the logical disk ID and the offset from the start of the logical disk. If so, step 706 is followed by step 710. Otherwise step 706 is followed by step 708.
  • In step 708, virtual volume layer 110-1 returns all zeros to host 118 in response to the read request. In one embodiment when the virtual volume (also referred to as a “Thin Provisioned Virtual Volume” or “TPVV”) is implemented using snapshot technology as a list of differences (exceptions) from a dummy base volume having all zeros. When a data block for a read request cannot be found in the virtual volume exception list, virtual volume layer 110-1 will look to the dummy base volume for the requested data block having the specified virtual volume block address. When searching for the requested data block in the dummy base volume, virtual volume layer 110-1 will find only zeros and thus return only zero data to host 118. For a description of a snapshot implemented as an exception list, please see U.S. patent application Ser. No. 10/655,951, entitled “Time-And-Space Efficient Mechanism To Create Virtual Storage Volume Copies,” attorney docket no. 3PD-M-8497 US; U.S. patent application Ser. No. 10/655,963, entitled “Efficient And Reliable Virtual Volume Mapping,” attorney docket no. 3PD-M-8498 US; and U.S. patent application Ser. No. 10/655,961, entitled “Read/Write Snapshot,” attorney docket no. 3PD-P100, which are incorporated by reference in their entirety. Step 708 is followed by step 714.
  • In step 710, virtual volume layer 110-1 issues a read command to the data block identified by the logical disk ID and the identified offset to logical disk layer 112-1. Step 708 is followed by step 712.
  • In step 712, logical disk layer 112-1 performs a normal read to the data block identified by the logical disk ID and the identified offset. Step 712 is followed by step 714.
  • In step 714, logical disk layer 112-1 ends method 600.
  • Various other adaptations and combinations of features of the embodiments disclosed are within the scope of the invention. Numerous embodiments are encompassed by the following claims.

Claims (6)

1. A method for writing to a virtual volume that is mapped to at least one logical disk, the virtual volume comprising an exception list, the method comprising:
receiving a request to write a data to the virtual volume at an address;
traversing the exception list with a path based on a portion of the address to find a pointer;
if the pointer does not exist,
determining if a number of available data pages in a data page pool is less than a threshold;
if the number of available data pages in the data page pool is not less than a threshold,
allocating one data page from the data page pool to form the data page;
writing the pointer in the exception list with the path based on the portion of the address;
writing the data in the data page at a location based on another portion of the address.
2. The method of claim 1, further comprising:
if the pointer exists,
following the pointer to the data page;
writing the data in the data page at the location based on another portion of the address.
3. The method of claim 1, further comprising:
if the number of available data pages in the data page pool is less than the threshold,
requesting additional data pages to be added to the data page pool;
determining if the additional data pages have been added before timing out;
if the additional data pages have been added before timing out,
allocating one data page from the data page pool to form the data page;
writing the pointer in the exception list with the path based on the portion of the address;
writing the data in the data page at the location based on another portion of the address.
4. The method of claim 3, further comprising:
if the additional data pages have not been added before timing out, returning a write failure message.
5. The method of claim 1, in response to said requesting additional data pages to be added to the data page pool, further comprising:
determining if said one logical disk has one storage unit (“logical disk region”) available;
if said one logic disk has one logical disk region available,
dividing said one logical disk region into data pages;
allocating the data pages to the data page pool.
6. The method of claim 4, further comprising:
if said one logic disk does not have one logical disk region available,
creating a new logical disk;
dividing a logical disk region from the new logical disk into data pages;
allocating the data pages to the data page pool.
US10/959,569 2003-05-12 2004-10-05 Method of managing virtual volumes in a utility storage server system Abandoned US20050044310A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US10/959,569 US20050044310A1 (en) 2003-05-12 2004-10-05 Method of managing virtual volumes in a utility storage server system

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US47001803P 2003-05-12 2003-05-12
US10/840,956 US6823442B1 (en) 2003-05-12 2004-05-07 Method of managing virtual volumes in a utility storage server system
US10/959,569 US20050044310A1 (en) 2003-05-12 2004-10-05 Method of managing virtual volumes in a utility storage server system

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US10/840,956 Continuation US6823442B1 (en) 2003-05-12 2004-05-07 Method of managing virtual volumes in a utility storage server system

