CN102566944B - Store path redundancy approach - Google Patents
Store path redundancy approach Download PDFInfo
- Publication number
- CN102566944B CN102566944B CN201110459634.XA CN201110459634A CN102566944B CN 102566944 B CN102566944 B CN 102566944B CN 201110459634 A CN201110459634 A CN 201110459634A CN 102566944 B CN102566944 B CN 102566944B
- Authority
- CN
- China
- Prior art keywords
- module
- infinite bandwidth
- store path
- path redundancy
- application
- 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.)
- Active
Links
Abstract
According to an aspect of the present invention, provide a kind of store path redundancy approach, comprising: step S1: application module is communicated by the first infinite bandwidth module and memory module; Step S2: application module detects that the first infinite bandwidth module breaks down; And step S3: application module is communicated by the second infinite bandwidth module and memory module.The present invention by InfiniBand framework applications in store path redundancy, by InfiniBand access data between application server and memory device, thus when ensure that storage system is normally run, greatly improve memory property, improve efficiency.
Description
Technical field
The present invention relates to server system substantially, more specifically, relates to a kind of store path redundancy approach.
Background technology
Disk array (Disk Array) is that make the read-write of multiple hard disk synchronous, minimizing mistake, increases the technology of efficiency and fiduciary level by a hard disk controller to control being interconnected of multiple hard disk.What this technology was realized is exactly disk array product, and common physical form is exactly contain the equipment such as several hard disks in a rectangular parallelepiped, provides the service of different stage with certain organizational form.Determine that one of key index of disk array properties of product is the host interface that disk array product uses, the host interface type of the disk array product of main flow of the prior art mainly contains: iSCSI, adopt the storage protocol based on TCP/IP, the interface rate of current main flow is 1Gb/s (based on gigabit Ethernet network), 10Gb/s (based on 10Gb ten thousand mbit ethernet network); FC (optical-fibre channel), existing market main flow FC interface rate has 4Gb/s, 8Gb/s two kinds; SAS (Serial Attached SCSI (SAS) interface), existing market main flow interface rate has 3Gb/s, 6Gb/s two kinds.
In traditional store path redundancy approach, employ above-mentioned legacy interface as path, thus do not reach application performance demand under harsh applied environment is required to memory property, cause disk array product to become the bottleneck of application system.Therefore, adopting other more high performance host interface type, is the necessary ways promoting disk array properties of product.
Summary of the invention
For above-mentioned defect of the prior art, the present invention proposes a kind of store path redundancy approach, by store path redundancy approach described in the invention, solve the technical matters how promoting memory property.
According to an aspect of the present invention, provide a kind of store path redundancy approach, comprising: step S1: application module is communicated by the first infinite bandwidth module and described memory module; Step S2: described application module detects that described first infinite bandwidth module breaks down; And step S3: described application module is communicated by the second infinite bandwidth module and described memory module.
In this store path redundancy approach, described step S2 comprises: by Device Mapper mechanism, described application module detects that described first infinite bandwidth module breaks down.
In this store path redundancy approach, described step S1 comprises: application module reads information by the first infinite bandwidth module from described memory module, or stores information in described memory module; And described step S3 comprises: application module reads information by the second infinite bandwidth module from described memory module, or stores information in described memory module.
In this store path redundancy approach, described memory module comprises mainboard, memory device, and described mainboard comprises PCIE slot.
In this store path redundancy approach, described first infinite bandwidth module and described second infinite bandwidth module are communicated by described PCIE slot and described memory module.
In this store path redundancy approach, described first infinite bandwidth module comprises the first infinite bandwidth host channel adapter and the first infinite bandwidth cable, and described second infinite bandwidth module comprises the second infinite bandwidth host channel adapter and the second infinite bandwidth cable.
In this store path redundancy approach, described memory device comprises disk array.
In this store path redundancy approach, described disk array comprises RAID card and multiple hard disk, and wherein, described multiple hard disk is connected with described RAID card, and described RAID card is connected with described mainboard.
In this store path redundancy approach, described application module comprises application server.
In this store path redundancy approach, described memory module, described application module and described first infinite bandwidth module or described second infinite bandwidth module composition storage area network.
By store path redundancy approach described in the invention, memory property of can greatly deducting a percentage.
Accompanying drawing explanation
Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for instructions, together with embodiments of the present invention for explaining the present invention, is not construed as limiting the invention.In the accompanying drawings:
Fig. 1 is the process flow diagram of the overview embodiment of store path redundancy approach having thus described the invention;
Fig. 2 is the process flow diagram of the specific embodiment of store path redundancy approach having thus described the invention;
Fig. 3 is the structural drawing of the embodiment of the storage organization realizing store path redundancy approach having thus described the invention.
Embodiment
Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are described, should be appreciated that preferred embodiment described herein is only for instruction and explanation of the present invention, is not intended to limit the present invention.
Fig. 1 is the process flow diagram of the overview embodiment of store path redundancy approach having thus described the invention.In FIG:
Step S100: application module is communicated by the first infinite bandwidth module and memory module.
Step S102: application module detects that the first infinite bandwidth module breaks down.
Step S104: application module is communicated by the second infinite bandwidth module and memory module.
Wherein, this first infinite bandwidth module comprises the first infinite bandwidth host channel adapter and the first infinite bandwidth cable, and the second infinite bandwidth module comprises the second infinite bandwidth host channel adapter and the second infinite bandwidth cable.Memory device comprises disk array, and preferably, this disk array can comprise RAID card and multiple hard disk, and wherein, multiple hard disk is connected with RAID card, and RAID card is connected with mainboard.Wherein, this application module comprises application server.
In one embodiment, memory module, application module and the first infinite bandwidth module or the second infinite bandwidth module composition storage area network.
Infinite bandwidth (InfiniBand) is a kind of support how concurrent link " Convertion cable " technology.InfiniBand High speed network provides point-to-point bidirectional linked list to connect at a high speed.The InfiniBand QDR network of current main flow, its theoretical bandwidth, up to 40Gb/s, postpones to only have 1 microseconds, is far superior to other network interconnection technologies.InfiniBand technology has played important effect in fields such as high-performance calculations.In the global TOP500 supercomputer list in June, 2011, have and use InfiniBand network more than 40%.
The present invention by InfiniBand framework applications in store path redundancy, by InfiniBand access data between application server and memory device, thus when ensure that storage system is normally run, greatly improve memory property, improve efficiency.
Fig. 2 is the process flow diagram of the specific embodiment of store path redundancy approach having thus described the invention.In fig. 2:
Step S200: application module reads information by the first infinite bandwidth module from memory module, or stores information in memory module; And
Step S202: by Device Mapper mechanism, application module detects that the first infinite bandwidth module breaks down.
Step S204: application module reads information by the second infinite bandwidth module from memory module, or stores information in memory module.
The present invention by InfiniBand framework applications in store path redundancy, by InfiniBand access data between application server and memory device, thus when ensure that storage system is normally run, greatly improve memory property, improve efficiency.In addition, by using the DeviceMapper mechanism of Linux, can monitor the situation of memory module and application module, ensure that storage system can be made a response in time when store path breaks down.
Fig. 3 is the structural drawing of the embodiment of the storage system realizing store path redundancy approach having thus described the invention.
Memory module 300 and application module 308 is comprised in storage system illustrated in fig. 3.This memory module 300 comprises mainboard 302, and this mainboard 302 comprises multiple PCIE slot 204.Fig. 3 shows three PCIE slots 204, but is not limited to this quantity.In three PCIE slots 204 illustrated in fig. 3, a PCIE slot 204, for connecting RAID card 310, this RAID card 310 is provided with multiple hard disk 212 (although illustrated three hard disks 212 in Fig. 3, being not limited to this quantity).Two other PCIE card is all for connecting wireless bandwidth (InfiniBand) module, these two InfiniBand modules both can use when not breaking down simultaneously, thus improve data-transmission efficiency by the mode increasing bandwidth, and also can not block the access of data when an InfiniBand module breaks down.Store path redundancy approach having thus described the invention, these two InfiniBand modules also can make redundancy mutually.When an InfiniBand module breaks down, start another InfiniBand module, ensure that the normal operation of storage system.
Wherein, the first infinite bandwidth module 306 comprises the first infinite bandwidth host channel adapter and the first infinite bandwidth cable, and the second infinite bandwidth module 314 comprises the second infinite bandwidth host channel adapter and the second infinite bandwidth cable.Wherein, first infinite bandwidth host channel adapter and the second infinite bandwidth host channel adapter are all arranged on memory module 300 and application module 308, then by the first infinite bandwidth cable, the first infinite bandwidth host channel adapter in memory module 300 is connected with the first infinite bandwidth host channel adapter in application module 308, by the second infinite bandwidth cable, the second infinite bandwidth host channel adapter in memory module 300 is connected with the second infinite bandwidth host channel adapter in application module 308.Application module 308 comprises application server.
The present invention by InfiniBand framework applications in store path redundancy, by InfiniBand access data between application server and memory device, thus when ensure that storage system is normally run, greatly improve memory property, improve efficiency.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (6)
1. a store path redundancy approach, is characterized in that, described method adopts InfiniBand framework, comprising:
Step S1: application module is communicated by the first infinite bandwidth module and memory module;
Step S2: described application module detects that described first infinite bandwidth module breaks down; And
Step S3: described application module is communicated by the second infinite bandwidth module and described memory module;
Wherein, described step S2 comprises: by Device Mapper mechanism, described application module detects that described first infinite bandwidth module breaks down;
Wherein, described memory module comprises mainboard, memory device, and described mainboard comprises PCIE slot;
Wherein, described memory device comprises disk array;
Wherein, described memory module, described application module and described first infinite bandwidth module or described second infinite bandwidth module composition storage area network.
2. store path redundancy approach according to claim 1, is characterized in that,
Described step S1 comprises: application module reads information by the first infinite bandwidth module from described memory module, or stores information in described memory module; And
Described step S3 comprises: application module reads information by the second infinite bandwidth module from described memory module, or stores information in described memory module.
3. store path redundancy approach according to claim 1, it is characterized in that, described first infinite bandwidth module and described second infinite bandwidth module are communicated by described PCIE slot and described memory module.
4. store path redundancy approach according to claim 3, it is characterized in that, described first infinite bandwidth module comprises the first infinite bandwidth host channel adapter and the first infinite bandwidth cable, and described second infinite bandwidth module comprises the second infinite bandwidth host channel adapter and the second infinite bandwidth cable.
5. store path redundancy approach according to claim 1, it is characterized in that, described disk array comprises RAID card and multiple hard disk, and wherein, described multiple hard disk is connected with described RAID card, and described RAID card is connected with described mainboard.
6. store path redundancy approach according to claim 5, it is characterized in that, described application module comprises application server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110459634.XA CN102566944B (en) | 2011-12-31 | 2011-12-31 | Store path redundancy approach |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110459634.XA CN102566944B (en) | 2011-12-31 | 2011-12-31 | Store path redundancy approach |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102566944A CN102566944A (en) | 2012-07-11 |
| CN102566944B true CN102566944B (en) | 2015-09-02 |
Family
ID=46412469
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110459634.XA Active CN102566944B (en) | 2011-12-31 | 2011-12-31 | Store path redundancy approach |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102566944B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101808123A (en) * | 2010-03-09 | 2010-08-18 | 浪潮(北京)电子信息产业有限公司 | Method and device for accessing storage resources in storage system |
| CN101488077B (en) * | 2009-02-24 | 2011-08-24 | 浪潮电子信息产业股份有限公司 | Redundancy magnetic disk control server system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005527898A (en) * | 2002-04-18 | 2005-09-15 | インターナショナル・ビジネス・マシーンズ・コーポレーション | How to provide redundancy against channel adapter failure |
| CN101056175B (en) * | 2007-04-26 | 2011-07-20 | 华为技术有限公司 | Disk array and its access right control method and device, server and server system |
-
2011
- 2011-12-31 CN CN201110459634.XA patent/CN102566944B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101488077B (en) * | 2009-02-24 | 2011-08-24 | 浪潮电子信息产业股份有限公司 | Redundancy magnetic disk control server system |
| CN101808123A (en) * | 2010-03-09 | 2010-08-18 | 浪潮(北京)电子信息产业有限公司 | Method and device for accessing storage resources in storage system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102566944A (en) | 2012-07-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9298648B2 (en) | Method and system for I/O flow management using RAID controller with DMA capabilitiy to directly send data to PCI-E devices connected to PCI-E switch | |
| KR101340176B1 (en) | Smart scalable storage switch architecture | |
| US9342413B2 (en) | SAS RAID head | |
| US8352649B2 (en) | Storage virtualization subsystem architecture | |
| CN104202197A (en) | Equipment management method and device | |
| US8677067B2 (en) | Storage system and method for controlling storage system | |
| CN101540685B (en) | PCIe shared storage blade for blade server | |
| CN202870800U (en) | Embedded type high-capacity network storage control module | |
| US8458306B1 (en) | Coalescing change notifications in an I/O virtualization system | |
| CN104216805B (en) | A kind of high-end disk array rear end disk chassis link failure protection system and method | |
| KR102171716B1 (en) | Chained, scalable storage devices | |
| KR101209916B1 (en) | Raid-based storage control board having fibre channel interface controller and driving method thereof | |
| CN102843284B (en) | ISCSI memory node, framework and reading, wiring method | |
| CN103929475B (en) | The hard disk storage system and hard disc data operating method of a kind of Ethernet architecture | |
| US9785372B2 (en) | Storage device and method for configuring raid group | |
| US9021166B2 (en) | Server direct attached storage shared through physical SAS expanders | |
| CN201725323U (en) | Independent redundant disk array | |
| EP1768026B1 (en) | Redundant storage virtualization subsystem having data path branching functionality | |
| CN105511990B (en) | Device based on fusion architecture dual redundant degree storage control node framework | |
| CN102566944B (en) | Store path redundancy approach | |
| US20110231674A1 (en) | Independent drive power control | |
| CN105700817A (en) | Just Bunch of Discs JBOD apparatus | |
| CN102868684A (en) | Fiber channel target and realizing method thereof | |
| CN202206413U (en) | iSCSI storage node and architecture | |
| US8943237B1 (en) | Performance improvement for attached multi-storage devices |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |