CN112395220B - Processing method, device and system of shared storage controller and storage controller - Google Patents

Abstract

The invention provides a processing method, a processing device and a processing system of a shared storage controller, a storage controller and a storage medium. The method comprises the following steps: the shared storage controller receives a request of a computing node for data; the shared storage controller analyzes the request to obtain a virtual address carried in the request; the shared storage controller carries out address mapping according to an address mapping relation table to obtain a physical address corresponding to the virtual address; the shared storage controller initiates hard disk access according to the mapped physical address; and the shared storage controller receives a response result of the hard disk and forwards the response result to the computing node which initiates the request. The invention can avoid the waste of storage resources.

Description

Processing method, device, system and storage controller for shared storage controller

technical field

The present invention relates to the technical field of computer storage, in particular to a processing method, device, and system for sharing a storage controller, a storage controller, and a storage medium.

Background technique

The shared storage controller can solve the problem that the high-density server storage system occupies too much cabinet space and the storage resources cannot be allocated on demand. The shared storage controller pools storage resources. Through software configuration, the entire storage resource can be allocated on demand according to the needs of different applications. It is flexible and simple on demand.

In the process of realizing the present invention, the inventor found that there are at least the following technical problems in the prior art:

Existing implementations of shared memory controllers are all based on static address mapping. On the one hand, this method needs to configure the size of each partition one by one for each computing node using the shared storage controller, which is very cumbersome. On the other hand, this method does not allocate storage resources according to the size of the hard disk actually used by each computing node, which may cause waste of storage resources.

Contents of the invention

The processing method, device, system, storage controller and storage medium provided by the present invention can avoid waste of storage resources.

In one aspect, the present invention provides a processing method for a shared storage controller, including:

The shared storage controller receives the computing node's request for data;

The shared memory controller parses the request, and parses out the virtual address carried in the request;

The shared memory controller performs address mapping according to the address mapping relationship table, and obtains a physical address corresponding to the virtual address;

The shared storage controller initiates hard disk access according to the mapped physical address;

The shared storage controller receives the response result of the hard disk, and forwards the response result to the computing node that initiated the request.

On the other hand, the present invention also provides a shared storage controller, including:

a first receiving unit, configured to receive a computing node's request for data;

A parsing unit, configured to parse the request, and parse out the virtual address carried in the request;

a mapping unit, configured to perform address mapping according to an address mapping relationship table, and obtain a physical address corresponding to the virtual address;

The access unit is used for initiating hard disk access according to the mapped physical address;

The second receiving unit is configured to receive the response result of the hard disk, and forward the response result to the computing node that initiates the request.

On the other hand, the present invention also provides a processing system for sharing a memory controller, including:

At least one shared storage controller; a hard disk and at least one computing node directly connected to each shared storage controller;

The shared storage controller is configured to receive a request from a computing node for data; parse the request to obtain a virtual address carried in the request; perform address mapping according to an address mapping table, and obtain the virtual address corresponding to the virtual address. physical address; initiate hard disk access according to the mapped physical address; receive a response result from the hard disk, and forward the response result to the computing node that initiated the request.

On the other hand, the present invention also provides a processing device for sharing a storage controller, including:

memory;

and a processor coupled to the memory, the processor configured to perform the method based on instructions stored in the memory.

On the other hand, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions implement the method when executed by a processor.

In the above solution, the shared storage controller based on dynamic address mapping independently establishes and maintains the address mapping relationship table throughout the process, and only allocates physical address space for virtual addresses with real access, avoiding waste of storage resources.

Description of drawings

FIG. 1 is a flowchart of a processing method for a shared memory controller according to the first embodiment of the present invention;

FIG. 2 is a schematic diagram of an implementation scheme of a dynamic address mapping shared memory controller in an embodiment of the present invention;

Fig. 3 is the schematic diagram of the data structure of hard disk entry DTE of the embodiment of the present invention;

FIG. 4 is a schematic diagram of a processing flow of a dynamic address mapping shared memory controller in an embodiment of the present invention.

Fig. 5 is the schematic diagram of Disk Table Cache in the embodiment of the present invention;

FIG. 6 is a schematic diagram of a processing system sharing a memory controller in an embodiment of the present invention;

FIG. 7 is a schematic diagram of a processing flow of a shared storage controller in an embodiment of the present invention;

FIG. 8 is a schematic diagram of connection of a shared storage controller according to the present invention.

FIG. 9 is a connection diagram of a processing device sharing a memory controller according to the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is only some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

First embodiment:

As shown in Figure 1, it is a processing method of a shared memory controller according to the present invention, including:

Step 11, use the address of the hard disk accessed by each computing node as a virtual address, and use the address mapped from the virtual address as a physical address; use the space of the virtual address and the space of the physical address as a unit of a block of a predetermined size division;

Step 12, establishing an address mapping relationship table in the shared storage controller, where the address mapping relationship table is used to record the mapping relationship between virtual addresses and physical addresses in the local hard disk of the shared storage controller.

Step 13, the shared storage controller receives the computing node's request for data;

Step 14, the shared storage controller parses the request, and parses out the virtual address carried in the request;

Step 15, the shared storage controller performs address mapping according to the address mapping relationship table, and obtains a physical address corresponding to the virtual address;

Step 16, the shared storage controller initiates hard disk access according to the mapped physical address;

Step 17, the shared storage controller receives the response result of the hard disk, and forwards the response result to the computing node that initiated the request.

In the above solution, the shared storage controller based on dynamic address mapping independently establishes and maintains the address mapping relationship table throughout the process, and only allocates physical address space for virtual addresses with real access, avoiding waste of storage resources.

During the period, the address mapping relationship table includes a hard disk entry DTE;

The data structure of the hard disk entry DTE includes:

PR, used to identify whether the hard disk entry is valid, when the bit is 1, it means that the entry is valid, and when it is 0, it means that the entry is invalid;

S: It is a shared flag bit. When the bit is 1, it means that the virtual address is shared by multiple computing nodes. When it is 0, it means that it is a private address;

The computing node identification bit is used to identify the identification of the computing node that initiates the request;

The virtual address bit is a bit identifying a virtual address accessed by the computing node;

The physical address bit is a bit that identifies the physical address mapped by the shared memory controller.

Optionally, the step 15 includes:

When the first hard disk entry is found in the address mapping table, the identifier of the computing node in the first hard disk entry is consistent with the identifier of the computing node that initiated the request, and the virtual The address is the same as the virtual address in the request, then converting the virtual address of the request address into the physical address recorded in the first hard disk entry, and completing the address mapping;

When not finding the first hard disk entry in the address mapping table, then create a second hard disk entry in the address mapping table; the Req ID of the second hard disk entry is requested The computing node, the virtual address of the second hard disk entry is the virtual address carried in the request, the physical address of the second hard disk entry is the first address of the next unallocated physical address block of the hard disk, and Complete address mapping.

Second embodiment:

Another embodiment of the processing method of a shared memory controller according to the present invention includes:

Step 21, use the address of the hard disk accessed by each computing node as a virtual address, and use the address mapped from the virtual address as a physical address; use the space of the virtual address and the space of the physical address as a unit of a block of a predetermined size division;

Step 22, establishing an address mapping relationship table in the shared storage controller, where the address mapping relationship table is used to record the mapping relationship between virtual addresses and physical addresses in the local hard disk of the shared storage controller.

Step 23, the shared storage controller receives the computing node's request for data;

Step 24, when the data requested by the computing node spans at least two blocks, split the request into at least two sub-requests, each sub-request corresponds to the access of one block; parse out the data carried in each of the sub-requests virtual address.

Step 25, the shared memory controller performs address mapping according to the address mapping relationship table, and obtains a physical address corresponding to the virtual address;

Step 26, the shared storage controller initiates hard disk access according to the mapped physical address;

Step 27: The shared storage controller receives at least two response results from the hard disk, combines the responses returned by at least two accesses to the hard disk into one response, and returns the combined response to the computing node.

In the foregoing embodiments, processing can be performed on a scenario where data requested by a computing node spans at least two blocks.

Third embodiment:

Another embodiment of the processing method of a shared memory controller according to the present invention includes:

Step 31, use the address of the hard disk accessed by each computing node as a virtual address, and use the address mapped from the virtual address as a physical address; the space of the virtual address and the space of the physical address are performed in units of blocks of a predetermined size division;

Step 32, establishing an address mapping relationship table in the shared storage controller, where the address mapping relationship table is used to record the mapping relationship between virtual addresses and physical addresses in the local hard disk of the shared storage controller. And the address mapping relationship table cache space is preset in the shared storage controller, and the address mapping relationship table cache space stores hard disk entries;

Step 33, the shared storage controller receives the computing node's request for data;

Step 34, the shared storage controller parses the request, and parses out the virtual address carried in the request;

Step 35, the shared memory controller performs address mapping according to the address mapping relationship table, and obtains the physical address corresponding to the virtual address; the step 15 includes: in the address mapping relationship table and the address mapping relationship The first hard disk entry is searched in the table cache space; when the first hard disk entry is found in the address mapping table, address mapping is performed, and the search in the address mapping table is stopped.

Step 36, the shared storage controller initiates hard disk access according to the mapped physical address;

Step 37, the shared storage controller receives the response result of the hard disk, and forwards the response result to the computing node that initiated the request.

In the above embodiment, the address mapping time can be accelerated by using the cache space of the address mapping relationship table.

Fourth embodiment:

Another embodiment of the processing method of a shared memory controller according to the present invention includes:

Step 41, use the address of the hard disk accessed by each computing node as a virtual address, and use the address mapped from the virtual address as a physical address; use the space of the virtual address and the space of the physical address as a unit of a block of a predetermined size Division; pre-configure the start address and space size of the shared virtual address space;

Step 42, establishing an address mapping relationship table in the shared storage controller, where the address mapping relationship table is used to record the mapping relationship between virtual addresses and physical addresses in the local hard disk of the shared storage controller.

Step 43, the shared storage controller receives the computing node's request for data;

Step 44, the shared storage controller parses the request, and parses out the virtual address carried in the request;

Step 45, the shared storage controller performs address mapping according to the address mapping relationship table, and obtains the physical address corresponding to the virtual address; the step 15 includes: when the virtual address in the second hard disk entry is the same as When the virtual address in the request is the same, and the S bit of the second hard disk entry indicates that the virtual address is in the shared virtual address space range, the virtual address of the request address is converted into the second hard disk table The physical address recorded in the item completes the address mapping.

Step 46, the shared storage controller initiates hard disk access according to the mapped physical address;

Step 47, the shared storage controller receives the response result of the hard disk, and forwards the response result to the computing node that initiated the request.

The foregoing embodiments may be applied to a processing scenario of sharing a virtual address space.

Fifth embodiment:

Another embodiment of the processing method of a shared memory controller according to the present invention includes:

Step 41, use the address of the hard disk accessed by each computing node as a virtual address, and use the address mapped from the virtual address as a physical address; use the space of the virtual address and the space of the physical address as a unit of a block of a predetermined size division;

Step 42, establishing an address mapping relationship table in the shared storage controller, where the address mapping relationship table is used to record the mapping relationship between virtual addresses and physical addresses in the local hard disk of the shared storage controller.

Step 43: The shared storage controller receives a computing node's request for data; when the request is a write request, and the virtual address of the request is within the range of the shared virtual address space, the shared storage controller determines whether it is the main controller Compute the request initiated by the node, and generate a judgment result; if the judgment result is yes, execute step 14; otherwise, discard the request without processing.

Step 44, the shared storage controller parses the request, and parses out the virtual address carried in the request;

Step 45, the shared memory controller performs address mapping according to the address mapping relationship table, and obtains a physical address corresponding to the virtual address;

Step 46, the shared storage controller initiates hard disk access according to the mapped physical address;

Step 47, the shared storage controller receives the response result of the hard disk, and forwards the response result to the computing node that initiated the request.

The foregoing embodiments are applicable to application scenarios of multiple computing nodes.

The application scenarios of the present invention are described below. The invention proposes a processing method of a shared memory controller based on dynamic address mapping. The method dynamically determines the mapping relationship between each computing node and each hard disk according to its access address and storage resource pool conditions when each computing node accesses the hard disk. Instead of fixing the mapping relationship between each computing node and each hard disk in advance. On the one hand, this implementation method does not need to configure the mapping relationship between each partition of each computing node and each hard disk in the storage resource pool in advance, and only needs to configure the total storage capacity and shared space address used by each computing node. On the other hand, this method only performs storage resource mapping for the hard disk addresses actually used by each computing node, and does not reserve resources for unused addresses in the storage resource pool, avoiding waste of storage resources. The shared storage controller can allocate the entire storage resource pool to each computing node on demand through software configuration. And the whole process is completely transparent to each computing node.

The whole system consists of three modules: computing nodes, shared storage controllers, and hard disks, as shown in Figure 2. In this solution, the size of the storage resources required by each computing node and the address of the shared space can be configured through software. The specific mapping relationship is dynamically established and maintained by the shared storage controller itself.

Figure 2 is the implementation scheme of the dynamic address mapping shared memory controller. The hard disk address accessed by each computing node is called a virtual address, and the address mapped by the shared storage controller is called a physical address. Divide the virtual address space and physical address space in units of blocks. For example: the size of each block can be 2M Bytes. If the capacity of a physical address space is 4T Bytes, then 2M Bytes are divided into blocks, and 2M block units can be divided.

The shared storage controller itself maintains an address mapping relationship table, called Disk Table, which is responsible for maintaining the address mapping related to the hard disk of the shared storage controller. The entries contained in the Disk Table are called Disk TableEntry (DTE), and each DTE is responsible for maintaining the address mapping relationship of a block unit. For example, if the hard disk capacity of the shared storage controller is 4T Bytes, then the Disk Table in the shared storage controller contains 2M DTEs. The format of DTE is shown in Figure 3.

PR: Indicates whether the entry is valid. If the bit is 1, it means that the entry is valid, and if it is 0, it means that the entry is invalid.

S: Shared flag bit, when the bit is 1, it means that the virtual address is shared by multiple computing nodes, and when it is 0, it means it is a private address.

Computing node ID Req ID: identifies the ID of the computing node that initiated the request.

The virtual address Virtual Address[47:21] identifies the upper 27 bits of the virtual address accessed by the computing node.

Reserved: reserved bit.

Physical address Physical Address[47:21]: identifies the upper 27 bits of the physical address mapped by the shared memory controller.

After each computing node initiates a hard disk request, the request will be sent to the shared storage controller. The processing flow of the shared memory controller is shown in Figure 4, which can be divided into the following steps:

Step 1, the shared storage controller parses the request of the computing node. Parse out the request command, address, length, data and other information initiated by the computing node.

Step 2, the shared memory controller performs table lookup according to the address in the request.

If an entry containing the request address corresponding to the computing node is found in the Disk Table, that is, the Req ID in the entry is consistent with the computing node that initiated the request, and the Virtual Address[47:21] in the entry is the same as If the upper 27 bits of the request address are the same, the upper 27 bits of the request address are converted to the Physical Address[47:21] recorded in the entry to complete the address mapping.

If the entry containing the request address corresponding to the computing node cannot be found in the Disk Table, create a new entry in the DiskTable. The Req ID of this entry is the requested computing node, the Virtual Address[47:21] of this entry is the upper 27 bits of the address corresponding to the request, and the Physical Address[47:21] of this entry is the next unallocated The upper 27 bits of the first address of the physical address block, and finally set the PR of the entry. After creating a new table entry, address mapping can be completed according to this table entry.

Step 3, the shared storage controller initiates hard disk access according to the mapped physical address.

Step 4, the shared storage controller receives the result of the hard disk response, and forwards the result to the computing node that initiated the request.

In one embodiment, if a data request of a computing node spans two blocks, the data request needs to be split into two requests, and each request corresponds to the access of one block. Then perform table lookup, address mapping, and access to the hard disk for the two requests in turn. Finally, the results returned by the two visits to the hard disk are combined, and a response is returned to the corresponding computing node.

In one embodiment, in order to further increase the response speed of data requests and reduce the time spent on table lookup, a Disk Table Cache is added to the shared storage controller. 2K DTEs can be stored in the Cache. After each data request is parsed, the node number and the upper 27 bits of the request address are used to look up the table while searching in the Disk Table Cache. If it hits in the Disk Table Cache, directly perform address mapping and stop searching in the Disk Table. If there is no hit in the Disk Table Cache, after finding the DTE in the Disk Table, copy the DTE to the Disk Table Cache, as shown in Figure 5, which is a schematic diagram of the Disk Table Cache. If the Disk Table Cache is full, the DTE that has not been hit for the longest time will be removed from the Disk Table Cache.

In one embodiment, the shared storage controller based on the dynamic address mapping method disclosed in this technical solution can also support the sharing of virtual addresses by multiple computing nodes. If you want to enable the shared virtual address function, you need to configure the start address and space size of the shared virtual address space during software configuration.

After the shared virtual address is enabled, when creating a new DTE entry in the Disk Table, it is necessary to check whether the virtual address in the data request initiated by the computing node is within the range of the shared virtual address space. If it is within the range of the shared virtual address space, then Set the S shared flag bit in the DTE entry.

When the shared storage controller processes data requests initiated by each computing node and performs table lookup. If the S bit of the DTE entry matching the Virtual Address is 1, it means that the address is in the range of the shared virtual address space, then there is no need to match the Req ID, and this DTE can be used to complete address mapping.

When different computing nodes share the same segment of address space, only computing node 1 has write permission to the shared address space, and other computing nodes only have read permission to the shared address space. If the data request initiated by the computing node is a write request and the address is within the range of the shared virtual address space, it needs to be checked whether the request is initiated by the computing node 1. If it is a request initiated by computing node 1, it will be processed according to the normal process. If the request is not initiated by computing node 1, the request is directly discarded, and a normal completion response is returned to the originating computing node.

In the existing shared memory controller implementation method, the address mapping relationship table is statically configured by means of software. In this configuration method, on the one hand, the content to be configured is very cumbersome, and it is easy to make configuration errors; on the other hand, even the unused virtual address space of each computing node will be allocated the corresponding physical address space, which will cause a waste of storage resources.

The present invention has the following beneficial effects:

1. In the present invention, the shared storage controller itself maintains the address mapping relationship table, establishes a Disk Table inside the shared storage controller, completes address mapping by looking up the table, and accesses across block boundaries.

2. The present invention improves the table look-up efficiency by increasing the Disk Table Cache.

3. The present invention is based on the multi-computing node sharing virtual address of dynamic address mapping, and the dynamic address mapping shared storage controller maintains the address mapping relationship table by the shared storage controller itself, and dynamically establishes the mapping relationship between virtual address and physical address. Therefore, the cumbersome software configuration can be avoided, and the physical address space can be allocated only to the virtual address space actually used by the computing nodes, so the utilization rate of storage resources can be improved and the waste of storage resources can be avoided.

The following uses a system with 32 computing nodes as an example to introduce an embodiment of a shared storage controller based on dynamic address mapping.

As shown in Figure 6, the entire system contains 4 shared storage controllers, each shared storage controller is directly connected to 8 computing nodes and 1 hard disk, and each shared storage controller is connected to the remaining 3 shared storage controllers. The four hard disks form a storage resource pool through the shared storage controller, which is used by 32 computing nodes on demand.

Taking the data request initiated by computing node 1 as an example, the processing flow of the shared storage controller is shown in FIG. 7 .

In step 1, the computing node 1 initiates a data request to the shared controller 1, and the shared controller 1 parses the data request, and parses out the command, address, length, data and other information of the address.

In step 2, the shared memory controller 1 performs table lookup according to the address in the request. If the entry is found in the address mapping table of the shared memory controller 1, address mapping is performed according to the entry. Then, access to the hard disk 1 is initiated according to the mapped physical address. Finally, the response from hard disk 1 is forwarded to computing node 1.

Step 3, if the entry is not found in the address mapping table of the shared storage controller 1, send the request to the shared storage controller 2. Then in the shared memory controller 2, a table lookup is performed according to the address in the request.

Step 4, if there is a hit in the address mapping table of the shared memory controller 2, address mapping is performed according to the hit entry, and the shared memory controller 2 initiates access to the hard disk 2 according to the mapped physical address, and then the hard disk 2 is accessed. The response is forwarded to shared storage controller 1, and finally shared storage controller 1 returns the result to computing node 1.

Step 5, if there is no hit in the address mapping table of the shared memory controller 2, then repeat steps 3 and 4 in the shared memory controller 3 and the shared memory controller 4 in turn.

Step 6, if the request has no hit in the address mapping tables of the four shared memory controllers, a new DTE is established for the request in the shared memory controller 1 . And complete the address mapping according to the newly created DTE, and then initiate access to the hard disk 1 according to the mapped physical address. Finally, the response from hard disk 1 is forwarded to computing node 1.

Step 7, if there is no block that can be allocated in the shared storage controller 1, then in the shared storage controllers 2, 3, and 4 in turn, query whether there is a block that can be allocated, and check in the shared storage controller that has a block that can be allocated , create a new DTE, complete address mapping, and initiate access to the corresponding hard disk. Then the corresponding shared storage controller returns the response of its hard disk to shared controller 1, and finally shared controller 1 returns the result to computing node 1.

In the above solution, the shared storage controller based on dynamic address mapping independently establishes and maintains the address mapping relationship table throughout the process, and only allocates physical address space for virtual addresses with real access, avoiding waste of storage resources.

As shown in Figure 8, it is a shared memory controller according to the present invention, including:

a first receiving unit, configured to receive a computing node's request for data;

A parsing unit, configured to parse the request, and parse out the virtual address carried in the request;

a mapping unit, configured to perform address mapping according to an address mapping relationship table, and obtain a physical address corresponding to the virtual address;

The access unit is used for initiating hard disk access according to the mapped physical address;

The second receiving unit is configured to receive the response result of the hard disk, and forward the response result to the computing node that initiates the request.

The device of this embodiment can be used to implement the technical solution of the above method embodiment, and its implementation principle and technical effect are similar, and will not be repeated here.

As shown in Figure 6, it is a processing system of a shared storage controller according to the present invention, including:

At least one shared storage controller; a hard disk and at least one computing node directly connected to each shared storage controller;

The shared storage controller is configured to receive a request from a computing node for data; parse the request to obtain a virtual address carried in the request; perform address mapping according to the address mapping table to obtain the virtual address A physical address corresponding to the address; initiating hard disk access according to the mapped physical address; receiving a response result from the hard disk, and forwarding the response result to the computing node that initiated the request.

As shown in FIG. 9, the present invention also provides a processing device for a shared memory controller, including:

memory;

and a processor coupled to the memory, the processor configured to perform the method based on instructions stored in the memory.

The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are executed by a processor to execute the method.

Those of ordinary skill in the art can understand that all or part of the processes in the above method embodiments can be completed by instructing related hardware through computer programs. The program can be stored in a computer-readable storage medium. The program During execution, it may include the processes of the embodiments of the above-mentioned methods. Wherein, the storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM) or a random access memory (Random Access Memory, RAM) and the like.

The above is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Anyone skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present invention. All should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims (11)

1. A processing method of a shared memory controller, comprising: Using the address of the hard disk accessed by each computing node as a virtual address, and using the address mapped from the virtual address as a physical address; dividing the space of the virtual address and the space of the physical address into blocks of a predetermined size; An address mapping relationship table is established in the shared storage controller, and the address mapping relationship table is used to record the mapping relationship between the virtual address and the physical address in the local hard disk of the shared storage controller; The shared storage controller receives the computing node's request for data; The shared memory controller parses the request, and parses out the virtual address carried in the request; The shared memory controller performs address mapping according to the address mapping relationship table, and obtains a physical address corresponding to the virtual address; The shared storage controller initiates hard disk access according to the mapped physical address; The shared storage controller receives the response result of the hard disk, and forwards the response result to the computing node that initiated the request. 2. The method according to claim 1, wherein the address mapping table includes a hard disk entry DTE; The data structure of the hard disk entry DTE includes: PR, used to identify whether the hard disk entry is valid, when the bit is 1, it means that the entry is valid, and when it is 0, it means that the entry is invalid; S: It is a shared flag bit. When the bit is 1, it means that the virtual address is shared by multiple computing nodes. When it is 0, it means that it is a private address; The computing node identification bit is used to identify the identification of the computing node that initiates the request; The virtual address bit is a bit identifying a virtual address accessed by the computing node; The physical address bit is a bit that identifies the physical address mapped by the shared memory controller. 3. The method according to claim 2, wherein the shared memory controller performs address mapping according to an address mapping table, and obtaining the physical address corresponding to the virtual address comprises: When the first hard disk entry is found in the address mapping table, the identifier of the computing node in the first hard disk entry is consistent with the identifier of the computing node that initiated the request, and the virtual The address is the same as the virtual address in the request, then converting the virtual address of the request address into the physical address recorded in the first hard disk entry, and completing the address mapping; When not finding the first hard disk entry in the address mapping table, then create a second hard disk entry in the address mapping table; the Req ID of the second hard disk entry is requested The computing node, the virtual address of the second hard disk entry is the virtual address carried in the request, the physical address of the second hard disk entry is the first address of the next unallocated physical address block of the hard disk, and Complete address mapping. 4. The method of claim 1, wherein, The shared storage controller parses the request, and the step of parsing the virtual address carried in the request command includes: when the data requested by the computing node spans at least two blocks, splitting the request into At least two sub-requests, each sub-request corresponds to the access of a block; respectively resolve the virtual addresses carried in the sub-requests; The shared storage controller receiving the response result of the hard disk, and forwarding the response result to the address of the computing node that initiated the request includes: the shared storage controller receiving at least two response results of the hard disk, Combining the responses returned by at least two accesses to the hard disk into one response, and returning the combined response to the computing node. 5. The method of claim 1, wherein, Before the step of the shared storage controller receiving the data request from the computing node, the method further includes: The shared memory controller is pre-set with an address mapping relationship table cache space, and the address mapping relationship table cache space stores hard disk entries; The shared memory controller performs address mapping according to the virtual address, and obtaining the physical address corresponding to the virtual address further includes: Searching for a first hard disk entry in the address mapping relationship table and the address mapping relationship table cache space respectively; When the first hard disk entry is found in the address mapping table, perform address mapping and stop searching in the address mapping table. 6. The method of claim 1, wherein, Before the step of the shared storage controller receiving the data request from the computing node, the method further includes: Pre-configure the start address and space size of the shared virtual address space; The shared memory controller performs address mapping according to the address mapping relationship table, and obtaining the physical address corresponding to the virtual address includes: When the virtual address in the second hard disk entry is the same as the virtual address in the request, and the S bit of the second hard disk entry indicates that the virtual address is in the shared virtual address space range, use the request address The virtual address of the virtual address is converted into the physical address recorded in the second hard disk entry, and the address mapping is completed. 7. The method according to claim 1, wherein the step of receiving the data request from the computing node by the shared storage controller further comprises: When the request is a write request, and the virtual address of the request is within the range of the shared virtual address space, the shared storage controller judges whether it is a request initiated by the main control computing node, and generates a judgment result; When the judgment result is yes, then perform the next step; Otherwise, the request is discarded. 8. A shared storage controller, characterized in that, comprising: The first receiving unit is configured to use the address of the hard disk accessed by each computing node as a virtual address, use the address mapped from the virtual address as a physical address, and use a predetermined size between the space of the virtual address and the space of the physical address The block is divided into units, and after the address mapping table for recording the mapping relationship between the virtual address and the physical address in the local hard disk of the shared storage controller is established in the shared storage controller, the receiving computing node's data ask; A parsing unit, configured to parse the request, and parse out the virtual address carried in the request; a mapping unit, configured to perform address mapping according to an address mapping relationship table, and obtain a physical address corresponding to the virtual address; The access unit is used for initiating hard disk access according to the mapped physical address; The second receiving unit is configured to receive the response result of the hard disk, and forward the response result to the computing node that initiates the request. 9. A processing system for a shared memory controller, comprising: At least one shared storage controller; a hard disk and at least one computing node directly connected to each shared storage controller; The shared storage controller is configured to use the hard disk address accessed by each computing node as a virtual address, and use the address mapped from the virtual address as a physical address; use the virtual address space and the physical address space in a predetermined size The block is divided into units; an address mapping table is established in the shared storage controller, and the address mapping table is used to record the mapping relationship between the virtual address and the physical address in the local hard disk of the shared storage controller; receiving calculation A request from a node for data; parsing the request to obtain the virtual address carried in the request; performing address mapping according to the address mapping table to obtain the physical address corresponding to the virtual address; according to the mapped physical address , initiating hard disk access; receiving a response result from the hard disk, and forwarding the response result to the computing node that initiated the request. 10. A processing device for sharing a memory controller, comprising: memory; and a processor coupled to the memory, the processor configured to perform the method according to any one of claims 1 to 7 based on instructions stored in the memory. 11. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images