CN107817953B - Method and device for accessing hard disk by double-control storage equipment - Google Patents
Method and device for accessing hard disk by double-control storage equipment Download PDFInfo
- Publication number
- CN107817953B CN107817953B CN201711160024.3A CN201711160024A CN107817953B CN 107817953 B CN107817953 B CN 107817953B CN 201711160024 A CN201711160024 A CN 201711160024A CN 107817953 B CN107817953 B CN 107817953B
- Authority
- CN
- China
- Prior art keywords
- controller
- hard disk
- link
- position information
- disk
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input 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
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0614—Improving the reliability of storage systems
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/16—Handling requests for interconnection or transfer for access to memory bus
- G06F13/1668—Details of memory controller
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input 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
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0626—Reducing size or complexity of storage systems
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input 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
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input 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
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0683—Plurality of storage devices
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Signal Processing For Digital Recording And Reproducing (AREA)
Abstract
The application provides a method and a device for accessing a hard disk by a double-control storage device, which are applied to the storage device, wherein a back plate of the storage device is inserted into a plurality of double-port NVMe hard disks and is provided with two controllers which are arranged oppositely, and each connector of each controller is connected with each hard disk on the back plate based on a principle of proximity; each controller pre-configures two link mapping tables, the method comprising: a CPU of a controller receives an IO request; the IO request carries disk position information of a target hard disk; the CPU selects a link mapping table corresponding to a level signal through the level signal on a designated pin; the values of level signals on the appointed pins of different controllers are different; and the CPU searches the selected link mapping table according to the disk position information in the IO request, determines the identifier of the physical link corresponding to the disk position information, and sends a high-speed PCIe signal to the target hard disk through the physical link. The method and the device reduce the length of the physical link, reduce the manufacturing cost of the back plate, and improve the reliability and stability of the storage system.
Description
Technical Field
The present application relates to the field of computers, and in particular, to a method and an apparatus for accessing a hard disk by a dual-control storage device.
Background
NVMe (Non-Volatile Memory standard) hard disks are a specification of SSD (Solid State Drives) using pcie (peripheral component interconnect express) channels. The NVMe hard disk is used for storing data, and has lower power consumption compared with an SAS (Serial Attached SCSI) hard disk, reduced access delay and higher performance.
In a storage system with high requirements on stability, a dual-controller redundancy architecture is usually adopted, and when one controller fails, the other controller can automatically take over the service of the other controller, so that the reliability of the storage system is improved. In addition, other components such as a power supply, a battery, a fan and the like are designed in a redundant mode, the system has no single-point fault, and data safety of a user is guaranteed.
The NVMe hard disk for the dual-controller redundancy architecture is a dual-port NVMe hard disk, and the dual controllers of the storage system are communicated with the hard disk through one port respectively.
Referring to fig. 1, which is a schematic structural diagram of a backplane of a storage device shown in this application, as shown in fig. 1, a dashed frame portion on the backplane is divided into two controllers, and portions indicated by letters a and B are main bodies of the controllers, which include devices such as a CPU (Central Processing Unit), a CP L D (Complex Programmable logic device), and a BMC (Baseboard Management Controller) of the controllers, and other portions of the controllers are connectors.
Referring to fig. 2, which is a schematic diagram illustrating the device on backplane of the present application, as shown in fig. 2, when the upper board (upper controller) is inserted onto the backplane in the same manner as the lower board (lower controller) by DIP (dual inline package) technology, the connectors of the upper board may penetrate through the backplane, which may cause the connectors of the upper board to interfere with the connectors of the hard disk drives 1-25, see the portion inside the dashed line frame, and thus the connectors of the upper board may not be completely inserted into the backplane, which increases the device height of the storage device and increases the cost.
To solve the above problem, referring to fig. 3, a schematic diagram of the device insertion on the backplane according to the present application is shown, as shown in fig. 3, the upper board is inserted into the backplane after being turned over by 180 degrees, so that the controller on the upper board can avoid the connectors of hard disks No. 1-25. When inserted in this manner, i.e., in the configuration shown in fig. 1, the controller body of the upper plate is at the rightmost side, and the controller body of the lower plate is at the leftmost side, with the two controllers facing each other on the back plate.
The controller is communicated with the hard disks through links between the connectors and the hard disks; the link includes a physical link and a logical link, the physical link refers to a passive point-to-point physical connection, and the logical link refers to a data link and is a physical link plus necessary communication procedures.
The lower board can communicate with each hard disk through a link between a connector near each hard disk and each hard disk, for example, taking fig. 1 as an example, the lower board communicates with the rightmost hard disks No. 24, No. 25, No. 49 and No. 50 through the connector 17, and communicates with the leftmost hard disks No. 1, No. 2, No. 26 and No. 27 through the connector 1. After the upper board is inserted into the back board after being turned over, the positions of the connectors of the upper board are changed, and if the upper board is still communicated with the hard disks according to the previous links, the adopted physical links are crossed. In this case, if the physical links are not to cross, the physical links on the backplane need to be printed in layers, increasing backplane manufacturing costs and reducing the processability of the backplane. In addition, the upper board communicates with each hard disk according to the previous link, and the physical link is too long, so that the upper board is easily interfered by external factors of the environment in practical application, and the reliability and the stability are insufficient.
Disclosure of Invention
In view of this, the present application provides a method and an apparatus for accessing a hard disk by a dual-control storage device, so as to reduce the length of a physical link and avoid crossing of the physical link, thereby reducing the manufacturing cost of a backplane and improving the reliability and stability of a storage system.
Specifically, the method is realized through the following technical scheme:
a method for accessing hard disks by a double-control storage device is applied to the storage device, a backboard of the storage device is inserted into a plurality of double-port NVMe hard disks, two controllers which are oppositely arranged exist, and each connector of each controller is connected with each hard disk on the backboard based on a principle of proximity; each controller pre-configures two link mapping tables, wherein each link mapping table comprises a mapping relation of identifiers of a logical link and a physical link, and the logical link comprises disk position information of each hard disk; the method comprises the following steps:
a CPU of the controller receives an IO request; the IO request carries disk position information of a target hard disk;
the CPU of the controller selects the link mapping table corresponding to the level signal through the level signal on the appointed pin; the values of the level signals on the appointed pins of different controllers are different;
and the CPU of the controller searches the selected link mapping table according to the disk position information in the IO request, determines the identification of a physical link corresponding to the disk position information, and sends a PCIe (peripheral component interconnect express) signal to the target hard disk through the physical link.
In the method for accessing a hard disk by the dual-control storage device, the method further includes:
the CP L D of the controller receives an in-place signal sent by the target hard disk after being inserted into the backboard;
the CP L D of the controller selects the link mapping table corresponding to the level signal through the level signal on the designated pin;
the CP L D of the controller searches the selected link mapping table based on the identification of the physical link receiving the on-position signal, and determines the inventory information corresponding to the identification of the physical link;
and the CP L D of the controller reports the in-place signal and the disk position information to the CPU of the controller, so that the CPU of the controller determines that the target hard disk is inserted into the backboard.
In the method for accessing a hard disk by the dual-control storage device, the method further includes:
the CPU of the controller receives the in-place signal and the disk position information of the target hard disk and sends a power-on instruction to a CP L D of the controller, wherein the power-on instruction carries the disk position information of the target hard disk;
the CP L D of the controller receives the power-on instruction and selects the link mapping table corresponding to the level signal through the level signal on the appointed pin;
and the CP L D of the controller searches the selected link mapping table according to the disk position information in the power-on command, determines the identification of a physical link corresponding to the disk position information, and sends a power-on signal to the target hard disk through the physical link.
In the method for accessing a hard disk by the dual-control storage device, the method further includes:
after the CPU of the controller sends the power-on command to a preset interval duration, a release reset command is sent to a CP L D of the controller, wherein the release reset command carries the disk position information of the target hard disk;
the CP L D of the controller receives the release reset instruction and selects the link mapping table corresponding to the level signal through the level signal on the appointed pin;
the CP L D of the controller searches the selected link mapping table through the disk position information in the release reset instruction, determines the identifier of the physical link corresponding to the disk position information, and sends a release reset signal to the target hard disk through the physical link.
In the method for accessing the hard disk by the dual-control storage device, the CPU of the controller is connected with the CP L D of the controller through an L PC interface.
A device for accessing hard disks by double-control storage equipment is applied to the storage equipment, a backboard of the storage equipment is inserted into a plurality of double-port NVMe hard disks, two controllers which are oppositely arranged exist, and each connector of each controller is connected with each hard disk on the backboard based on a principle of proximity; each controller pre-configures two link mapping tables, wherein each link mapping table comprises a mapping relation of identifiers of a logical link and a physical link, and the logical link comprises disk position information of each hard disk; the device comprises:
a first receiving unit, configured to receive an IO request; the IO request carries disk position information of a target hard disk;
a first selection unit for selecting the link mapping table corresponding to a level signal through the level signal on a designated pin; the values of the level signals on the appointed pins of different controllers are different;
and the first searching unit is used for searching the selected link mapping table according to the disk position information in the IO request, determining the identifier of a physical link corresponding to the disk position information, and sending a high-speed PCIe signal to the target hard disk through the physical link.
In the apparatus for accessing a hard disk by the dual-control storage device, the apparatus further comprises:
the second receiving unit is used for receiving an in-place signal sent by the target hard disk after being inserted into the backboard;
a second selection unit, configured to select, through a level signal on the designated pin, the link mapping table corresponding to the level signal;
the second searching unit is used for searching the selected link mapping table based on the identification of the physical link receiving the in-place signal and determining the inventory information corresponding to the identification of the physical link;
and the reporting unit is used for reporting the in-place signal and the disk position information to a CPU (central processing unit) of the controller so that the CPU of the controller determines that the target hard disk is inserted into the backboard.
In the apparatus for accessing a hard disk by the dual-control storage device, the apparatus further comprises:
the first receiving unit is further configured to receive the in-place signal and the disk position information of the target hard disk, and send a power-on instruction to a CP L D of the controller, where the power-on instruction carries the disk position information of the target hard disk;
the second receiving unit is further configured to receive the power-on instruction, and select the link mapping table corresponding to the level signal through the level signal on the designated pin;
the second searching unit is further configured to search the selected link mapping table according to the disk position information in the power-on instruction, determine an identifier of a physical link corresponding to the disk position information, and send a power-on signal to the target hard disk through the physical link.
In the apparatus for accessing a hard disk by the dual-control storage device, the apparatus further comprises:
the first receiving unit is further configured to send a release reset instruction to the CP L D of the controller after the power-on instruction is sent for a preset interval duration, where the release reset instruction carries the disk location information of the target hard disk;
the second receiving unit is further configured to receive the release reset instruction, and select the link mapping table corresponding to the level signal through the level signal on the designated pin;
the second searching unit is further configured to search the selected link mapping table through the disk position information in the release reset instruction, determine an identifier of a physical link corresponding to the disk position information, and send a release reset signal to the target hard disk through the physical link.
In the device for accessing the hard disk by the dual-control storage equipment, the CPU of the controller is connected with the CP L D of the controller through an L PC interface.
In the technical scheme of the application, connectors of two controllers on a back plate of a storage device are connected with hard disks on the back plate nearby, and the two controllers are preconfigured with two link mapping tables, wherein the link mapping tables comprise mapping relations of logical links and physical links, and the logical links comprise disk position information of the hard disks; the CPU of the controller receives an IO request, the IO request carries the disk position information of a target hard disk, the CPU of the controller can select the link mapping table corresponding to the level signal through the level signal on a designated pin, then the disk position information in the IO request searches the selected link mapping table, the identification of a physical link corresponding to the disk position information is determined, and a high-speed PCIe signal is sent to the target hard disk through the physical link;
because each connector of the two controllers is connected with each hard disk based on the principle of proximity, the physical link corresponding to the disk position information of the target hard disk in the link mapping table selected by the CPU of the controller is the shortest physical link of the controller connected with the target hard disk; crossed physical links can not appear on the backboard, and the physical links do not need to be printed in a layered mode, so that the length of the physical links of the backboard is reduced, the manufacturing cost of the backboard is reduced, and the reliability and the stability of the storage system are also improved.
Drawings
FIG. 1 is a schematic structural diagram of a backplane of a memory device shown in the present application;
FIG. 2 is a schematic illustration of a device on backplane plug-in package shown herein;
FIG. 3 is a schematic diagram of another device on backplane plug-in assembly shown in the present application;
FIG. 4 is a schematic diagram illustrating the connection of devices on a backplane of a memory device shown in the present application;
FIG. 5 is a flow chart illustrating a method for a dual-provisioned storage device to access a hard disk according to the present application;
fig. 6 is a block diagram illustrating an embodiment of an apparatus for accessing a hard disk by a dual-control storage device according to the present application.
Detailed Description
In order to make the technical solutions in the embodiments of the present invention better understood and make the above objects, features and advantages of the embodiments of the present invention more comprehensible, the following description of the prior art and the technical solutions in the embodiments of the present invention with reference to the accompanying drawings is provided.
The upper plate of the backboard of the storage device is inserted onto the backboard through 180-degree overturning, and is arranged on the backboard in opposite direction with the lower plate. The lower board can communicate with each hard disk through the link between the connector near each hard disk and each hard disk. After the upper board is turned over, the positions of the connectors are changed, and if the communication is still performed with the hard disks according to the original links, the physical links between the connectors and the hard disks are crossed. Still taking fig. 1 as an example, if the upper board communicates with hard disk No. 1 and hard disk No. 25 according to the original link, the upper board is connected to the 25 hard disk No. 25 on the rightmost side through connector 17 and connected to the 1 hard disk on the leftmost side through connector 1, and at this time, the two physical links intersect.
If the physical links are to be left uncrossed, the physical links on the backplane need to be printed in layers, increasing backplane manufacturing costs and reducing backplane processability.
In addition, the upper board communicates with each hard disk according to the original link, and the physical link between the controller and the hard disk is too long, so that the physical link is easily interfered by external factors of the environment in practical application, and the reliability and stability of the storage system are insufficient.
In view of this, the present application provides a method for accessing a hard disk by a dual-control storage device, where each connector of a controller is connected to each hard disk on a backplane based on a principle of proximity, and two link mapping tables are preconfigured on the controller, where the link mapping tables include mapping relationships between identifiers of a logical link and a physical link, and the logical link includes disk location information of each hard disk; the controller can select a corresponding link mapping table through a level signal on a designated pin, and then send a signal to the target hard disk through a physical link corresponding to the disk position information of the target hard disk in the selected link mapping table.
Referring to fig. 4, a schematic diagram of the connection of devices on a backplane of a storage device shown in the present application, where a CPU and a CP L D both belong to a controller, the CPU is connected to a CP L D through a L pc (low pin count) interface, a dual-port NVMe hard disk includes two ports (i.e., an a port and a B port in the figure), where the two ports may be standard SFF-8639 ports, the two controllers are respectively connected to the dual-port NVMe hard disk through the two ports, after each connector of the controller is connected to each hard disk on the backplane based on the proximity principle, the CPU of the controller and the CP L D may be respectively connected to each hard disk nearby, where a high-speed PCIe link of the CPU is a physical link from the CPU to the connector plus a physical link from the connector to the hard disk, and a low-speed signal link of the CP L D is a physical link from the CP L D to the connector plus a physical link from the connector to the hard disk.
With continued reference to fig. 5, for the method for accessing a hard disk by a dual-control storage device shown in the present application, the method is applied to a controller of the storage device, a plurality of dual-port NVMe hard disks are inserted into a backplane of the storage device, and two controllers are oppositely arranged, and each connector of each controller is connected to each hard disk on the backplane based on a proximity principle; each controller is preconfigured with two link mapping tables, wherein each link mapping table comprises a mapping relation of identifiers of a logical link and a physical link, and the logical link comprises disk position information of each hard disk; the method comprises the following steps:
step 501: a CPU of the controller receives an IO request; and the IO request carries the disk position information of the target hard disk.
Step 502: the CPU of the controller selects the link mapping table corresponding to the level signal through the level signal on the appointed pin; and the values of the level signals on the appointed pins of different controllers are different.
Step 503: and the CPU of the controller searches the selected link mapping table according to the disk position information in the IO request, determines the identification of a physical link corresponding to the disk position information, and sends a PCIe (peripheral component interconnect express) signal to the target hard disk through the physical link.
The proximity principle may be a preset length range (for example, 5 cm), and if the distance between the connector of the controller and the connector of the hard disk is within 5 cm, the two connectors may be connected by a physical link. In practical application, corresponding adjustment can be performed according to the device position of the back plate, and the principle of proximity is flexibly applied.
Still taking fig. 1 as an example for illustration, after the connectors of the controllers and the connectors of the hard disks are connected based on the principle of proximity, the connector 17 of the upper board can be connected with the hard disk No. 1 and the hard disk No. 26, the connector 16 can be connected with the hard disks No. 2, No. 3, No. 27 and No. 28, the connector 15 can be connected with the hard disks No. 4 and No. 29, and so on, the connector 2 is connected with the hard disks No. 22, No. 23, No. 47 and No. 48, and the connector 1 is connected with the hard disks No. 24, No. 25, No. 49 and No. 50;
The two controllers are respectively pre-configured with two link mapping tables, the two link mapping tables respectively correspond to the upper plate and the lower plate, and the link mapping tables comprise mapping relations of the logical links and the physical links of the hard disks; the logical link includes disk information of the hard disk, which can indicate the position of the hard disk on the backplane, and other necessary communication procedures. The controller can correctly recognize the identity of the physical link and thereby select the corresponding physical link.
Still taking fig. 1 as an example, the identifiers of the physical links may be serial numbers 1 to 50, which respectively correspond to 50 physical links of the controller connected to the hard disk. It should be noted that, because the connection modes of the upper board and the lower board are different, in the two link mapping tables, the physical links corresponding to the disk location information of the same hard disk are different, for example, the identifier of the physical link corresponding to hard disk No. 1 in the link mapping table corresponding to the upper board is 50, and the identifier of the physical link corresponding to hard disk No. 1 in the link mapping table corresponding to the lower board is 1.
The back board can respectively send level signals to designated pins of the two controllers, and the level signals of the upper board and the lower board are different. Such as: the upper plate is high and the lower plate is low. The controller can also be pre-configured with the mapping relationship of the link mapping table and the level signal, and the subsequent controller can select the corresponding link mapping table based on the read level signal.
In the embodiment of the present application, after the dual-port NVMe hard disk is plugged into the backplane, the dual-port NVMe hard disk sends in-place signals to the two controllers respectively, and the CP L D of the controller receives the in-place signals and then can look up the signals in the pre-configured link mapping table according to the identifier of the physical link receiving the in-place signals.
Specifically, CP L D first needs to read a level signal of a designated pin, then selects a link mapping table corresponding to the level signal according to the read level signal, and finds an identifier of a physical link receiving the above-mentioned on-site signal in the link mapping table, and determines the disk information corresponding to the identifier of the physical link.
And after the CPU receives the disk position information and the in-place signal, the CPU can determine that a target hard disk is inserted in the disk position indicated by the disk position information.
In the embodiment of the application, the CPU can power on the target hard disk inserted into the backboard after receiving the on-site signal, and the CPU can send a power-on instruction to the CP L D through an L PC interface, wherein the power-on instruction carries the disk position information of the target hard disk.
CP L D receives the above-mentioned power-on command, first reads the level signal of the appointed pin, then selects the corresponding link mapping table according to the read level signal, and searches the disk position information in the above-mentioned power-on command in the link mapping table, and determines the identification of the physical link corresponding to the above-mentioned disk position information CP L D can determine the physical link based on the found identification, and sends the power-on signal to the above-mentioned target hard disk through the physical link.
Taking fig. 1 as an example for illustration, on one hand, if the CPU of the upper board receives the disk location information and the on-site information of the hard disk No. 1, and sends a power-on command carrying the disk location information to the CP L D of the upper board through the L PC interface, and the CP L D receives the power-on command, may select the link mapping table corresponding to the upper board according to the level signal of the specified pin, and then searches the link mapping table according to the disk location information of the hard disk No. 1, and determines that the identifier of the corresponding physical link is 50, then the CP L D may send the power-on signal to the hard disk No. 1 through the physical link identified as 50.
On the other hand, if the CPU of the lower board receives the disk position information and the on-position information of the hard disk No. 1, a power-on command carrying the disk position information may be sent to the CP L D of the lower board through the L PC interface, the CP L D may receive the power-on command, may select the link mapping table corresponding to the lower board according to the level information of the specified pin, then searches the link mapping table according to the disk position information of the hard disk No. 1, and determines that the identifier of the corresponding physical link is 1, and then the CP L D may send the power-on signal to the hard disk No. 1 through the physical link identified as 1.
And the target hard disk receives the power-on signal, is in a reset state by default after being successfully powered on, and can be released to reset if the controller needs to perform further operation on the target hard disk subsequently.
The method includes that an interval duration is preset on a CPU of the controller, and represents the waiting time required after the CPU of the controller sends a power-on command, the interval duration can be preconfigured based on practical application conditions, for example, the interval duration is set to 500 milliseconds, the CPU of the controller can count time after sending the power-on command for the target hard disk, and sends a release reset command to a CP L D through an L PC interface after reaching the preset interval duration, wherein the release reset command carries disk position information of the target hard disk.
CP L D receives the above-mentioned release reset instruction, first reads the level signal of the appointed pin, then selects the corresponding link mapping table according to the read level signal, and searches the disk position information in the above-mentioned release reset instruction in the link mapping table, and determines the identification of the physical link corresponding to the above-mentioned disk position information CP L D can determine the physical link based on the found identification, and sends the release reset signal to the above-mentioned target hard disk through the physical link, thereby realizing the release reset process of the target hard disk.
And after the controller releases and resets the target hard disk, the controller can subsequently process the IO request aiming at the target hard disk.
In the embodiment of the application, when upper software of the storage device stores or acquires data to the target hard disk, an IO request may be sent to the controller; and the IO request carries the disk position information of the target hard disk. And the CPU of the controller receives the IO request.
In this embodiment of the present application, after receiving an IO request, a CPU of the controller may first read a level signal of a designated pin of the controller, determine whether the controller in which the CPU is located is an upper board or a lower board based on the read level signal, and then select a link mapping table corresponding to the level signal according to the read level signal. Because the mapping relation between the level signal and the link mapping table is pre-configured on the controller, the CPU can select the link mapping table corresponding to the controller.
In this embodiment of the present application, the CPU of the controller may search the selected link mapping table according to the disk position information in the IO request, and determine an identifier of a physical link corresponding to the disk position information. The CPU may send a PCIe high-speed signal to the target according to the found physical link. The PCIe high-speed signal may be a data signal.
Still taking fig. 1 as an example for explanation, on one hand, if the CPU of the upper board receives an IO request for the hard disk No. 1, the corresponding link mapping table may be selected according to the level signal of the designated pin, then the link mapping table is searched according to the disk location information of the hard disk No. 1, and it is determined that the identifier of the corresponding physical link is 50, the CPU of the upper board may send a PCIe high-speed signal to the hard disk No. 1 through the physical link identified as 50.
On the other hand, if the CPU of the lower board receives the IO request for the hard disk No. 1, the corresponding link mapping table may be selected according to the level signal of the designated pin, and then the link mapping table is searched according to the disk location information of the hard disk No. 1, and it is determined that the identifier of the corresponding physical link is 1, the CPU of the lower board may send the PCIe high-speed signal to the hard disk No. 1 through the physical link identified as 1.
In summary, in the technical solution of the present application, a backplane of the storage device is inserted into a plurality of dual-port NVMe hard disks, two controllers are oppositely disposed on the backplane, and each connector of each controller is connected to each hard disk on the backplane based on a principle of proximity; each controller is pre-configured with two link mapping tables, the two link mapping tables respectively correspond to an upper board (an upper controller) and a lower board (a lower controller) of a backboard, and each link mapping table comprises a mapping relation of identifications of a logical link and a physical link;
a CPU of a controller receives an IO request, wherein the IO request carries disk position information of a target hard disk, and the CPU of the controller can select a link mapping table corresponding to a level signal through the level signal on a designated pin, wherein values of the level signal on the designated pin of different controllers are different; the CPU of the controller may search the selected link mapping table according to the disk position information in the IO request, determine an identifier of a physical link corresponding to the disk position information, and send a PCIe at high speed to the target hard disk through the physical link;
because each connector of the two controllers is connected with each hard disk based on the principle of proximity, the physical link corresponding to the disk position information of the target hard disk in the link mapping table selected by the controller is the shortest physical link of the controller connected with the target hard disk;
in addition, the CP L D of the controller receives a power-on instruction or a release reset instruction sent by the CPU, and may also select a link mapping table corresponding to the level signal through the level signal on the designated pin, and search the link mapping table according to the disk location information carried by the power-on instruction or the release reset instruction, thereby selecting a corresponding physical link to communicate with the target hard disk;
because the connectors of the upper plate and the lower plate of the back plate are connected with the hard disks based on the principle of proximity, crossed physical links can not appear on the back plate, and the physical links do not need to be printed in a layered mode, so that the length of the physical links is reduced, the manufacturing cost of the back plate is reduced, and the reliability and the stability of the storage system are improved.
Corresponding to the embodiment of the method for accessing the hard disk by the dual-control storage device, the application also provides an embodiment of a device for accessing the hard disk by the dual-control storage device.
Referring to fig. 6, a block diagram of an embodiment of an apparatus for accessing a hard disk by a dual-controlled storage device according to the present application is shown, where the apparatus 60 for accessing a hard disk by a dual-controlled storage device includes:
a first receiving unit 610, configured to receive an IO request; and the IO request carries the disk position information of the target hard disk.
A first selection unit 620 for selecting the link mapping table corresponding to a level signal through the level signal on a designated pin; and the values of the level signals on the appointed pins of different controllers are different.
A first searching unit 630, configured to search the selected link mapping table according to the disk information in the IO request, determine an identifier of a physical link corresponding to the disk information, and send a PCIe at high speed to the target hard disk through the physical link.
In this example, the apparatus further comprises:
a second receiving unit 640 (not shown in the figure) for receiving an in-place signal sent by the target hard disk inserted into the backplane.
A second selection unit 650 (not shown in the figure) for selecting the link mapping table corresponding to the level signal through the level signal on the designated pin.
A second lookup unit 660 (not shown in the figure) configured to lookup the selected link mapping table based on the identity of the physical link receiving the on-bit signal, and determine the parking information corresponding to the identity of the physical link.
A reporting unit 670 (not shown in the figure) for reporting the on-position signal and the disk position information to the CPU of the controller, so that the CPU of the controller determines that the target hard disk is inserted into the backplane.
In this example, the apparatus further comprises:
the first receiving unit 610 is further configured to receive the in-place signal and the disk position information of the target hard disk, and send a power-on instruction to the CP L D of the controller, where the power-on instruction carries the disk position information of the target hard disk.
The second receiving unit 640 (not shown in the figure) is further configured to receive the power-on instruction, and select the link mapping table corresponding to the level signal through a level signal on a designated pin.
The second searching unit 660 (not shown in the figure) is further configured to search the selected link mapping table according to the disk location information in the power-on instruction, determine an identifier of a physical link corresponding to the disk location information, and send a power-on signal to the target hard disk through the physical link.
In this example, the apparatus further comprises:
the first receiving unit 610 is further configured to send a release reset instruction to the CP L D of the controller after the power-on instruction is sent for a preset interval duration, where the release reset instruction carries the disk location information of the target hard disk;
the second receiving unit 640 (not shown in the figure) is further configured to receive the release reset instruction, and select the link mapping table corresponding to the level signal through the level signal on the designated pin;
the second searching unit 660 (not shown in the figure) is further configured to search the selected link mapping table through the disk position information in the release reset instruction, determine an identifier of a physical link corresponding to the disk position information, and send a release reset signal to the target hard disk through the physical link.
In this example, the CPU of the controller is connected to the CP L D of the controller through an L PC interface.
The implementation process of the functions and actions of each unit in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the application. One of ordinary skill in the art can understand and implement it without inventive effort.
The above description is only exemplary of the present application and should not be taken as limiting the present application, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the scope of protection of the present application.
Claims (10)
1. A method for accessing hard disks by a double-control storage device is applied to the storage device and is characterized in that a backboard of the storage device is inserted into a plurality of dual-port NVMe hard disks, two controllers which are oppositely arranged exist, and each connector of each controller is connected with each hard disk on the backboard based on a principle of proximity; the controller structures of the two controllers are the same, each controller is preconfigured with two link mapping tables, wherein each link mapping table comprises a mapping relation of the identifiers of a logical link and a physical link, and the logical link comprises disk position information of each hard disk; the method comprises the following steps:
a CPU of the controller receives an IO request; the IO request carries disk position information of a target hard disk;
the CPU of the controller selects the link mapping table corresponding to the level signal through the level signal on the appointed pin; the values of the level signals on the appointed pins of different controllers are different;
and the CPU of the controller searches the selected link mapping table according to the disk position information in the IO request, determines the identification of a physical link corresponding to the disk position information, and sends a PCIe (peripheral component interconnect express) signal to the target hard disk through the physical link.
2. The method of claim 1, further comprising:
the CP L D of the controller receives an in-place signal sent by the target hard disk after being inserted into the backboard;
the CP L D of the controller selects the link mapping table corresponding to the level signal through the level signal on the designated pin;
the CP L D of the controller searches the selected link mapping table based on the identification of the physical link receiving the on-position signal, and determines the inventory information corresponding to the identification of the physical link;
and the CP L D of the controller reports the in-place signal and the disk position information to the CPU of the controller, so that the CPU of the controller determines that the target hard disk is inserted into the backboard.
3. The method of claim 2, further comprising:
the CPU of the controller receives the in-place signal and the disk position information of the target hard disk and sends a power-on instruction to a CP L D of the controller, wherein the power-on instruction carries the disk position information of the target hard disk;
the CP L D of the controller receives the power-on instruction and selects the link mapping table corresponding to the level signal through the level signal on the appointed pin;
and the CP L D of the controller searches the selected link mapping table according to the disk position information in the power-on command, determines the identification of a physical link corresponding to the disk position information, and sends a power-on signal to the target hard disk through the physical link.
4. The method of claim 3, further comprising:
after the CPU of the controller sends the power-on command to a preset interval duration, a release reset command is sent to a CP L D of the controller, wherein the release reset command carries the disk position information of the target hard disk;
the CP L D of the controller receives the release reset instruction and selects the link mapping table corresponding to the level signal through the level signal on the appointed pin;
the CP L D of the controller searches the selected link mapping table through the disk position information in the release reset instruction, determines the identifier of the physical link corresponding to the disk position information, and sends a release reset signal to the target hard disk through the physical link.
5. The method of claim 4, wherein the CPU of the controller is connected to the CP L D of the controller via a low pin count L PC interface.
6. A device for accessing hard disks by a double-control storage device is applied to the storage device and is characterized in that a back plate of the storage device is inserted into a plurality of dual-port NVMe hard disks, two controllers which are oppositely arranged exist, each connector of each controller is connected with each hard disk on the back plate based on a principle of proximity, and the controller structures of the two controllers are the same; each controller pre-configures two link mapping tables, wherein each link mapping table comprises a mapping relation of identifiers of a logical link and a physical link, and the logical link comprises disk position information of each hard disk; the device comprises:
a first receiving unit, configured to receive an IO request; the IO request carries disk position information of a target hard disk;
a first selection unit for selecting the link mapping table corresponding to a level signal through the level signal on a designated pin; the values of the level signals on the appointed pins of different controllers are different;
and the first searching unit is used for searching the selected link mapping table according to the disk position information in the IO request, determining the identifier of a physical link corresponding to the disk position information, and sending a high-speed PCIe signal to the target hard disk through the physical link.
7. The apparatus of claim 6, further comprising:
the second receiving unit is used for receiving an in-place signal sent by the target hard disk after being inserted into the backboard;
a second selection unit, configured to select, through a level signal on the designated pin, the link mapping table corresponding to the level signal;
the second searching unit is used for searching the selected link mapping table based on the identification of the physical link receiving the in-place signal and determining the inventory information corresponding to the identification of the physical link;
and the reporting unit is used for reporting the in-place signal and the disk position information to a CPU (central processing unit) of the controller so that the CPU of the controller determines that the target hard disk is inserted into the backboard.
8. The apparatus of claim 7, further comprising:
the first receiving unit is further configured to receive the in-place signal and the disk position information of the target hard disk, and send a power-on instruction to a CP L D of the controller, where the power-on instruction carries the disk position information of the target hard disk;
the second receiving unit is further configured to receive the power-on instruction, and select the link mapping table corresponding to the level signal through the level signal on the designated pin;
the second searching unit is further configured to search the selected link mapping table according to the disk position information in the power-on instruction, determine an identifier of a physical link corresponding to the disk position information, and send a power-on signal to the target hard disk through the physical link.
9. The apparatus of claim 8, further comprising:
the first receiving unit is further configured to send a release reset instruction to the CP L D of the controller after the power-on instruction is sent for a preset interval duration, where the release reset instruction carries the disk location information of the target hard disk;
the second receiving unit is further configured to receive the release reset instruction, and select the link mapping table corresponding to the level signal through the level signal on the designated pin;
the second searching unit is further configured to search the selected link mapping table through the disk position information in the release reset instruction, determine an identifier of a physical link corresponding to the disk position information, and send a release reset signal to the target hard disk through the physical link.
10. The apparatus of claim 9, wherein the CPU of the controller is connected to the CP L D of the controller through an L PC interface.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711160024.3A CN107817953B (en) | 2017-11-20 | 2017-11-20 | Method and device for accessing hard disk by double-control storage equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711160024.3A CN107817953B (en) | 2017-11-20 | 2017-11-20 | Method and device for accessing hard disk by double-control storage equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107817953A CN107817953A (en) | 2018-03-20 |
| CN107817953B true CN107817953B (en) | 2020-07-28 |
Family
ID=61609529
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711160024.3A Active CN107817953B (en) | 2017-11-20 | 2017-11-20 | Method and device for accessing hard disk by double-control storage equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107817953B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI679531B (en) * | 2018-04-18 | 2019-12-11 | 緯穎科技服務股份有限公司 | Method of hot-plugging identification and server having function of hot-plugging identification |
| CN109032851B (en) * | 2018-06-26 | 2021-01-12 | 华为技术有限公司 | A method and device for determining a link failure |
| CN112286458B (en) | 2018-06-27 | 2022-09-23 | 华为技术有限公司 | Disk, storage system and disk configuration method |
| CN110806989A (en) * | 2018-08-06 | 2020-02-18 | 浙江宇视科技有限公司 | Storage server |
| CN109407998B (en) * | 2018-11-09 | 2022-02-18 | 郑州云海信息技术有限公司 | Method, system and related assembly for IO stream synchronization in cache |
| CN112162706A (en) * | 2020-09-30 | 2021-01-01 | 新华三云计算技术有限公司 | Hard disk management method, device, equipment and machine readable storage medium |
| CN113688084B (en) * | 2021-07-22 | 2023-08-22 | 苏州浪潮智能科技有限公司 | A circuit, method and storage medium for realizing automatic allocation of PCIE resources |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101719051A (en) * | 2009-11-09 | 2010-06-02 | 中兴通讯股份有限公司 | Multi-control disk array and implementation method thereof |
| CN101833522A (en) * | 2010-03-31 | 2010-09-15 | 杭州华三通信技术有限公司 | Communication method and equipment of SAS (Serial Attached SCSI) link |
| CN104011691A (en) * | 2011-12-29 | 2014-08-27 | 英特尔公司 | Non-volatile ram disk |
| CN104426814A (en) * | 2013-08-30 | 2015-03-18 | 英特尔公司 | Numa node peripheral switch |
| CN106326154A (en) * | 2016-08-09 | 2017-01-11 | 杭州宏杉科技有限公司 | Control circuit and control method of hard disk address |
| CN106326171A (en) * | 2016-08-24 | 2017-01-11 | 联想(北京)有限公司 | Method and device for recognizing hard disk type of hard disk back plate |
| CN106919531A (en) * | 2015-12-25 | 2017-07-04 | 华为技术有限公司 | Exchange method and equipment based on non-volatile memories bus protocol |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9298648B2 (en) * | 2013-05-08 | 2016-03-29 | Avago Technologies General Ip (Singapore) Pte Ltd | 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 |
-
2017
- 2017-11-20 CN CN201711160024.3A patent/CN107817953B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101719051A (en) * | 2009-11-09 | 2010-06-02 | 中兴通讯股份有限公司 | Multi-control disk array and implementation method thereof |
| CN101833522A (en) * | 2010-03-31 | 2010-09-15 | 杭州华三通信技术有限公司 | Communication method and equipment of SAS (Serial Attached SCSI) link |
| CN104011691A (en) * | 2011-12-29 | 2014-08-27 | 英特尔公司 | Non-volatile ram disk |
| CN104426814A (en) * | 2013-08-30 | 2015-03-18 | 英特尔公司 | Numa node peripheral switch |
| CN106919531A (en) * | 2015-12-25 | 2017-07-04 | 华为技术有限公司 | Exchange method and equipment based on non-volatile memories bus protocol |
| CN106326154A (en) * | 2016-08-09 | 2017-01-11 | 杭州宏杉科技有限公司 | Control circuit and control method of hard disk address |
| CN106326171A (en) * | 2016-08-24 | 2017-01-11 | 联想(北京)有限公司 | Method and device for recognizing hard disk type of hard disk back plate |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107817953A (en) | 2018-03-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107817953B (en) | Method and device for accessing hard disk by double-control storage equipment | |
| JP6799706B2 (en) | Storage system and computing system supporting multi-protocol input/output interface | |
| US9367510B2 (en) | Backplane controller for handling two SES sidebands using one SMBUS controller and handler controls blinking of LEDs of drives installed on backplane | |
| US12093120B2 (en) | Lighting method and apparatus based on AMD platform, device and readable medium | |
| US20160188216A1 (en) | Hard Disk and Management Method | |
| US20150058506A1 (en) | Server system and method for transferring at least one chassis-specific configuration value | |
| KR20180012181A (en) | Self-Configuring SSD Multi-Protocol Support in Host-less Environment | |
| US9705984B2 (en) | System and method for sharing data storage devices | |
| CN105446657A (en) | Method for monitoring RAID card | |
| CN103929475B (en) | The hard disk storage system and hard disc data operating method of a kind of Ethernet architecture | |
| CN111684765B (en) | Server system | |
| CN103793003B (en) | A kind of power board and blade server | |
| CN206249150U (en) | A kind of storage server | |
| US8219714B2 (en) | Storage area network and method for provisioning therein | |
| US20160246746A1 (en) | Sas configuration management | |
| US9021166B2 (en) | Server direct attached storage shared through physical SAS expanders | |
| TWI802269B (en) | Server equipment and input and output device | |
| US20200233831A1 (en) | Self-configuring ssd multi-protocol support in host-less environment | |
| CN104486384A (en) | Storage system and exchange expansion device | |
| CN101052136B (en) | A backplane and its implementation method | |
| US7716440B2 (en) | Storage system and management method thereof | |
| US20110231674A1 (en) | Independent drive power control | |
| CN105700817A (en) | Just Bunch of Discs JBOD apparatus | |
| US7111066B2 (en) | Method of operating a storage device | |
| US7228338B2 (en) | Multi-service platform module |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |