JP2005276135A - Disk management method and raid storage system - Google Patents

Disk management method and raid storage system Download PDF

Info

Publication number
JP2005276135A
JP2005276135A JP2004092654A JP2004092654A JP2005276135A JP 2005276135 A JP2005276135 A JP 2005276135A JP 2004092654 A JP2004092654 A JP 2004092654A JP 2004092654 A JP2004092654 A JP 2004092654A JP 2005276135 A JP2005276135 A JP 2005276135A
Authority
JP
Japan
Prior art keywords
temporary
area
storage system
disk
time
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.)
Pending
Application number
JP2004092654A
Other languages
Japanese (ja)
Inventor
Kiyohisa Miyamoto
Mamoru Sato
守 佐藤
清久 宮本
Original Assignee
Hitachi Computer Peripherals Co Ltd
日立コンピュータ機器株式会社
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 Hitachi Computer Peripherals Co Ltd, 日立コンピュータ機器株式会社 filed Critical Hitachi Computer Peripherals Co Ltd
Priority to JP2004092654A priority Critical patent/JP2005276135A/en
Publication of JP2005276135A publication Critical patent/JP2005276135A/en
Application status is Pending legal-status Critical

Links

Images

Abstract

A disk management method and a RAID storage system are provided.
In a RAID storage system 16 in which physical disks 11a to 11d are connected and a RAID group is configured by the same stripe columns of the plurality of physical disks 11a to 11d, the RAID groups of the physical disks 11a to 11d A plurality of physical disks 11a to 11d in which temporary block areas 15a to 15l are set by a plurality of virtual logical partitions in smaller units, and a temporary block mode management table 8 for managing the temporary block areas of the physical disks; A timer 7 for monitoring a monitoring time from receiving a command from a host to a response, and when the monitoring time from receiving the command to the response by the timer 7 exceeds a predetermined management time, What temporarily blocks.
[Selection] Figure 1

Description

  The present invention relates to a disk management method and a RAID storage system of a RAID storage system composed of a plurality of magnetic disk devices, and more particularly to a disk management method and a RAID storage system that can avoid a failure due to a violation of time rules.

  In recent years, for example, four magnetic disk devices are provided side by side, data is distributed and stored in three magnetic disk devices, and parity data is stored in one other magnetic disk device, so that a logical storage device is configured as a whole. RAID storage systems have been put into practical use. This RAID storage system is generally specified to be driven under the environment of a general-purpose OS. For example, when connected to a host computer driven by a special OS for forming a transaction system, Special specification requirements different from those of the general-purpose OS must be met. As this special specification requirement, for example, “input / output (I / O) time regulation” different from that of the general-purpose OS is known.

  This “input / output (I / O) time regulation” is, for example, an allowable time from when a host computer issues a single I / O request (read request or write request) to a storage device until a response is made. For example, in the host computer driven by the special OS, the “input / output (I / O) time regulation” is set to 10 seconds, and in the host computer driven by the general-purpose OS, the “input / output (I / O)” is set. Is set to 100 seconds.

  On the other hand, the RAID storage system can form, for example, more than 100 logical disks in one RAID group due to the recent increase in storage capacity of magnetic disk devices. By using a cache, it is necessary to configure so as not to violate the “input / output (I / O) time rule”.

  Also, two types of time monitoring are performed in a RAID storage system connected to a host computer driven by an OS that forms the transaction system. For example, as shown in FIG. The front-end processor 2 for processing the request and the back-end processor 3 for processing the physical disks 11a to 11d are arranged in two series, and a plurality of the I / O requests are stored, and high-speed processing is performed by optimizing the execution order. When the I / O queue 20 and the cache memory (not shown) are configured, the “input / output (I / O) time regulation” between the front-end processor 2 and the back-end processor 3 is within 9 seconds. "Input / output (I / I) between backend processor 3 and physical disks 11a-11d" ) Time the provisions of the "is set to within 4.5 seconds. The queue depth of each physical disk is about several hundreds, that is, several hundred I / O requests are held.

  When a read command I / O request is issued from the host computer 1, the RAID storage system 16 configured as described above temporarily stores the I / O request in the I / O queue 20 and stores this I / O request in the backend processor. 3 is sequentially taken out and issued to each of the physical disks 11a to 11c, data is read from each of the physical disks 11a to 11c and stored in the cache memory, and this data read notification is notified to the front-end processor 2, and this notification The front-end processor 2 receiving the operation operates to transfer the data from the cache memory to the host computer. Here, in the above-mentioned “input / output (I / O) time regulation”, the front end processor 2 receives the I / O request from the host computer 1 and then the processing completion notification from the back end processor 3 is returned. In a host computer driven by a special OS, it is set within 9 seconds.

  When the “input / output (I / O) time rule” is violated, the front-end processor 2 notifies the host computer 1 of the occurrence of a failure in the internal time monitoring by the front-end processor 2.

  On the other hand, the time from when the back-end processor 3 takes out an I / O request stored in the I / O queue 20 and issues a read command to the physical disks 11a to 11c until the completion notification is returned is also “input / output (I This time is 4.5 seconds. This time rule is a time rule by the back-end processor 3. When the time rule is exceeded, the back-end processor 3 reads the data read from the other physical disks that did not exceed the time rule and the physical disk 11d. It operates to perform a failure process called correction read that reproduces data using parity data.

  As described above, the RAID storage system according to the prior art includes a front-end processor 2 that performs processing for the host computer 1, a back-end processor 3 that performs processing for the physical disks 11 a to 11 d, and an I / O between the processors 2 and 3. By providing an I / O queue 20 for temporarily storing an / O instruction and performing an optimum processing procedure, a large number of I / O processes are realized.

The following patent documents are cited as documents describing the RAID storage system composed of the plurality of magnetic disk devices.
Japanese Patent Laid-Open No. 11-95933 JP-A-6-51915

  The RAID storage system configured as described above can respond to the host computer 1 in about 4.5 seconds by executing a failure process called the collection read when any failure occurs in the physical disk. However, if the failure of the physical disk is not irrecoverable and temporary, the processing by the back-end processor 3 is added every 4.5 seconds for I / O to the corresponding part from multiple hosts. The I / O behind the I / O queue 20 has a problem in that failure processing is reported to the host computer 1 because it violates the internal time regulation by the front-end processor 2 described above. For example, there is a problem in that the host computer 1 may cause the corresponding logical disk or the corresponding logical path to be offline, that is, perform a blocking process related to the logical disk and stop the entire system due to the failure.

  Further, in the prior art, since the host computer 1 repeatedly issues the same I / O request to the physical disk in which the failure has occurred within a predetermined number of failure reports, the I / O request is sent to the I / O queue 20. There was a problem that it was saved and the time monitoring was resumed. That is, in the prior art, there is a problem that the front-end processor 2 repeats timeout by monitoring the internal time. There is also a problem that the host computer 1 may stop when the failure of the RAID storage system is repeated.

  An object of the present invention is to eliminate the problems caused by the prior art, and to provide a disk management method and a RAID storage system that can avoid a failure due to a time regulation violation by the RAID storage system.

  The present invention relates to a disk management method for a RAID storage system in which a plurality of physical disks are connected and a RAID group is configured by the same stripe column of the plurality of physical disks, and is smaller in the RAID group based on the stripe column of the physical disk. When a temporary block area is set by a unit of a plurality of logical partitions in units, and a failure occurs in which the monitoring time from the reception of a command from the host to the response exceeds a predetermined management time, the temporary block where the failure has occurred The first feature is that the area is temporarily blocked. In this disk management method, when an I / O command for the temporarily blocked temporary blocked area is issued from the upper level, the same stripe as the temporarily blocked temporary blocked area is used. It is determined whether the temporary blockage area of another physical disk corresponding to the column is temporarily blocked. There is judged whether it is possible to correction read using the data of the temporary occlusion region, when it is judged possible by the determination, the second feature to perform the correction read.

  Further, the present invention provides a RAID storage system in which a plurality of physical disks are connected and a RAID group is configured by the same stripe column of the plurality of physical disks. Monitors a plurality of physical disks in which temporary block areas are set in units of a plurality of logical partitions, a temporary block mode management table for managing the temporary block areas of the physical disks, and a monitoring time from the reception of a command from the host to a response. And a time monitoring unit, wherein when the time monitoring unit determines that the monitoring time from the receipt of the command to the response exceeds a predetermined management time, the temporary blocking region where the failure has occurred is temporarily blocked. In this RAID storage system, the I / O command for the temporarily blocked temporary blocked area is Is determined by referring to the temporary block mode management table to determine whether or not the temporary block area of another physical disk corresponding to the same stripe column as the temporarily blocked temporary block area is temporarily blocked. According to the fourth feature, it is determined whether or not correction read is possible using data of a temporary block region that has not been temporarily blocked, and when it is determined that determination is possible, correction read is executed.

  The disk management method and RAID storage system according to the present invention provide a plurality of logical partitions in which a physical disk block unit is a smaller virtual / temporary RAID group in a real RAID group based on a stripe array of physical disks. By performing temporary blockage in units of a plurality of logical partitions as the temporary RAID group, it is possible to avoid a failure due to a time regulation violation by the RAID storage system. Further, according to the present invention, it is determined by referring to the temporary block mode management table that the temporary block can be reproduced by collection read, and the data is reproduced when it can be reproduced.

  Hereinafter, a disk management method and a RAID storage system according to the present invention will be described in detail with reference to the drawings. First, the principle of the present invention will be described.

  First, in the RAID storage system according to the prior art, when a failure occurs in one physical disk as described above, the entire logical disk may be moved offline. The failure that occurs is generally a physical failure of a part of the disk recording surface, and attention is paid to the fact that data can be normally recorded and reproduced on other recording surfaces.

  In order to solve the inconvenience, the inventors set a block unit as a plurality of logical partition units, which are virtual / temporary RAID groups of smaller units in the existing RAID group based on the stripe column of the physical disk, The present invention invented the present invention capable of preventing the entire system from being stopped by performing temporary blockage in units of a plurality of logical partitions as the temporary RAID group.

Hereinafter, a RAID storage system to which a disk management method according to the present invention is applied will be described in detail with reference to FIG.
The RAID storage system 16 shown in FIG. 1 includes a front-end time management table 8 and a plurality of physical disks 11a including a front-end processor 2 in charge of processing for the host computer 1 and a temporary block area 15a described later. -11d, a back-end time management table 9 and an error identification determination unit 17, and a back-end processor 3 responsible for executing an I / O command for the physical disks 11a-11d, an internal I / O An internal I / O queue counter 5 that counts the O queue 4, a timer 7 that generates the current time, and a cache memory 12 that stores a corresponding record of data read from the physical disks 11a to 11d, an I / O instruction, and the like. And corresponding records such as data stored in the cache memory 12 A cache pointer 13 for storing the pointer, a cache memory 12 for temporarily storing data read from the plurality of physical disks 11a to 11c, and a temporary blockage area in the physical disks 11a to 11d. A temporary block mode management table 10 for managing whether or not the medium is in use.

  The plurality of physical disks 11a to 11d temporarily block a plurality of logical partition units, which are virtual and temporary RAID groups of smaller units, within an existing RAID group (logical disk unit) based on a stripe array of physical disks. Areas 15a to 15l are set.

  In the storage system configured as described above, when the host computer 1 first issues a read request to an arbitrary logical disk to the RAID storage system 16, the front-end processor 2 detects the request, and the cache pointer corresponding to the corresponding record. 13, whether the data requested by the host exists in the cache memory 12, and if the corresponding record has already been stored in the cache memory 12, the front end processor 2 requests the host computer 1 from the cache memory 12. The processed data is transferred to complete the process.

  If the record is not stored in the cache memory 12 as a result of checking the cache pointer 13, the front-end processor 2 determines which physical disk 11 the I / O request is for, and The I / O request parameters (physical location / size / read / write, etc.) are determined and pushed to the internal I / O queue 4 of the target physical disk 11a, for example, and the internal I / O queue counter 5 The counter value is added, the current time is acquired from the timer 7, and a timeout time (for example, current time + 9 seconds) is set in the front-end time management table 8.

  Next, in the present system, the back-end processor 3 pops the I / O request from the internal I / O queue 4 and subtracts the counter value of the internal I / O queue counter 5 to set the I / O request parameter ( (Physical location / size / read / write, etc.) is recognized, and if the I / O request is read, the temporary physical block area 15a to 15l in which the designated physical location exists is calculated.

Next, in the present system, the back-end processor 3 refers to the temporary block mode management table 10 and recognizes whether or not the temporary block area 15 where the designated physical location exists is temporarily blocked. In the present embodiment, the following description will be given on the assumption that the temporary blocking areas 15a and 15e are registered in the temporary blocking mode management table 10 as being temporarily blocked at this time.
<When the target record exists in 15a>

  When the backend processor 3 recognizes by referring to the temporary block mode management table 10 that the temporary block area 15a in which the designated physical location exists is being temporarily blocked, the back-end processor 3 determines that the corresponding temporary block area 15a is Since the temporary block is in progress, a read command is issued to the other physical disks 11b / 11c / 11d for collection read, the current time is acquired from the timer 7, and the time-out time (current time + 4. 5 seconds).

  The back-end processor 3 reads the data from the temporary blockage area 15d / temporary blockage area 15g / temporary blockage area 15j, which are the same stripe row of the temporary blockage area 15a, stores the data in the cache memory 12, and then the front-end processor 2 Notify that the data is ready. Receiving this notification, the front-end processor 2 transfers the requested data from the cache memory 12 to the host computer 1 and completes the processing.

That is, when the data requested by the host exists in the temporary block area 15a of the physical disk 11a, the RAID storage subsystem 16 according to this embodiment reads the corresponding data of the other physical disks 11b to 11d and reproduces the data. , Act to respond to the host.
<When the target record exists in 15c>

  [1] When the temporary blockage area 15c in which the designated physical location exists is not temporarily blocked with reference to the temporary blockage mode management table 10, the back-end processor 3 recognizes the corresponding temporary blockage area 15c. Is not temporarily blocked, a read command is issued to the physical disk 11a, the current time is obtained from the timer 7, and a timeout time (current time + 4.5 seconds) is set in the back-end time management table 9.

  Next, when the data is read from the physical location of the designated physical disk 11a, the back-end processor 3 transfers the data to the cache memory 12, and notifies the front-end processor 2 that the data is ready. Receiving this, the front end processor 2 transfers the requested data from the cache memory 12 to the host computer 1 and completes the processing.

  [2] If a failure occurs at the physical location designated here, and a retry is performed, and the back-end processor 3 cannot read data normally by the time-out time set above, the back-end processor 3 refers to the temporary blockage mode management table 10 in order to determine whether or not the temporary blockage region 15c can be set to the temporary blockage mode. At this time, the temporary blockage region 15f / temporary block having the same stripe row as the temporary blockage region 15c is referred to. Since the temporary blockage mode is not set in the blockage region 15i / temporary blockage region 15l, it is determined that correction read is possible with this stripe row even if the temporary blockage mode is set for the temporary blockage region 15c. .

  Next, the back-end processor 3 acquires the current time from the timer 7 in order to set the temporary blockage area 15 c where the failure has occurred to the temporary blockage mode, and temporarily stores the temporary blockage area 15 c in the area corresponding to the temporary blockage area 15 c in the temporary blockage mode management table 10. When the block mode is being set, the set time is written, and in the same manner as in the above <when the target record exists in 15a>, a read command is issued to the physical disks 11b / 11c / 11d in order to perform a collection read. To work.

  [3] If a failure occurs in the physical location designated here, the retry is performed, and if the back-end processor 3 cannot read data normally by the time-out time already set, the back-end processor 3 Refers to the temporary block mode management table 10 to determine whether the temporary block region 15c can be set to the temporary block mode, and refers to the temporary block region 15f / temporary block region 15i / temporary in the same stripe row as the temporary block region 15c. If the temporary blockage mode is set in any of the blocked areas 15l, the temporary blockage mode is canceled and then the temporary blockage area 15c is set in the temporary blockage mode. The management table 10 is updated.

  In the above-described embodiment, the case where the temporary block area 15c is set to the temporary block mode has been described. However, when the physical disk 11a is actually blocked, another physical disk 11b / physical disk 11c / physical disk 11d is used. In order to release the temporary blockage area 15 of all the areas, the corresponding temporary blockage mode management table 10 is updated.

  In the above embodiment, a description is given on the assumption that a RAID group is configured with four physical disks and one physical disk is a parity disk, and the temporary block area 15c is set to the temporary block mode. As described above, for example, when a RAID group is configured with eight physical disks and two of them are used as ECC disks, it is possible to recover up to two physical disk failures in one RAID group by collection. Conceivable.

  In the case of such eight RAID storage systems, in order to determine whether or not the temporary blockage area 15 can be set to the temporary blockage mode, when the temporary blockage mode management table 10 is referred to, the same stripe column as the corresponding temporary blockage area If the temporary block mode has already been set in two areas, cancel all the temporary block modes in the same stripe row, or temporarily block the old ones of the two set times Judgment / selection can be made from the mode management table 10 to release the temporary blockage mode.

  In addition to the individual / local failures on the disk medium surface, there are common failures in the mechanical system as physical disk failures. In this case, the error code returned by the physical disk is determined by the error type determination unit 17, and in the case of the former failure, the temporary blockage area 15 is managed as described above, or the temporary block mode is canceled. It can also be performed at the normal end of the write operation for the closed area 15. Further, according to the present invention, when it is determined that the failure is related to the mechanical system, it is possible to set all the temporary block areas 15 of the physical disk in the temporary block mode management table 10 to the temporary block mode, and further configure the transaction system. In the host computer 1, the temporary block mode has the purpose of preventing the logical disk from being offline due to the occurrence of a continuous timeout. Therefore, the internal I / O queue counter 5 of the corresponding physical disk 11 is referred to and the standby I / O Even if it does not exist, the temporary closing mode can be canceled.

  As described above, according to the present invention, the physical disk block unit is a plurality of logical partition units which are smaller virtual / temporary RAID groups in the actual RAID group based on the physical disk stripe column. When temporary blockage is performed in units of a plurality of logical partitions in the RAID group, it is determined with reference to the temporary blockage mode management table 10 that the temporary blockage is reproducible by collection read. By performing data reproduction, it is possible to avoid a failure due to a time regulation violation by the RAID storage system.

The figure for demonstrating the RAID storage system to which the disk management method by this invention is applied. The figure for demonstrating the RAID storage system by a prior art.

Explanation of symbols

1: host computer, 2: front end processor, 3: back end processor, 4: queue, 5: queue counter, 7: timer, 8: front end time management table, 9: back end time management table, 10: temporary block Mode management table, 11a to 11d: physical disk, 12: cache memory, 13: cache pointer, 15a to 15l: temporary block area, 16: RAID storage system, 17: error type determination unit.

Claims (4)

  1. A disk management method for a RAID storage system in which a plurality of physical disks are connected and a RAID group is configured by the same stripe column of the plurality of physical disks, the virtual disk having a smaller unit in the RAID group by the stripe column of the physical disk. A temporary block area is set in units of multiple logical partitions, and when a failure occurs in which the monitoring time from the receipt of a command from the host to the response exceeds the predetermined management time, the temporary block area where the failure has occurred is temporarily blocked And a disk management method.
  2. 2. The disk management method according to claim 1, wherein when an I / O command for the temporarily blocked temporarily blocked area is issued from a higher level, another physical disk corresponding to the same stripe column as the temporarily blocked temporarily blocked area. When the temporary blockage area is temporarily blocked, it is determined whether or not correction read is possible using the data of the temporary blockage area that is not temporarily blocked by the determination. A disk management method characterized by executing collection read.
  3. In a RAID storage system in which a plurality of physical disks are connected and a RAID group is configured by the same stripe column of the plurality of physical disks, a plurality of virtual logical partitions in smaller units in the RAID group by the stripe column of the physical disks A plurality of physical disks in which a temporary block area is set in units, a temporary block mode management table for managing the temporary block area of the physical disk, and a time monitoring unit for monitoring a monitoring time from command reception to response from a host. And a RAID storage system characterized in that, when the time monitoring means determines that the monitoring time from command reception to response exceeds a predetermined management time, the temporary blocking area where the failure has occurred is temporarily blocked.
  4. 3. The RAID storage system according to claim 2, wherein when an I / O command for the temporarily blocked temporary blocked area is issued from an upper level, another physical disk corresponding to the same stripe column as the temporarily blocked temporary blocked area To determine whether or not the collection read is possible using the data of the temporary blockage area that is not temporarily blocked by the determination. A RAID storage system, wherein correction reading is executed when it is determined that the determination is possible.
JP2004092654A 2004-03-26 2004-03-26 Disk management method and raid storage system Pending JP2005276135A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2004092654A JP2005276135A (en) 2004-03-26 2004-03-26 Disk management method and raid storage system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2004092654A JP2005276135A (en) 2004-03-26 2004-03-26 Disk management method and raid storage system

Publications (1)

Publication Number Publication Date
JP2005276135A true JP2005276135A (en) 2005-10-06

Family

ID=35175709

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2004092654A Pending JP2005276135A (en) 2004-03-26 2004-03-26 Disk management method and raid storage system

Country Status (1)

Country Link
JP (1) JP2005276135A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100827677B1 (en) 2006-06-20 2008-05-07 한국과학기술원 A method for improving I/O performance of RAID system using a matrix stripe cache

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS58175064A (en) * 1982-04-07 1983-10-14 Fujitsu Ltd Processing system of multiplex volume
JPS60258659A (en) * 1984-06-05 1985-12-20 Nec Corp Control system for partial file closure
JPH09212311A (en) * 1996-01-31 1997-08-15 Brother Ind Ltd Disk array device
JPH11259242A (en) * 1998-03-10 1999-09-24 Nec Corp Controller for disk array and method therefor
JPH11338648A (en) * 1998-02-26 1999-12-10 Nec Corp Disk array device, its error control method, and recording medium where control program thereof is recorded
JP2000132413A (en) * 1998-10-22 2000-05-12 Nec Corp Error retry method, error retry system and its recording medium
JP2000242438A (en) * 1999-02-19 2000-09-08 Matsushita Electric Ind Co Ltd Disk array system capable of executing reassign processing
JP2002373059A (en) * 2001-06-14 2002-12-26 Toshiba Corp Method for recovering error of disk array, and controller and device for disk array

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS58175064A (en) * 1982-04-07 1983-10-14 Fujitsu Ltd Processing system of multiplex volume
JPS60258659A (en) * 1984-06-05 1985-12-20 Nec Corp Control system for partial file closure
JPH09212311A (en) * 1996-01-31 1997-08-15 Brother Ind Ltd Disk array device
JPH11338648A (en) * 1998-02-26 1999-12-10 Nec Corp Disk array device, its error control method, and recording medium where control program thereof is recorded
JPH11259242A (en) * 1998-03-10 1999-09-24 Nec Corp Controller for disk array and method therefor
JP2000132413A (en) * 1998-10-22 2000-05-12 Nec Corp Error retry method, error retry system and its recording medium
JP2000242438A (en) * 1999-02-19 2000-09-08 Matsushita Electric Ind Co Ltd Disk array system capable of executing reassign processing
JP2002373059A (en) * 2001-06-14 2002-12-26 Toshiba Corp Method for recovering error of disk array, and controller and device for disk array

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100827677B1 (en) 2006-06-20 2008-05-07 한국과학기술원 A method for improving I/O performance of RAID system using a matrix stripe cache

Similar Documents

Publication Publication Date Title
US8190846B2 (en) Data management method in storage pool and virtual volume in DKC
US8898385B2 (en) Methods and structure for load balancing of background tasks between storage controllers in a clustered storage environment
US8239584B1 (en) Techniques for automated storage management
US6978325B2 (en) Transferring data in virtual tape server, involves determining availability of small chain of data, if large chain is not available while transferring data to physical volumes in peak mode
US8423739B2 (en) Apparatus, system, and method for relocating logical array hot spots
JP4220724B2 (en) Storage device
US8775731B2 (en) Write spike performance enhancement in hybrid storage systems
US7451346B2 (en) Storage control device and data recovery method for storage control device
US8635427B2 (en) Data storage control on storage devices
US8959305B1 (en) Space reclamation with virtually provisioned devices
JP2005235058A (en) Snapshot acquisition method, snapshot acquisition device, and computer program provided with snapshot acquisition function
US8495288B2 (en) Storage controller and duplicated data detection method using storage controller
US6341342B1 (en) Method and apparatus for zeroing a transfer buffer memory as a background task
US7069465B2 (en) Method and apparatus for reliable failover involving incomplete raid disk writes in a clustering system
JP4922496B2 (en) Method for giving priority to I / O requests
US20020099908A1 (en) Apparatus and method for reallocating logical to physical disk devices using a storage controller, with access frequency and sequential access ratio calculations and display
JP2004206623A (en) Storage system and its dynamic load management method
US8751699B1 (en) Systems and methods for indication of activity status of a storage device
JP3078972B2 (en) Disk array device
US7743206B2 (en) Dynamic loading of virtual volume data in a virtual tape server
EP0718766A2 (en) Method of operating a disk drive array
US7447832B2 (en) Automated on-line capacity expansion method for storage device
CN101566931B (en) Virtual disk drive system and method
JP2004005634A (en) Method for uniformly distributing load over plural storage mediums in storage device array
US9690487B2 (en) Storage apparatus and method for controlling storage apparatus

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20070323

RD02 Notification of acceptance of power of attorney

Free format text: JAPANESE INTERMEDIATE CODE: A7422

Effective date: 20091211

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20100203

A521 Written amendment

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20100330

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20110121

A02 Decision of refusal

Free format text: JAPANESE INTERMEDIATE CODE: A02

Effective date: 20110706