Publications (1)

Publication Number Publication Date
US20050044310A1 true US20050044310A1 (en) 2005-02-24

Family

ID=33423942

Family Applications (2)

Application Number Title Priority Date Filing Date
US10/840,956 Active US6823442B1 (en) 2003-05-12 2004-05-07 Method of managing virtual volumes in a utility storage server system
US10/959,569 Abandoned US20050044310A1 (en) 2003-05-12 2004-10-05 Method of managing virtual volumes in a utility storage server system

Family Applications Before (1)

Application Number Title Priority Date Filing Date
US10/840,956 Active US6823442B1 (en) 2003-05-12 2004-05-07 Method of managing virtual volumes in a utility storage server system

Country Status (2)

Country Link
US (2) US6823442B1 (en)
WO (1) WO2004102391A1 (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070061540A1 (en) * 2005-06-06 2007-03-15 Jim Rafert Data storage system using segmentable virtual volumes
US20070174673A1 (en) * 2006-01-25 2007-07-26 Tomohiro Kawaguchi Storage system and data restoration method thereof
US20100100700A1 (en) * 2008-10-16 2010-04-22 International Business Machines Corporation Adaptively preventing out of memory conditions
US20110082834A1 (en) * 2007-01-24 2011-04-07 Hitachi, Ltd. Storage control device to backup data stored in virtual volume
US20110191563A1 (en) * 2010-02-02 2011-08-04 International Business Machines Corporation On demand conversion of standard logical volumes to thin-provisioned logical volumes
US20130173880A1 (en) * 2011-12-28 2013-07-04 International Business Machines Corporation Dedicated large page memory pools
US8825963B1 (en) * 2010-01-06 2014-09-02 Netapp, Inc. Dynamic balancing of performance with block sharing in a storage system

Families Citing this family (52)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8032701B1 (en) * 2004-03-26 2011-10-04 Emc Corporation System and method for managing provisioning of storage resources in a network with virtualization of resources in such a network
US7162600B2 (en) * 2005-03-29 2007-01-09 Hitachi, Ltd. Data copying method and apparatus in a thin provisioned system
JP4283004B2 (en) * 2003-02-04 2009-06-24 株式会社日立製作所 Disk control device and control method of disk control device
JP4646526B2 (en) 2004-02-18 2011-03-09 株式会社日立製作所 Storage control system and control method thereof
US7130960B1 (en) * 2005-04-21 2006-10-31 Hitachi, Ltd. System and method for managing disk space in a thin-provisioned storage subsystem
JP4699808B2 (en) * 2005-06-02 2011-06-15 株式会社日立製作所 Storage system and configuration change method
US8055938B1 (en) * 2005-06-10 2011-11-08 American Megatrends, Inc. Performance in virtual tape libraries
JP4723921B2 (en) * 2005-06-13 2011-07-13 株式会社日立製作所 Storage control device and control method thereof
US20070112895A1 (en) * 2005-11-04 2007-05-17 Sun Microsystems, Inc. Block-based incremental backup
JP4927412B2 (en) * 2006-02-10 2012-05-09 株式会社日立製作所 Storage control method and control method thereof
JP4993928B2 (en) 2006-03-23 2012-08-08 株式会社日立製作所 Storage system, storage area release method, and storage system
JP4940738B2 (en) * 2006-04-12 2012-05-30 株式会社日立製作所 Storage area dynamic allocation method
JP5037881B2 (en) * 2006-04-18 2012-10-03 株式会社日立製作所 Storage system and control method thereof
JP2007304794A (en) * 2006-05-10 2007-11-22 Hitachi Ltd Storage system and storage control method in storage system
JP2007316725A (en) * 2006-05-23 2007-12-06 Hitachi Ltd Storage area management method and management computer
GB0613660D0 (en) * 2006-07-08 2006-08-16 Ibm Reserve pool management in virtualized storage systems
JP4975396B2 (en) * 2006-08-24 2012-07-11 株式会社日立製作所 Storage control device and storage control method
US7590812B2 (en) * 2006-09-25 2009-09-15 International Business Machines Corporation Apparatus, system, and method for archiving a log
JP2008097502A (en) * 2006-10-16 2008-04-24 Hitachi Ltd Capacity monitoring method and computer system
US7949847B2 (en) * 2006-11-29 2011-05-24 Hitachi, Ltd. Storage extent allocation method for thin provisioning storage
EP1931108B1 (en) * 2006-12-08 2011-02-09 Deutsche Telekom AG Method and system for peer-to-peer content dissemination
JP2008146574A (en) 2006-12-13 2008-06-26 Hitachi Ltd Storage controller and storage control method
US20080282047A1 (en) * 2007-05-08 2008-11-13 Hitachi, Ltd. Methods and apparatus to backup and restore data for virtualized storage area
US8177922B2 (en) * 2007-09-04 2012-05-15 Hitachi Metals, Ltd. R-Fe-B anisotropic sintered magnet
JP5481030B2 (en) 2008-01-30 2014-04-23 ルネサスエレクトロニクス株式会社 Semiconductor device
US8244868B2 (en) * 2008-03-24 2012-08-14 International Business Machines Corporation Thin-provisioning adviser for storage devices
JP4616899B2 (en) * 2008-05-26 2011-01-19 株式会社日立製作所 Management server, pool expansion method and computer system
US8024442B1 (en) * 2008-07-08 2011-09-20 Network Appliance, Inc. Centralized storage management for multiple heterogeneous host-side servers
JP2010072753A (en) * 2008-09-16 2010-04-02 Hitachi Ltd Storage system equipped with automatic extension volume and power saving function
US8359444B2 (en) 2008-09-24 2013-01-22 Hitachi, Ltd. System and method for controlling automated page-based tier management in storage systems
JP4762289B2 (en) * 2008-10-01 2011-08-31 株式会社日立製作所 A storage system that controls allocation of storage areas to virtual volumes that store specific pattern data
JP5138530B2 (en) * 2008-10-08 2013-02-06 株式会社日立製作所 Fault management method in storage capacity virtualization technology
US8504524B1 (en) * 2009-02-09 2013-08-06 American Megatrends, Inc. Reclaiming space from a file system
CN107247565B (en) 2009-03-18 2020-06-09 株式会社日立制作所 Storage control device and virtual volume control method
DE202009018958U1 (en) 2009-03-18 2014-10-22 Hitachi, Ltd. Memory controller
US8539184B2 (en) 2009-09-23 2013-09-17 International Business Machines Corporation On demand storage group management with recapture
CN102549542A (en) 2009-10-09 2012-07-04 株式会社日立制作所 Storage system and control method thereof, implementing data reallocation in case of load bias
US8402215B2 (en) * 2009-10-15 2013-03-19 Hitachi, Ltd. Storage system, storage apparatus, and optimization method of storage areas of storage system having page migration
JP5378604B2 (en) * 2010-04-27 2013-12-25 株式会社日立製作所 Mainframe storage device to which Thin Provisioning is applied
WO2011135617A1 (en) 2010-04-27 2011-11-03 株式会社日立製作所 Method and storage device for limiting total capacity of virtual volumes associated with pool
US20110276776A1 (en) * 2010-05-07 2011-11-10 Tatu Ylonen Oy Ltd Addressing for Huge Direct-Mapped Object Systems
US8356147B2 (en) 2010-08-20 2013-01-15 Hitachi, Ltd. Tiered storage pool management and control for loosely coupled multiple storage environment
US8468303B2 (en) 2010-09-27 2013-06-18 Hitachi, Ltd. Method and apparatus to allocate area to virtual volume based on object access type
US8683120B2 (en) 2011-03-28 2014-03-25 Hitachi, Ltd. Method and apparatus to allocate area to virtual volume
US9430367B1 (en) * 2011-04-18 2016-08-30 American Megatrends, Inc. Systems and methods for active raid
US8806154B1 (en) 2011-05-06 2014-08-12 Chelsio Communications, Inc. Thin provisioning row snapshot with reference count map
US8913335B2 (en) * 2011-05-23 2014-12-16 HGST Netherlands B.V. Storage device with shingled data and unshingled cache regions
US9009437B1 (en) * 2011-06-20 2015-04-14 Emc Corporation Techniques for shared data storage provisioning with thin devices
US9229637B2 (en) 2011-11-18 2016-01-05 Hitachi, Ltd. Volume copy management method on thin provisioning pool of storage subsystem
US9146853B2 (en) 2013-03-28 2015-09-29 Microsoft Technology Licensing, Llc Managing capacity of a thinly provisioned storage system
CN106462491B (en) * 2014-12-31 2020-08-14 华为技术有限公司 Management method of stored data, storage manager and storage system
US10459663B2 (en) 2017-01-23 2019-10-29 International Business Machines Corporation Thin provisioning of raid storage

Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5394537A (en) * 1989-12-13 1995-02-28 Texas Instruments Incorporated Adaptive page placement memory management system
US5537585A (en) * 1994-02-25 1996-07-16 Avail Systems Corporation Data storage management for network interconnected processors
US6092157A (en) * 1995-07-11 2000-07-18 Canon Kabushiki Kaisha Processing method, apparatus and system for memory management using shared memories with common data areas
US6209070B1 (en) * 1998-03-20 2001-03-27 International Business Machines Corporation System for multiple data transfer operations
US6260129B1 (en) * 1998-09-08 2001-07-10 International Business Machines Corportion Management of fixed pages in memory for input/output operations
US6421787B1 (en) * 1998-05-12 2002-07-16 Sun Microsystems, Inc. Highly available cluster message passing facility
US20020112113A1 (en) * 2001-01-11 2002-08-15 Yotta Yotta, Inc. Storage virtualization system and methods
US6477612B1 (en) * 2000-02-08 2002-11-05 Microsoft Corporation Providing access to physical memory allocated to a process by selectively mapping pages of the physical memory with virtual memory allocated to the process
US20030140210A1 (en) * 2001-12-10 2003-07-24 Richard Testardi Dynamic and variable length extents
US20030204597A1 (en) * 2002-04-26 2003-10-30 Hitachi, Inc. Storage system having virtualized resource
US6658478B1 (en) * 2000-08-04 2003-12-02 3Pardata, Inc. Data storage system
US20030225884A1 (en) * 2002-05-31 2003-12-04 Hayden Mark G. Distributed network storage system with virtualization
US20040111485A1 (en) * 2002-12-09 2004-06-10 Yasuyuki Mimatsu Connecting device of storage device and computer system including the same connecting device

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4947319A (en) * 1988-09-15 1990-08-07 International Business Machines Corporation Arbitral dynamic cache using processor storage
US6421767B1 (en) * 2000-02-23 2002-07-16 Storage Technology Corporation Method and apparatus for managing a storage system using snapshot copy operations with snap groups
JP2003202964A (en) * 2002-01-09 2003-07-18 Hitachi Ltd Method for controlling computer system, computer system, method for controlling storage device and storage device

Patent Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5394537A (en) * 1989-12-13 1995-02-28 Texas Instruments Incorporated Adaptive page placement memory management system
US5537585A (en) * 1994-02-25 1996-07-16 Avail Systems Corporation Data storage management for network interconnected processors
US6092157A (en) * 1995-07-11 2000-07-18 Canon Kabushiki Kaisha Processing method, apparatus and system for memory management using shared memories with common data areas
US6209070B1 (en) * 1998-03-20 2001-03-27 International Business Machines Corporation System for multiple data transfer operations
US6421787B1 (en) * 1998-05-12 2002-07-16 Sun Microsystems, Inc. Highly available cluster message passing facility
US6260129B1 (en) * 1998-09-08 2001-07-10 International Business Machines Corportion Management of fixed pages in memory for input/output operations
US6477612B1 (en) * 2000-02-08 2002-11-05 Microsoft Corporation Providing access to physical memory allocated to a process by selectively mapping pages of the physical memory with virtual memory allocated to the process
US6658478B1 (en) * 2000-08-04 2003-12-02 3Pardata, Inc. Data storage system
US20020112113A1 (en) * 2001-01-11 2002-08-15 Yotta Yotta, Inc. Storage virtualization system and methods
US20030140210A1 (en) * 2001-12-10 2003-07-24 Richard Testardi Dynamic and variable length extents
US20030204597A1 (en) * 2002-04-26 2003-10-30 Hitachi, Inc. Storage system having virtualized resource
US20030225884A1 (en) * 2002-05-31 2003-12-04 Hayden Mark G. Distributed network storage system with virtualization
US20050010618A1 (en) * 2002-05-31 2005-01-13 Lefthand Networks, Inc. Distributed Network Storage System With Virtualization
US20040111485A1 (en) * 2002-12-09 2004-06-10 Yasuyuki Mimatsu Connecting device of storage device and computer system including the same connecting device

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070061540A1 (en) * 2005-06-06 2007-03-15 Jim Rafert Data storage system using segmentable virtual volumes
US20070174673A1 (en) * 2006-01-25 2007-07-26 Tomohiro Kawaguchi Storage system and data restoration method thereof
US7594137B2 (en) * 2006-01-25 2009-09-22 Hitachi, Ltd. Storage system and data restoration method thereof
US20100011239A1 (en) * 2006-01-25 2010-01-14 Tomohiro Kawaguchi Storage system and data restoration method thereof
US8453011B2 (en) 2006-01-25 2013-05-28 Hitachi, Ltd. Storage system and data restoration method thereof
US8010837B2 (en) 2006-01-25 2011-08-30 Hitachi, Ltd. Storage sub system and data restoration method thereof
US20110082834A1 (en) * 2007-01-24 2011-04-07 Hitachi, Ltd. Storage control device to backup data stored in virtual volume
US8271444B2 (en) * 2007-01-24 2012-09-18 Hitachi, Ltd. Storage control device to backup data stored in virtual volume
US8245012B2 (en) * 2008-10-16 2012-08-14 International Business Machines Corporation Adaptively preventing out of memory conditions
US20100100700A1 (en) * 2008-10-16 2010-04-22 International Business Machines Corporation Adaptively preventing out of memory conditions
US8825963B1 (en) * 2010-01-06 2014-09-02 Netapp, Inc. Dynamic balancing of performance with block sharing in a storage system
US20110191563A1 (en) * 2010-02-02 2011-08-04 International Business Machines Corporation On demand conversion of standard logical volumes to thin-provisioned logical volumes
US8578087B2 (en) 2010-02-02 2013-11-05 International Business Machines Corporation On demand conversion of standard logical volumes to thin-provisioned logical volumes
US8656096B2 (en) 2010-02-02 2014-02-18 International Business Machines Corporation On demand conversion of standard logical volumes to thin-provisioned logical volumes
US20130173880A1 (en) * 2011-12-28 2013-07-04 International Business Machines Corporation Dedicated large page memory pools
US8868876B2 (en) * 2011-12-28 2014-10-21 International Business Machines Corporation Dedicated large page memory pools

Also Published As

Publication number Publication date
US6823442B1 (en) 2004-11-23
US20040230766A1 (en) 2004-11-18
WO2004102391A1 (en) 2004-11-25

Similar Documents

Publication Publication Date Title
US6823442B1 (en) Method of managing virtual volumes in a utility storage server system
US9946460B2 (en) Storage subsystem and storage system architecture performing storage virtualization and method thereof
US7392365B2 (en) Dynamically changeable virtual mapping scheme
US7650480B2 (en) Storage system and write distribution method
US9696932B1 (en) Virtual provisioning space reservation
US8984248B2 (en) Data migration system and data migration method
US9152349B2 (en) Automated information life-cycle management with thin provisioning
US8627038B2 (en) Storage controller and storage control method
US6804755B2 (en) Apparatus and method for performing an instant copy of data based on a dynamically changeable virtual mapping scheme
US6779094B2 (en) Apparatus and method for instant copy of data by writing new data to an additional physical storage area
US6779095B2 (en) Apparatus and method for instant copy of data using pointers to new and original data in a data location
US7594084B2 (en) File storage control device and method
US8131969B2 (en) Updating system configuration information
US8694563B1 (en) Space recovery for thin-provisioned storage volumes
US20070174574A1 (en) System and method for managing disk space in a thin-provisioned storage subsystem
US7987328B2 (en) Data archive system
US20080104150A1 (en) Method and system for priority-based allocation in a storage pool
US7032093B1 (en) On-demand allocation of physical storage for virtual volumes using a zero logical disk
KR20100077156A (en) Thin provisioning migration and scrubbing
US8566554B2 (en) Storage apparatus to which thin provisioning is applied and including logical volumes divided into real or virtual areas

Legal Events

Date Code Title Description
STCB Information on status: application discontinuation

Free format text: ABANDONED -- FAILURE TO PAY ISSUE FEE