Parallel read-write method, product, medium and equipment of an automatic storage management system

Abstract

The present application relates to database technology, in particular to a parallel read-write method, product, medium and equipment of an automatic storage management system. The parallel read-write method comprises: each node of a cluster database generates a respective node historical snapshot and enters a waiting state; a pre-designated coordinator node monitors the state of each node and, in response to all nodes having entered the waiting state, notifies all nodes to enter a snapshot completion state; the coordinator node forms a global historical snapshot of the cluster database based on the node historical snapshots of all nodes, so that each node performs a logical decoding operation based on the global historical snapshot. The present application uses a bitmap array in the allocation unit to lock the data in the allocation unit space in units of data pages, thereby achieving fine-grained locking and avoiding lock contention conflicts caused by locking the entire allocation unit, and thus improving the parallel read-write efficiency of the automatic storage management system.

Description

Technical Field

[0001] This invention relates to database technology, and in particular to a parallel read / write method, product, medium, and device for an automatic storage management system. Background Technology

[0002] Automatic storage management system is a user-space distributed file system specifically for databases. It provides file system and disk management capabilities, which can help users manage databases more conveniently, especially shared storage database files.

[0003] Currently, automated storage and retrieval systems (AS / RS) typically employ striping technology to distribute data across multiple disks and logically divide storage space using allocation units as the basic unit of striping to achieve parallel I / O. Furthermore, AS / RS uses allocation units as the granularity for locking control; during read operations, each allocation unit independently acquires a shared lock to achieve parallel reading, while during write operations, an exclusive lock is applied to the entire allocation unit.

[0004] However, the current solution uses the allocation unit as the granularity for locking control, which serializes write operations for the same allocation unit. Data areas belonging to the same allocation unit can only be written serially, resulting in difficulty in improving the parallel read and write efficiency of the system. Summary of the Invention

[0005] In view of the above problems, a parallel read / write method, product, medium and device for an automatic storage management system that overcomes or at least partially solves the above problems is proposed.

[0006] One object of the present invention is to provide a parallel read / write method for an automatic storage management system, so as to improve the parallel read / write efficiency of the automatic storage management system.

[0007] A further objective of this invention is to improve the utilization rate of storage space.

[0008] Another further objective of this invention is to improve the efficiency of locating lock status information.

[0009] Specifically, the present invention provides a parallel read / write method for an automatic storage management system, comprising: The automatic storage management system receives access requests for target data pages, and locks the data within the allocation unit on a page-by-page basis. Obtain the lock mode status of the target data page based on the access request; Update the lock mode state of the target data page according to the type of access request; Perform access operations on the target data page to read or write the target data page.

[0010] Optionally, the automatic storage management system locks data within an allocation unit on a page-by-page basis, including the following steps: The automatic storage management system uses a bitmap array to lock each data page within each allocation unit. The bitmap array includes multiple sets of bits, each set of bits corresponding to a data page within the allocation unit, and each set of bits is used to store the lock mode state of a corresponding data page.

[0011] Optionally, the automatic storage management system uses a bitmap array to lock individual data pages within each allocation unit, including the following steps: The automatic storage management system adds a bitmap array to the metadata of each allocation unit; Map each group of bits in each bitmap array to each data page in the allocation unit according to the address offset; Each data page within each allocation unit is locked, and the lock mode state of each data page is stored in a set of bits corresponding to it.

[0012] Optionally, the access request for the target data page includes the address of the target data page; and The steps to obtain the lock mode status of the target data page based on the access request include: The page index of the target data page within its respective allocation unit is determined based on the address of the target data page; Based on the page index, obtain a set of bits corresponding to the target data page to determine the lock mode state of the target data page.

[0013] Optionally, the step of obtaining a set of bits corresponding to the target data page based on the page index to obtain the lock mode state of the target data page includes: Calculate the offset of a set of bits in the bitmap array corresponding to the target data page based on the page index; Read a set of bits corresponding to the target data page based on the offset; The read bits are decoded to obtain the lock mode state of the target data page.

[0014] Optionally, the step of determining the page index of the target data page within its respective allocation unit based on the address of the target data page includes: The control information of the target data page and its corresponding allocation unit is obtained through the address of the target data page, wherein the control information of the allocation unit includes the starting address of the allocation unit; Obtain the page index of the target data page based on the address of the target data page and the starting address of the allocation unit.

[0015] Optionally, the step of updating the lock mode state of the target data page according to the type of access request includes: Determine the target lock mode state based on the type of access request; Determine if there is a conflict between the lock mode state of the target data page and the target lock mode state; If not, then based on the comparison and exchange instruction, the lock mode state of the target data page is updated to the target lock mode state to complete the locking of the target data page.

[0016] According to another aspect of the present invention, a computer program product is also provided, comprising a computer program that, when executed by a processor, implements the steps of any of the above-described parallel read / write methods.

[0017] According to another aspect of the present invention, a computer-readable storage medium is also provided, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the steps of any of the above-described parallel read / write methods.

[0018] According to another aspect of the present invention, a computer device is also provided, which includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of any of the above-described parallel read / write methods.

[0019] The parallel read / write method of the automatic storage management system of the present invention locks the data within the allocation unit on a data page-by-data page basis. After receiving an access request for a target data page, the lock mode state of the target data page is obtained according to the access request, the lock mode state of the target data page is updated according to the type of access request, and the access operation on the target data page is performed to read and write the target data page. This achieves fine-grained lock management, thereby ensuring fine-grained lock management and avoiding lock contention caused by locking the entire allocation unit as the unit, thus improving the parallel read / write efficiency of the automatic storage management system.

[0020] Furthermore, the parallel read / write method of the automatic storage management system of the present invention locks each data page in each allocation unit by using a bitmap array, so that multiple sets of bits in the bitmap array correspond one-to-one with multiple data pages in the allocation unit, and each set of bits is used to store the lock mode state of a corresponding data page, thereby achieving compact storage of lock mode states. This significantly reduces metadata overhead while ensuring fine-grained lock management, thereby improving the utilization rate of storage space.

[0021] Furthermore, the parallel read / write method of the automatic storage management system of the present invention adds a bitmap array to the metadata of the allocation unit, maps each group of bits to the data page according to the address offset, and stores the lock mode state of each data page in the corresponding group of bits, thereby realizing the precise association between the lock mode state and the data page, thus improving the execution efficiency of the locking operation, so as to facilitate the rapid location of the lock state information in the future, and thus improving the location efficiency of the lock state information.

[0022] The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments of the invention in conjunction with the accompanying drawings. Attached Figure Description

[0023] The following sections will describe some specific embodiments of the invention in detail by way of example and not limitation, with reference to the accompanying drawings. The same reference numerals in the drawings denote the same or similar parts or portions. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings: Figure 1 This is a schematic diagram of a parallel read / write method for an automatic storage management system according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the allocation unit and bitmap array in the parallel read / write method of an automatic storage management system according to an embodiment of the present invention; Figure 3 This is a control flowchart of a parallel read / write method for an automatic storage management system according to an embodiment of the present invention; Figure 4 This is a schematic diagram of a computer program product according to an embodiment of the present invention; Figure 5 This is a schematic diagram of a computer-readable storage medium according to an embodiment of the present invention; Figure 6 This is a schematic block diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0024] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0025] To address the aforementioned problems, this invention proposes a parallel read / write method for an automatic storage management system. Figure 1 This is a schematic diagram of a parallel read / write method for an automatic storage management system according to an embodiment of the present invention. Figure 1 As shown, the parallel read / write method of this embodiment generally includes: In step S102, the automatic storage management system receives an access request for the target data page. The automatic storage management system locks the data within the allocation unit on a page-by-page basis. It should be noted that the upper-layer application of the automatic storage management system is a database, which can access data for reading and writing on a page-by-page basis. In this embodiment, the automatic storage management system can define the locking granularity during reading and writing as a page, and use a separate read / write lock for each page.

[0026] Step S104: Obtain the lock mode status of the target data page based on the access request. In this embodiment, the lock mode status may include an unlocked state and a locked state. Additionally, the locked state may include one or more of shared locks, exclusive locks, and intention locks.

[0027] Step S106: Update the lock mode state of the target data page according to the type of access request. It should be noted that the access request type can include read operations and write operations. For read operations, a shared lock is required on the target page; for write operations, an exclusive lock is required on the target page.

[0028] Step S108 involves performing an access operation on the target data page to read or write the target data page. In other words, after updating the lock mode state of the target page, read or write operations can be performed on the target page.

[0029] It's important to note that an Allocation Unit (AU) is the basic striped unit in an Automatic Storage Management (AS / RS) system, logically partitioned from disk space. Furthermore, in AS / RS, the default size for each AU is 1MB, while the default size for each Page is 4KB, 8KB, or 16KB. This means that each AU can store multiple Pages, and different Pages within each AU can be accessed concurrently.

[0030] Therefore, the parallel read / write method of the automatic storage management system in this embodiment of the invention locks the data within the AU on a page-by-page basis. After receiving an access request for the target page, it first obtains the lock mode state of the target page according to the access request, then updates the lock mode state of the target page according to the type of access request, and performs the access operation on the target page to read and write the target page. This achieves fine-grained lock management, thereby ensuring fine-grained lock management and avoiding lock contention caused by locking the entire AU as the unit, thus improving the parallel read / write efficiency of the automatic storage management system.

[0031] Figure 2 This is a schematic diagram of the allocation unit and bitmap array in a parallel read / write method of an automatic storage management system according to an embodiment of the present invention. Figure 2 As shown, each AU in the automatic storage management system of the present invention may include multiple Pages, and the lock mode states of the multiple Pages are stored in multiple sets of bits in a bitmap array.

[0032] Based on this, the step of locking the data in the AU by the automatic storage management system in step S102 above may include the following steps: the automatic storage management system uses a bitmap array to lock each page in each AU.

[0033] like Figure 2 As shown, the bitmap array can include multiple sets of bits. Each set of bits corresponds one-to-one with multiple pages within the AU, and each set of bits stores the lock mode state of its corresponding page. Furthermore, the entire bitmap array can be represented using uint64 atomic variables, and each set of bits can be encoded to store the lock mode state of its corresponding page.

[0034] For example, each group of bits can include 1 bit, and the encoding of each group of bits can be 0 or 1, where 0 can represent no lock and 1 can represent a locked state. Therefore, by adding a bitmap array to the metadata of the AU and using 1 bit of lock mode encoding, only 2 uint64 spaces need to be added to the metadata of each AU.

[0035] For example, each group of bits can include 2 bits, and the encoding of each group of bits can be 00, 01, 10, and 11. 00 can represent no lock, 01 can represent a shared lock (S lock), 10 can represent an exclusive lock (X lock), and 11 can represent an intention lock. Therefore, by adding a bitmap array to the AU's metadata and using 2-bit lock mode encoding, only 4 uint64 spaces need to be added to the metadata of each AU.

[0036] Therefore, the parallel read / write method of the automatic storage management system in this embodiment of the invention locks each Page in each AU by using a bitmap array, so that multiple sets of bits in the bitmap array correspond one-to-one with multiple Pages in the AU, and each set of bits is used to store the lock mode state of a Page corresponding to it, thereby achieving space compression and compact storage of lock mode states. While ensuring fine-grained lock management, it significantly reduces metadata overhead and thus improves the utilization rate of storage space.

[0037] In some embodiments, the step of the automatic storage management system using a bitmap array to lock each Page within each AU may include the following steps: the automatic storage management system adds a bitmap array to the metadata of each AU; maps each group of bits in each bitmap array to each Page in the AU according to the address offset; locks each Page within each AU, and stores the lock mode state of each Page in a corresponding group of bits.

[0038] In other words, the automatic storage management system of the present invention will no longer use AU or stripe as the unit to control the parallel reading and writing of file data. Instead, it will add a bitmap array to the metadata of the AU to lock the AU space in units of pages. That is, the lock mode is compressed into several bits to facilitate the fast marking, traversal and conflict detection of page-level locks.

[0039] Therefore, the parallel read / write method of the automatic storage management system of the present invention, by adding a bitmap array to the metadata of the AU, maps each group of bits to the Page according to the address offset, and stores the lock mode state of each Page in the corresponding group of bits, realizes the precise association between the lock mode state and the Page, thereby improving the execution efficiency of the locking operation, so as to facilitate the rapid location of the lock state information in the future, and thus improves the location efficiency of the lock state information.

[0040] In some embodiments, the access request for the target Page may include the address of the target Page. Based on this, step S104 may include the following steps: determining the page index of the target Page within its respective AU based on the address of the target Page; and obtaining a set of bits corresponding to the target Page based on the page index to obtain the lock mode state of the target Page.

[0041] In other words, the automatic storage management system of the present invention can obtain the page index in the AU through the page address information, and obtain the 1-2 bits corresponding to the target Page in the bitmap array based on the page index, so as to obtain the lock mode state of the target Page according to the encoding of the 1-2 bits.

[0042] Therefore, the parallel read / write method of the automatic storage management system in this embodiment of the invention determines the page index in the AU based on the address of the target Page and obtains the corresponding lock mode state based on the index, thereby achieving rapid location of lock state information, reducing the time overhead of lock query, and thus improving the response speed of the system.

[0043] In some embodiments, the step of determining the page index of the target page within its AU based on the address of the target page may include the following steps: obtaining control information of the AU corresponding to the target page through the address of the target page, wherein the control information of the AU may include the starting address of the AU; and obtaining the page index of the target page based on the address of the target page and the starting address of the AU.

[0044] Specifically, the steps for obtaining the page index of the target page based on the address of the target page and the starting address of the AU can be implemented as follows: calculate the difference between the address of the target page and the starting address of the AU; use this difference to perform a modulo operation on the page size to obtain the page index.

[0045] Therefore, the parallel read / write method of the automatic storage management system in this embodiment of the invention calculates the page index by using the target Page address and the AU starting address, thereby achieving precise positioning of the Page within the AU, providing an accurate index basis for subsequent lock operations, and thus ensuring the reliability of the entire concurrency control process.

[0046] In some embodiments, the step of obtaining a set of bits corresponding to the target Page based on the page index to obtain the lock mode state of the target Page may include the following steps: calculating the offset of a set of bits corresponding to the target Page in the bitmap array based on the page index; reading a set of bits corresponding to the target Page based on the offset; and decoding the read bits to obtain the lock mode state of the target Page.

[0047] In this embodiment, the step of calculating the offset of a set of bits corresponding to the target Page in the bitmap array based on the page index may include the following steps: obtaining the offset of the bit corresponding to the target Page in the bitmap array based on the page index and the width of the bits in the bitmap array.

[0048] Specifically, the automatic storage management system of the present invention can calculate the product between the page index and the width of each group of bits as the offset of the group of bits in the bitmap array. It can also perform a bitwise AND operation using the offset and the width of each group of bits to obtain the current lock mode state of the target page.

[0049] In addition, the above steps of decoding the read bits to obtain the lock mode state of the target page may include the following steps: performing a bitwise AND operation using the offset and the width of the bits to obtain the current lock mode state of the target page.

[0050] Therefore, the parallel read / write method of the automatic storage management system in this embodiment of the invention obtains the lock mode state by calculating the offset, reading the bits and decoding them, thereby accurately extracting the lock information of each page from the bitmap array, ensuring the accuracy of the lock state acquisition, and providing a reliable basis for subsequent concurrency control.

[0051] In some embodiments, step S106 may include the following steps: determining the target lock mode state based on the type of access request; determining whether there is a conflict between the lock mode state of the target Page and the target lock mode state; if not, updating the lock mode state of the target Page to the target lock mode state based on the comparison and exchange instruction, so as to complete the locking of the target Page.

[0052] It should be noted that lock mode states can include unlocked, shared lock, exclusive lock, and intention lock, while target lock mode states can include shared lock, exclusive lock, and intention lock. For example, there is a conflict between shared lock and exclusive lock, and an exclusive lock conflicts with any other lock.

[0053] In this embodiment, the Compare-And-Swap (CAS) instruction is a hardware-level atomic instruction used to achieve lock-free concurrency. Based on this, the step of updating the lock mode state of the target Page to the target lock mode state using the CAS instruction to complete the locking of the target Page can be specifically executed as follows: save the current lock mode state of the target Page as old and modify the corresponding bit state to new; use CAS to modify the bitmap information; determine whether the modification was successful; if yes, then determine that the locking of the target Page is complete; if not, return to step S104 above.

[0054] In addition, after the above steps of determining whether there is a conflict between the lock mode state of the target Page and the target lock mode state, the parallel read and write method of the automatic storage management system of the present invention may further include the following steps: if there is a conflict between the lock mode state of the target Page and the target lock mode state, wait and retry, and return to the above step S104.

[0055] Therefore, the parallel read / write method of the automatic storage management system in this embodiment of the invention achieves lock-free parallel writing by determining the target lock mode state according to the access request type and performing lock-free updates based on CAS instructions, thereby avoiding the computational overhead of traditional lock mechanisms and improving system operating efficiency.

[0056] Furthermore, the parallel read / write method of the automatic storage management system in this embodiment of the invention achieves efficient conflict detection by using CAS instructions to operate on the bitmap when updating the lock mode encoding. Since this is an atomic operation, a successful operation indicates no conflict, thus achieving efficient conflict detection.

[0057] Following step S108 above, the parallel read / write method of the automatic storage management system of the present invention may further include the following step: unlocking the target Page. Specifically, the step of unlocking the target Page may be executed by updating the lock mode state of the target Page to the target lock mode state based on a CAS instruction, thereby completing the unlocking of the target Page.

[0058] In other words, both unlocking and locking the target page involve updating the bitmap array. When unlocking the target page, the first 1-2 bits can be cleared to update the target page's lock mode state to an unlocked state.

[0059] Figure 3 This is a control flowchart of a parallel read / write method for an automatic storage management system according to an embodiment of the present invention. The following is in conjunction with... Figure 3 The process steps of this embodiment will be described in detail below.

[0060] In step S302, the automatic storage management system receives an access request for the target page. It should be noted that the access request for the target page may include the address of the target page.

[0061] Step S304: Obtain the control information of the AU corresponding to the target Page through its address. It should be noted that the AU's control information may include the AU's starting address.

[0062] Step S306: Calculate the difference between the address of the target Page and the starting address of the AU.

[0063] Step S308: Perform a modulo operation on the Page size using the difference to obtain the page index.

[0064] Step S310: Based on the page index and the width of the bits in the bitmap array, obtain the offset of the bit corresponding to the target Page in the bitmap array.

[0065] Step S312: Read a set of bits corresponding to the target Page based on the offset.

[0066] Step S314: Perform a bitwise AND operation using the offset and the width of the bit to obtain the current lock mode state of the target Page.

[0067] Step S316: Determine the target lock mode state based on the type of access request.

[0068] Step S318: Determine whether there is a conflict between the lock mode state of the target Page and the target lock mode state. If yes, return to step S304; otherwise, execute step S320.

[0069] Step S320: Save the current lock mode state of the target Page as old, and modify the corresponding bit state to new.

[0070] Step S322: Determine whether the modification was successful. If yes, proceed to step S324; otherwise, return to step S304.

[0071] Step S324: Perform access operations on the target Page to read and write the target Page.

[0072] Step S326: Unlock the target Page. This completes the parallel read / write process of the automatic storage management system, and this process ends.

[0073] Therefore, the parallel read / write method of the automatic storage management system in this embodiment of the invention locks the Page within the AU by using a bitmap array, that is, compressing the lock mode state into several bits and attaching it to the metadata of the AU, and updating the lock mode state through CAS instructions. This achieves fine-grained lock granularity and lock-free concurrent read / write operations, thus taking into account the goals of space compression, lock-free parallel writing and conflict detection. It can support a large number of concurrent read / write operations and provides the concurrency performance of the automatic storage management system.

[0074] This embodiment also provides a computer program product 10, a computer-readable storage medium 20, and a computer device 30. Figure 4 This is a schematic diagram of a computer program product 10 according to an embodiment of the present invention. Figure 5 This is a schematic diagram of a computer-readable storage medium 20 according to an embodiment of the present invention. Figure 6 This is a schematic block diagram of a computer device 30 according to an embodiment of the present invention.

[0075] Computer program product 10 includes computer program 11, which, when executed by processor 32, implements the steps of any of the above-described parallel read / write methods. Computer-readable storage medium 20 stores the computer program 11 thereon, which, when executed by processor 32, implements the steps of any of the above-described parallel read / write methods. Computer device 30 may include memory 31, processor 32, and computer program 11 stored in memory 31 and running on processor 32.

[0076] The computer program 11 used to perform the operations of the present invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, status setting data, integrated circuit configuration data, or source code or object code written in any combination of one or more programming languages ​​and procedural programming languages.

[0077] Computer program 11 may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer may be connected to the user's computer via any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, to perform aspects of the invention, electronic circuitry including, for example, programmable logic circuitry, a Field-Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA) may execute computer-readable program instructions to personalize the electronic circuitry by utilizing state information of the computer-readable program instructions.

[0078] For the purposes of this embodiment, computer program product 10 is a related product that includes computer program 11.

[0079] For the purposes of this embodiment, computer-readable storage medium 20 is a tangible device capable of holding and storing computer program 11. It can be any device capable of containing, storing, communicating, propagating, or transmitting program 811 for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable storage medium 20 include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable optical disc read-only memory (CD-ROM), digital versatile disc (DVD), memory stick, floppy disk, mechanical encoding device, and any suitable combination thereof.

[0080] Therefore, those skilled in the art should recognize that although numerous exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications conforming to the principles of the present invention can be directly determined or derived from the disclosure of the present invention without departing from the spirit and scope of the invention. Thus, the scope of the present invention should be understood and construed as covering all such other variations or modifications.

Claims

1. A parallel read / write method for an automatic storage management system, comprising: The automatic storage management system receives access requests for target data pages, wherein the automatic storage management system locks the data within the allocation unit on a per-data-page basis; Obtain the lock mode status of the target data page based on the access request; Update the lock mode state of the target data page according to the type of the access request; Perform an access operation on the target data page to read and write the target data page.

2. The parallel read / write method of the automatic storage management system according to claim 1, wherein, The automatic storage management system locks data within an allocation unit on a page-by-page basis, including the following steps: The automatic storage management system uses a bitmap array to lock each data page within each allocation unit. The bitmap array includes multiple sets of bits, each set of bits corresponding one-to-one with a plurality of data pages within the allocation unit, and each set of bits is used to store the lock mode state of a corresponding data page.

3. The parallel read / write method for the automatic storage management system according to claim 2, wherein, The automatic storage management system uses a bitmap array to lock each data page within each allocation unit, including the following steps: The automatic storage management system adds the bitmap array to the metadata of each allocation unit; Each group of bits in each bitmap array is mapped to each data page in the allocation unit according to the address offset; Each data page within each allocation unit is locked, and the lock mode state of each data page is stored in a corresponding set of bits.

4. The parallel read / write method of the automatic storage management system according to claim 2, wherein, The access request for the target data page includes the address of the target data page; and The step of obtaining the lock mode state of the target data page according to the access request includes: The page index of the target data page within its respective allocation unit is determined based on the address of the target data page; Based on the page index, a set of bits corresponding to the target data page is obtained to determine the lock mode state of the target data page.

5. The parallel read / write method of the automatic storage management system according to claim 4, wherein, The step of obtaining a set of bits corresponding to the target data page based on the page index to obtain the lock mode state of the target data page includes: Calculate the offset of a set of bits corresponding to the target data page in the bitmap array based on the page index; Read a set of bits corresponding to the target data page based on the offset; The read bits are decoded to obtain the lock mode state of the target data page.

6. The parallel read / write method of the automatic storage management system according to claim 4, wherein, The step of determining the page index of the target data page within its respective allocation unit based on the address of the target data page includes: The control information of the allocation unit corresponding to the target data page is obtained through the address of the target data page, wherein the control information of the allocation unit includes the starting address of the allocation unit; Based on the address of the target data page and the starting address of the allocation unit, obtain the page index of the target data page.

7. The parallel read / write method of the automatic storage management system according to claim 1, wherein, The step of updating the lock mode state of the target data page according to the type of the access request includes: Determine the target lock mode state based on the type of the access request; Determine whether there is a conflict between the lock mode state of the target data page and the target lock mode state; If not, then based on the comparison and exchange instruction, the lock mode state of the target data page is updated to the target lock mode state to complete the locking of the target data page.

8. A computer program product comprising a computer program, wherein, When the computer program is executed by the processor, it implements the steps of the parallel read / write method of the automatic storage management system according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the parallel read / write method of the automatic storage management system according to any one of claims 1 to 7.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory, wherein, The processor executes the computer program to implement the steps of the parallel read / write method of the automatic storage management system according to any one of claims 1 to 7.

Images