CN1722074A - Fault-tolerant type disk array controller - Google Patents
Fault-tolerant type disk array controller Download PDFInfo
- Publication number
- CN1722074A CN1722074A CNA2004100280988A CN200410028098A CN1722074A CN 1722074 A CN1722074 A CN 1722074A CN A2004100280988 A CNA2004100280988 A CN A2004100280988A CN 200410028098 A CN200410028098 A CN 200410028098A CN 1722074 A CN1722074 A CN 1722074A
- Authority
- CN
- China
- Prior art keywords
- disk array
- controller
- array controller
- order
- data
- 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
Links
Images
Abstract
The invention relates to an error-permitting type disk array controller, which uses multicast transmitting technology of express advanced switch technology to order or do multicast the data, so that the received and executed order or data of main disk array controller can do synchronic imaging to assistant disk array controller; assistant disk array controller dose on and on error measuring to main disk array controller; when it measures the error of main disk array controller, the assistant disk array controller immediately dose the work of main disk array controller and keeps the writing connection of data.
Description
Technical field
The present invention is a kind of fault tolerant formula disk array controller, particularly be applied on the disk array framework, a kind of by PCI EXPRESS Advanced Switch in multicast (multicast) mode with synchronous renewal mirror image data to keep the conforming fault tolerant type disk array controller of high-speed cache.
Background technology
So-called disk array briefly, be that several hard disks are chained together into a virtual big hard disk with specific technology, data can be cut into many blocks, when reading or write data, many hard disks are carried out access action simultaneously in parallel mode, when a hard disk number the more the time, more can increase access speed.Disk array uses just that several are cheap, the less hard disk (HDD) of capacity is combined into a single high power capacity storage facilities in fact, and can accelerate access speed and tolerance fault (FAULT TOLERANCE).
And so-called disk array controller (RAID Controller) promptly be to control disk array write and read action, because disk array controller is the action that many hard disks are read and write, so traditional fault tolerant type disk array controller (as shown in Figure 1), usually by two disk array controller 50,60 form, wherein a disk array controller is a master disk array control unit 50, another disk array controller then is secondary disk array controller 60, the order or the data that receive main frame 41 by master disk array control unit 50 read and write activity disk array 42, secondary disk array controller 60 is then taken over when master disk array control unit 50 faults and is continued execution work, thereby reaches the purpose of REDUNDANT.
The purpose of REDUNDANT is to use secondary disk array controller 60 to deposit unnecessary (REDUNDANCY) data, being used for when master disk array control unit 50 damages can data reconstruction, the meaning of disk array controller tolerance fault is when one of them disk array controller generation problem, another disk array controller still can provide service, and can rebuild the REDUNDANCY data to deal with the damage of next time.
And the traditional magnetic disk array control unit reaches the method for fault tolerance by REDUNDANT, the built-in controller 52 that is master disk array control unit 50 is handled and is temporarily stored in its high-speed cache 53 by order or data that host interface 51 receives main frame 41, secondary disk array controller 60 then is connected to the host interface 51 or two disk array controller 50 of master disk array control unit 50 by its host interface 61, the bus (not shown) that 60 inside are connected carries out error detection and data image that the high-speed cache 53 of master disk array control unit 50 is the stored high-speed cache 62 to itself to master disk array control unit 50, in case the master disk array control unit breaks down, just can take over work immediately.
Yet, no matter be two host interface 51,61 or internal bus between signal contact, because the restriction of known technology all is load modes of taking unicast (master-to-slave), this kind load mode makes the built-in controller 52 of master disk array control unit 50 at storage data during to high speed buffer memory 53, can't be stored into the high-speed cache 62 of secondary disk array controller 60 synchronously, must carry out access in two steps suddenly, two high-speed caches 53 just, 62 Data Update is to have a mistiming, the work disposal situation of master disk array control unit 50 also can't be upgraded synchronously with secondary disk array controller 60, when in case master disk array control unit 50 breaks down by secondary disk array controller 60 catchers, being easy to take place script master disk array control unit 50 is handling and uncompleted work, secondary disk array controller 60 is promptly judged and has been finished dealing with and continue to carry out next work, thereby cause the mistake of data, distortion.
Summary of the invention
Fundamental purpose of the present invention is to provide a kind of fault tolerant type disk array controller, and the technology by multicast can be upgraded the order of master disk array control unit and secondary disk array controller or data synchronously.
For reaching above-mentioned purpose, fault tolerant type disk array controller of the present invention, be to be connected with an at least one main frame and a disk array respectively, in order to order or the data that receive main frame disk array is read and write activity, and the fault tolerant type disk array controller is made up of two disk array controller that are connected, and selected one disk array controller is as the master disk array control unit, another disk array controller is then as secondary disk array controller, the order or the data that receive main frame by the master disk array control unit when normal start read and write activity disk array, and will order with the data sync mirror image to secondary disk array controller, secondary disk array controller then constantly carries out error detection to the master disk array control unit, in case the master disk array control unit breaks down or when wrong, secondary disk array controller i.e. the performed work of catcher master disk array control unit automatically.
Disk array controller then includes a host interface, an advanced exchange transport module, a built-in controller, a high-speed cache and a disk array driving interface and forms, and each element is to utilize PCI EXPRESSAdvanced Switch bus to be connected, and makes each interelement can carry out the multicast of order or data.
Then order or the data exported of main frame are that the built-in controller that transfers to the master disk array control unit is handled, and synchronous mirror to the built-in controller of secondary disk array controller carries out formation, when the built-in controller for the treatment of the master disk array control unit is handled the order or the data of main frame and is exported a response signal to main frame, at synchronous mirror to the built-in controller of secondary disk array controller order or data deletion with formation; In like manner, the built-in controller of master disk array control unit exports the order of disk array driving interface or data to and also exchanges transport module mirror image to the disk array driving interface of secondary disk array controller through the advanced person and carry out formation, when a response signal is finished and is exported in the disk array driving interface work for the treatment of the master disk array control unit to built-in controller, at synchronous mirror to the disk array driving interface of secondary disk array controller order or data deletion with formation.
Simultaneously, the advanced person of secondary disk array controller exchanges transport module, and to be constantly output one error detection signal exchange transport module to the advanced person of master disk array control unit, when detecting that the master disk array control unit breaks down or when wrong, because the built-in controller in the secondary disk array controller and the order and the data of disk array driving interface institute formation are to upgrade synchronously with the master disk array control unit, so the continuity that secondary disk array controller can be taken over the performed work of master disk array control unit immediately and keep data to write.
Description of drawings
Fig. 1 is the circuit box synoptic diagram of the disclosed disk array controller of prior art;
Fig. 2 is the circuit box synoptic diagram of the disclosed disk array controller of the present invention.
Embodiment
As shown in Figure 2, be circuit box synoptic diagram of the present invention.This fault tolerant type disk array controller 10 is to be connected with an at least one main frame 11 and a redundant array of inexpensive disks 12 respectively, in order to order or the data that receive main frame 11 redundant array of inexpensive disks 12 is read and write activity.
This redundant array of inexpensive disks 12 is to include several hard disks (HDD) to form, and is stored in respectively in each hard disk in order to data are cut into many blocks, in parallel mode many hard disks is carried out access action simultaneously when reading or write data.
And this fault tolerant type disk array controller 10, then include two disk array controller that are connected (RAIDController) 20,30, and select a disk array controller as master disk array control unit 20, another disk array controller is then as secondary disk array controller 30.When normal start, the order or the data that receive main frame 11 by master disk array control unit 20 read and write activity redundant array of inexpensive disks 12, and will order with data sync mirror image (mirror) to secondary disk array controller 30, secondary disk array controller 30 then carries out error detection (Heartbeat Detection) to master disk array control unit 20, in case master disk array control unit 20 breaks down or when wrong, secondary disk array controller 30 i.e. the performed work of catcher master disk array control unit 20 automatically, thereby reach the purpose of restoration after a failure (failover).
This master disk array control unit 20 then includes a host interface 21, one advanced exchange transport module 22, a built-in controller 23, a high-speed cache 24 and a disk array driving interface 25 and forms.
This host interface 21 is to be connected to main frame 11, exports after receiving order that main frame 11 transmitted or data.
This advanced exchange transport module (advanced switch) 22, be to be connected to host interface 21 with PCI EXPRESS AS bus, after receiving order that host interface 21 transmitted or data, export, and the order that will import and data image (mirror) secondary disk array controller 30 extremely.
This built-in controller 23 is to be connected to advanced exchange transport module 22 with PCI EXPRESS AS bus, in order to handle and to carry out the order or the data of input.
This high-speed cache (cache memory) 24 is to be connected to built-in controller 23, in order to storing the order or the data of frequent access, and the efficient that reads and write with raising.
This disk array driving interface 25 is to be connected to advanced exchange transport module 22 with PCI EXPRESS AS bus, in order to receive order or the data that advanced exchange transport module 22 is transmitted, exports after being converted to the specification of disk array.
And secondary disk array controller 30 also includes a host interface 31, an advanced person exchanges transport module 32, a built-in controller 33, a high-speed cache 34 and a disk array driving interface 35 and forms.
This host interface 31 is to be connected to main frame 11, exports after receiving order that main frame 11 transmitted or data.
This advanced exchange transport module 32, be to exchange transport module 22 with the advanced person that PCI EXPRESS AS bus is connected to host interface 31 and master disk array control unit 20 respectively, after receiving order that host interface 31 transmitted or data, export, and export an error detection signal (heartbeat) and exchange transport module 22 to the advanced person of master disk array control unit 20, when detecting that master disk array control unit 20 breaks down or when wrong, secondary disk array controller 30 i.e. the performed work of catcher master disk array control unit 20 automatically.
This built-in controller 33 is to be connected to advanced exchange transport module 32 with PCI EXPRESS AS bus, in order to handle and to carry out the order or the data of input.
This high-speed cache (cache memory) 34 is to be connected to built-in controller 33, in order to storing the order or the data of frequent access, and the efficient that reads and write with raising.
This disk array driving interface 35 is to be connected to advanced exchange transport module 32 with PCI EXPRESS AS bus, in order to receive order or the data that advanced exchange transport module 32 is transmitted, exports after being converted to the specification of disk array.
The above-mentioned connected mode of utilizing PCI EXPRESS AS bus is to make each interelement can carry out the multicast (MLITICAST) of order or data, and other bus that can carry out point-to-point (POINT-TO-POINT) transmission also is practical range of the present invention.
In addition, the disk array driving interface 25 of master disk array control unit 20 and the disk array driving interface 35 of secondary disk array controller 30 all are connected with a multiplexer 13,13 of this multiplexers are connected to redundant array of inexpensive disks 12, receive the path of order and data with decision redundant array of inexpensive disks 12 by this multiplexer 13.
Therefore, main frame 11 output commands or data are during to fault tolerant type disk array controller 10, if master disk array control unit 20 is working properly, promptly after the host interface 21 reception orders or data by master disk array control unit 20, transfer to advanced exchange transport module 22 again, and advanced exchange transport module 22 is after receiving the order or data that host interface 21 transmitted, being about to order and data transmission to built-in controller 23 handles, and will order synchronously and data image (mirror) exchanges transport module 32 to the advanced person of secondary disk array controller 30, and the advanced person of secondary disk array controller 30 exchanges transport module 32 after order that receives mirror image or data, and soon order or data transmission to the built-in controller 33 of secondary disk array controller 30 carries out formation (QUEUE).
After the built-in controller 23 for the treatment of master disk array control unit 20 is handled the order or data of main frame 11, promptly export a response signal and exchange transport module 22 and host interface 21 to main frame 11 through the advanced person, exchange transport module 22 by the advanced person simultaneously the advanced person that the response signal mirror image transfers to secondary disk array controller 30 is exchanged transport module 32, and then transfer to the built-in controller 33 of secondary disk array controller 30, then built-in controller 33 can be with the order or the data deletion of formation.
In addition, the built-in controller 23 of master disk array control unit 20 exports redundant array of inexpensive disks 12 to order or data, be to exchange transport module 22 through the advanced person to transfer to disk array driving interface 25 and handle, and will order synchronously and data exchange transport module 22 mirror images (mirror) through the advanced person and exchange transport module 32 to the advanced person of secondary disk array controller 30, and the advanced person of secondary disk array controller 30 exchanges transport module 32 after order that receives mirror image or data, is about to order or data transmission to the disk array driving interface 35 of secondary disk array controller 30 and carries out formation (QUEUE).
After the specification that the disk array driving interface 25 for the treatment of master disk array control unit 20 will be ordered or data are converted to disk array exports redundant array of inexpensive disks 12 to, promptly export a response signal and exchange transport module 22 to built-in controller 23 through the advanced person, exchange transport module 22 by the advanced person simultaneously the advanced person that the response signal mirror image transfers to secondary disk array controller 30 is exchanged transport module 32, and then transfer to the disk array driving interface 35 of secondary disk array controller 30, then disk array driving interface 35 can be with the order or the data deletion of formation.
At the same time, the advanced person of secondary disk array controller 30 exchanges transport module 32, and to be constantly output one error detection signals (heartbeat) exchange transport module 22 to the advanced person of master disk array control unit 20, when detecting that master disk array control unit 20 breaks down or when wrong, because the built-in controller 33 in the secondary disk array controller 30 and the order and the data of 35 formations of disk array driving interface are to upgrade synchronously with master disk array control unit 20, so secondary disk array controller 30 can be taken over the continuity that the performed work of master disk array control unit 20 and maintenance data write immediately.
Claims (7)
1. fault tolerant type disk array controller, be to be connected with an at least one main frame and a disk array respectively, in order to order or the data that receive main frame disk array is read and write activity, and this fault tolerant type disk array controller is made up of two disk array controller that are connected (RAID Controller), and selected one disk array controller is as the master disk array control unit, and another disk array controller is as secondary disk array controller, the order or the data that receive this main frame by this master disk array control unit read and write activity this disk array, should then when this master disk array control unit fault, take over the continuation execution work by the pair disk array controller, it is characterized in that this disk array controller includes:
One host interface is to be connected to this main frame by a network, exports after receiving order that this main frame transmits or data;
One advanced exchange transport module is to be connected to this host interface, after receiving order that this host interface transmits or data, export, and the order that will import and data with multicast (multicast) to this pair disk array controller;
One built-in controller is to be connected to this advanced person to exchange transport module, in order to handle and to carry out the order or the data of input;
One high-speed cache is to be connected to this built-in controller, in order to store the order or the data of frequent access; And
One disk array driving interface is to be connected to this advanced person respectively to exchange transport module and this disk array, exchanges order or the data that transport module transmits in order to receive this advanced person, exports this disk array to after being converted to the specification of disk array.
2. fault tolerant type disk array controller as claimed in claim 1 is characterized in that, it is to be connected to this host interface with PCI EXPRESS AS bus that this advanced person exchanges transport module.
3. fault tolerant type disk array controller as claimed in claim 1 is characterized in that, this built-in controller is to be connected to this advanced person with PCI EXPRESS AS bus to exchange transport module.
4. fault tolerant type disk array controller as claimed in claim 1 is characterized in that, this disk array driving interface is to be connected to this advanced person with PCI EXPRESS AS bus to exchange transport module.
5. fault tolerant type disk array controller as claimed in claim 1 is characterized in that, it is to exchange transport module with the advanced person that PCI EXPRESS AS bus is connected to this master disk array control unit that the advanced person of this pair disk array controller exchanges transport module.
6. fault tolerant type disk array controller as claimed in claim 1 is characterized in that, is to be connected with a multiplexer between this two disk arrays driving interface and this disk array, and receives the path of order and data by this multiplexer to determine this disk array.
7. fault tolerant type disk array controller as claimed in claim 1, it is characterized in that, the advanced person of this pair disk array controller exchanges transport module, and to be constantly output one error detection signal (heartbeat) exchange transport module to the advanced person of this master disk array control unit, detect this master disk array control unit and whether break down or mistake, thereby and take over the performed work of this master disk array control unit immediately.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2004100280988A CN1722074A (en) | 2004-07-16 | 2004-07-16 | Fault-tolerant type disk array controller |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2004100280988A CN1722074A (en) | 2004-07-16 | 2004-07-16 | Fault-tolerant type disk array controller |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1722074A true CN1722074A (en) | 2006-01-18 |
Family
ID=35912409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2004100280988A Pending CN1722074A (en) | 2004-07-16 | 2004-07-16 | Fault-tolerant type disk array controller |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1722074A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100410866C (en) * | 2006-09-06 | 2008-08-13 | 威盛电子股份有限公司 | Computer system with magnetic disk array control function and magnetic disk array control method |
| CN100416484C (en) * | 2006-09-06 | 2008-09-03 | 威盛电子股份有限公司 | Core logic unit with magnetic disk array control function and magnetic disk array control method |
| CN102215245A (en) * | 2010-04-06 | 2011-10-12 | 英业达股份有限公司 | Configuration information synchronization method of dual-controller of storage area network |
| US8041890B2 (en) | 2006-04-19 | 2011-10-18 | Hangzhou H3C Technologies Co., Ltd. | Method for accessing target disk, system for expanding disk capacity and disk array |
| CN101621465B (en) * | 2009-07-28 | 2011-12-07 | 成都市华为赛门铁克科技有限公司 | Method, device and system for sending data packet and exchanging equipment |
| CN101640695B (en) * | 2008-07-29 | 2012-08-08 | 英业达股份有限公司 | Method for providing customization replies by using disk array |
| CN101776983B (en) * | 2009-01-13 | 2015-09-16 | 中兴通讯股份有限公司 | The synchronous method of information of double controllers in disk array and disc array system |
| CN105007307A (en) * | 2015-06-18 | 2015-10-28 | 浪潮(北京)电子信息产业有限公司 | Storage control method and system |
-
2004
- 2004-07-16 CN CNA2004100280988A patent/CN1722074A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8041890B2 (en) | 2006-04-19 | 2011-10-18 | Hangzhou H3C Technologies Co., Ltd. | Method for accessing target disk, system for expanding disk capacity and disk array |
| CN100410866C (en) * | 2006-09-06 | 2008-08-13 | 威盛电子股份有限公司 | Computer system with magnetic disk array control function and magnetic disk array control method |
| CN100416484C (en) * | 2006-09-06 | 2008-09-03 | 威盛电子股份有限公司 | Core logic unit with magnetic disk array control function and magnetic disk array control method |
| CN101640695B (en) * | 2008-07-29 | 2012-08-08 | 英业达股份有限公司 | Method for providing customization replies by using disk array |
| CN101776983B (en) * | 2009-01-13 | 2015-09-16 | 中兴通讯股份有限公司 | The synchronous method of information of double controllers in disk array and disc array system |
| CN101621465B (en) * | 2009-07-28 | 2011-12-07 | 成都市华为赛门铁克科技有限公司 | Method, device and system for sending data packet and exchanging equipment |
| CN102215245A (en) * | 2010-04-06 | 2011-10-12 | 英业达股份有限公司 | Configuration information synchronization method of dual-controller of storage area network |
| CN105007307A (en) * | 2015-06-18 | 2015-10-28 | 浪潮(北京)电子信息产业有限公司 | Storage control method and system |
| CN105007307B (en) * | 2015-06-18 | 2019-02-22 | 浪潮(北京)电子信息产业有限公司 | A kind of storage controlling method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7529965B2 (en) | Program, storage control method, and storage system | |
| US7444541B2 (en) | Failover and failback of write cache data in dual active controllers | |
| US7975168B2 (en) | Storage system executing parallel correction write | |
| US7124244B2 (en) | Storage system and a method of speeding up writing data into the storage system | |
| US5566316A (en) | Method and apparatus for hierarchical management of data storage elements in an array storage device | |
| JP2548480B2 (en) | Disk device diagnostic method for array disk device | |
| US20140304461A1 (en) | Storage subsystem | |
| WO2009107213A1 (en) | Storage device, storage controller, data transfer integrated circuit, and method of controlling storage | |
| JPWO2006123416A1 (en) | Disk failure recovery method and disk array device | |
| JP2002108572A (en) | Backup system and duplicating device | |
| CN1598755A (en) | Jbod subsystem and external emulation controller thereof | |
| JP4343923B2 (en) | DMA circuit and data transfer method | |
| JPH06119724A (en) | Array type recording device | |
| CN1773443A (en) | Disk array apparatus, method of data recovery, and computer product | |
| CN1825269A (en) | Disk array apparatus and backup method of data | |
| CN103761058A (en) | RAID1 and RAID4 hybrid structure network storage system and method | |
| CN1573705A (en) | Memory controller and data storage method | |
| JP2007122476A (en) | Data storage system, data storage controller, and write failure diagnosis method for its disk | |
| WO2014094250A1 (en) | Data processing method and device | |
| US6434711B1 (en) | Disk array apparatus that avoids premature declarations of faults | |
| CN1722074A (en) | Fault-tolerant type disk array controller | |
| US20080034155A1 (en) | Storage apparatus and conversion board | |
| JP7125602B2 (en) | Data processing device and diagnostic method | |
| CN106610788A (en) | Hard disk array control method and device | |
| JPH1195933A (en) | Disk array system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |