TWI768476B - Method and apparatus for performing mapping information management regarding redundant array of independent disks, and associated storage system - Google Patents

Abstract

A method and apparatus for performing mapping information management regarding a RAID are provided. The method includes: writing data into a data region of the RAID in a Redirect-on-Write (ROW) manner, and recording mapping information between logical addresses of the data and protected-access-unit addresses (p-addresses) of protected access units in the data region into a logical-address-to-p-address (L2p) table within a table region of the RAID; when partial data of the data is updated, maintaining an updating list including a set of L2p table entries for the partial data in a RAM, and maintaining a recovery log corresponding to the updating list in a log region of the RAID, for power failure recovery; and according to the updating list, detecting whether a number of same-location L2p table entries in the set of L2p table entries reaches a predetermined threshold, to selectively updating the L2p table with the same-location L2p table entries.

Description

Method and apparatus for managing mapping information about fault-tolerant disk arrays and storage system

The present invention relates to a storage system, and more particularly, to a method and an apparatus for managing mapping information about a Redundant Array of Independent Disks (RAID for short).

According to the related art, a fault-tolerant disk array (Redundant Array of Independent Disks, may be referred to as RAID) can be implemented in a server. In particular, through various types of RAID solutions, data can be protected at a corresponding level. For example, for data backup purposes, the server may be designed with a Copy-on-Write (COW) architecture. Due to the nature of the COW architecture, the performance of this server may degrade over time. As another example, the server may be designed with a Redirect-on-Write (ROW) architecture to avoid the problem of performance degradation over time. However, certain problems, such as side effects, may occur. Therefore, there is a need for a novel method and related architecture to ensure that the storage system can operate correctly under various conditions, respectively.

An object of the present invention is to provide a method for managing mapping information about a fault-tolerant disk array (Redundant Array of Independent Disks, may be referred to as RAID). and device to solve the above problems.

Another object of the present invention is to provide a method and apparatus for managing mapping information about a RAID to ensure that the storage system can operate correctly under various conditions, respectively.

Another object of the present invention is to provide a method and apparatus for managing mapping information about a RAID, so as to solve the related technical problems without any side effects or less likely to introduce side effects.

At least one embodiment of the present invention provides a method for managing mapping information about a RAID. The method may include: writing data into a data area of the RAID by way of Redirect-on-Write (ROW for short), and recording the logical address of the data and a plurality of protections in the data area Mapping information between the respective protected-access-unit addresses (p-addresses, p-addresses) of the protected access units to a table area ( A logical address pair protection access unit address table (logical-address-to-p-address (L2p) table, L2p table for short) in the table region), in which subsets of the data are written into the Protect the access unit; when the local data of the data is updated, maintain an update list for the local data including a set of logical address pairs to protect the access unit address table entry (L2p table entry, may be referred to as L2p table entry) in In a random access memory (Random Access Memory, RAM for short), a recovery log corresponding to the update list is maintained in a log area of the RAID for use in power failure recovery; and according to the update list, the detection testing whether the number of at least two co-located L2p table entries in the set of L2p table entries reaches a predetermined threshold, to selectively update the L2p table with the at least two co-located L2p table entries, wherein the at least two A co-located L2p table entry represents an L2p table entry in the set of L2p table entries that should be stored in the same protection access unit in the table area.

According to some embodiments, the present invention further provides a storage system operating according to the above method, wherein the storage system includes the RAID.

At least one embodiment of the present invention provides a method for mapping information about a RAID management device. The apparatus may include a processing circuit, wherein the processing circuit is located in a storage system and is used to control the operation of the storage system. For example, the operations of the storage system may include: writing data into a data area of the RAID in a ROW manner, and recording the logical address of the data and the respective protection of a plurality of protection access units in the data area Mapping information between access unit addresses to an L2p table in a table area of the RAID, where subsets of the data are written into the protected access units respectively; when the partial data of the data is updated, it is The local data maintenance includes an update list of a set of L2p table entries in a RAM, and maintains a recovery log corresponding to the update list in a log area of the RAID for use in power failure recovery; and according to the update list , detecting whether the number of at least two co-located L2p table entries in the set of L2p table entries reaches a predetermined threshold, to selectively update the L2p table with the at least two co-located L2p table entries, wherein the at least two co-located L2p table entries are The two co-located L2p table entries represent the L2p table entries in the set of L2p table entries that should be stored in the same protection access unit in the table area.

The method and apparatus of the present invention can ensure that the storage system can operate normally under various conditions, respectively. In addition, the method and apparatus of the present invention provide a powerful mapping information management mechanism. As a result, various goals such as performance optimization, high security, budget control, etc. can be achieved. In addition, the method and apparatus of the present invention can solve the related technical problems in a manner that does not introduce any side effects or is less likely to introduce side effects.

10: User device

11,111: Processing Circuits

12, 112, 122: Interface Circuits

100: Storage System

110: Busbar

111P: Code

121: Random Access Memory

121T: Management table

130~134: Storage device

200: Method

210, 220, 230: Steps

310, 320, 330: Protected Blocks

410: L2p table

510, 610: Update list

520, 620: Recovery log

RAID: Fault Tolerant Disk Array

DR: data area

TR: table area

LR: log area

D: information

P: check code

S1~S1023: Auxiliary information

FIG. 1 is a schematic diagram of a storage system and a user device according to an embodiment of the present invention.

FIG. 2 is a flowchart of a method for performing mapping information management with respect to a RAID (such as that shown in FIG. 1 ) according to an embodiment of the present invention.

FIG. 3 illustrates a plurality of protected access units involved in the method shown in FIG. 2 in an embodiment, wherein an example of the plurality of protected access units may be a protected block.

FIG. 4 shows the L2p table involved in the method shown in FIG. 2 in one embodiment.

FIG. 5 shows an update list and a recovery log involved in the method shown in FIG. 2 in an embodiment.

FIG. 6 shows an update list and a recovery log involved in the method shown in FIG. 2 in another embodiment.

FIG. 1 is a schematic diagram of a storage system 100 and a user device 10 according to an embodiment of the present invention. The user device 10 may include a processing circuit 11 (eg, at least one processor and related circuits), and may further include an interface circuit 12 coupled to the processing circuit 11 and a storage device. The storage system 100 may include a processing circuit 111 (eg, at least one processor and related circuits), and may further include interface circuits 112 and 122 coupled to the processing circuit 111 through the bus bar 110 and a random access memory (Random Access Memory, may be referred to as RAM) 121, and storage devices {130, 131, 132, 133, 134} (such as hard disk drives, and/or solid state disks) can be installed in the storage system 100 through the interface circuit 122 to form a RAID in which the programs executed on the processing circuit 111 The code 111P can be read from the storage device 130 (eg, the system disk), and can maintain (eg, create, store, and/or update) the management table 121T in the RAM 121 to perform operations related to the method of the present invention to manage the data area DR, table area TR and log area LR. The interface circuits 12 and 112 may be implemented as wired network interfaces and/or wireless network interfaces to allow the storage system 100 and the user device 10 to exchange information. The user can access (read or write) user data in the storage system 100 through the user device 10 . Examples of user device 10 may include, but are not limited to, a multifunction mobile phone, a tablet computer, a wearable device, and personal computers (such as desktop computers and notebook computers). Examples of storage system 100 may include, but are not limited to, a server such as a storage server. According to some embodiments, the architecture of the storage system 100 may vary. For example, the code 111P can be implemented through dedicated hardware provided in the interface circuit 122 to perform the related operations of the method of the present invention.

FIG. 2 is a flowchart of a method 200 for managing mapping information with respect to a RAID (such as the RAID shown in FIG. 1 ) according to an embodiment of the present invention. The method 200 can be applied to that shown in FIG. 1 The storage system 100, the processing circuit 111 for executing the code 111P, and related components.

In step 210 , the storage system 100 (eg, the processing circuit 111 ) writes data into the data area DR of the RAID in a Redirect-on-Write (ROW) manner, and records the logical address of the data and the mapping between the respective protected-access-unit addresses (p-address, may be referred to as p-address) of a plurality of protected access units (protected access units) in the data area DR information to a logical-address-to-p-address (L2p) table in the table area TR of the RAID to protect the access unit address table (logical-address-to-p-address (L2p) table, may be referred to as L2p table), wherein the subsets of the data are respectively are written to the protected access units.

In step 220, when the local data of the data is updated, the storage system 100 (eg, the processing circuit 111) maintains a set of logical address pairs to protect the access unit address table entry (L2p table entry, may be referred to for short) for the local data An update list of L2p table entries) is stored in the RAM 121, and a recovery log corresponding to the update list is maintained in the log area LR of the RAID for power failure recovery.

In step 230, according to the update list, the storage system 100 (eg, the processing circuit 111) detects whether the number of at least two co-located L2p table entries in the set of L2p table entries reaches a predetermined threshold (eg, 2, or other predetermined values) to selectively update the L2p table with the at least two co-located L2p table entries, wherein the at least two co-located L2p table entries represent that the set of L2p table entries should be stored in the table region TR L2p table entry of the same protection access unit. For example, the L2p table entry may be an updated version used to update the L2p table, and may contain the latest storage location corresponding to the ROW operation.

According to the present embodiment, in step 220, when the partial data of the data is updated, the storage system 100 (eg, the processing circuit 111) can correspondingly record auxiliary information in the update list for detecting the L2p table entries in the set Whether the at least two co-located L2p table entries exist. For example, the auxiliary information may indicate a set of restore log entries in the restore log, and the set of restore log entries may point to a set of protected access units in the table region TR to direct the L2p table to Updated.

According to some embodiments, the update list may contain one of the L2p table entries respectively corresponding to the set A set of pointers, the set of pointers may respectively point to a set of restoration log entries in the restoration log, and the set of restoration log entries may point to a set of protection access units in the table region TR. For example, the L2p table may contain at least one sub-table that should be updated by the set of L2p table entries, and the position of the at least one sub-table in the table region TR may correspond to the set of protected access units. As another example, the L2p table may contain at least one sub-table that should be updated by the set of L2p table entries. In addition, the update list may include at least two pointers respectively corresponding to the at least two co-located L2p table entries, the at least two pointers may respectively point to the same recovery log entry in the recovery log, and the same recovery log entry The recovery log entry may point to the same protection access unit in the table region TR.

FIG. 3 shows a plurality of protected access units involved in the method 200 shown in FIG. 2 in an embodiment, wherein the example of the plurality of protected access units can be the protected blocks 310, 320 and 330, but this The invention is not limited to this. For example, for any one of the protection blocks 310, 320, and 330, the symbol "D" may represent the data in the protection block, and the symbol "P" may represent the data in the protection block Check code. Through the check code P, the data D can be protected. According to some embodiments, the type and/or level of protection of the RAID may be varied, wherein user data may be protected by the corresponding type and/or level.

FIG. 4 illustrates the L2p table 410 involved in the method 200 shown in FIG. 2 in one embodiment (which can be used as an example of the L2p table). The L2p table 410 may include a plurality of L2p sub-tables, wherein a first row (row) of the L2p sub-tables may respectively map pages 0-511 (in particular, logical addresses 0-511) to their respective storage locations (eg some protected access units such as protected blocks), the second column of L2p sub-tables can map pages 512-1023 (in particular, logical addresses 512-1023) to their respective storage locations (eg: some A protected access unit such as a protected block), but the present invention is not limited thereto. According to some embodiments, these storage locations may be considered ROW locations.

FIG. 5 shows the update list 510 and the recovery log 520 involved in the method 200 shown in FIG. 2 in an embodiment, wherein the auxiliary information {S4, S1021, S1022, S1023, S1, S2, ...} can be indicator. The contents of this embodiment similar to those of the foregoing embodiments are not repeated here.

FIG. 6 shows the update list 610 and the recovery log 620 involved in the method 200 shown in FIG. 2 in another embodiment, wherein the auxiliary information {S8, S3, S5, S1, S6, S1, ...} can be for the indicator. For example, in step 230, when the auxiliary information S1 appears twice, the storage system 100 (eg, the processing circuit 111 ) detects that the at least two co-located L2p table entries (such as the two in FIG. 6 contain the auxiliary information S1 ) If the number of L2p table entries) reaches the predetermined threshold (eg: 2), the L2p table is updated with the at least two co-located L2p table entries. Therefore, the storage system 100 updates the L2p table in advance, and does not need to wait until the recovery log 620 is full before updating the L2p table. Therefore, the device of the present invention can avoid the overall performance drop caused by the operation of updating the L2p table greatly. . Therefore, the method and apparatus of the present invention can ensure that the storage system 100 can operate normally under various conditions, respectively. In addition, the method and apparatus of the present invention provide a powerful mapping information management mechanism, which can achieve various goals, such as performance optimization, high security, budget control, and the like. In addition, the method and apparatus of the present invention can solve the related technical problems in a manner that does not introduce any side effects or is less likely to introduce side effects.

The above descriptions are only preferred embodiments of the present invention, and all equivalent changes and modifications made according to the scope of the patent application of the present invention shall fall within the scope of the present invention.

10: User device

11,111: Processing Circuits

12, 112, 122: Interface Circuits

100: Storage System

110: Busbar

111P: Code

121: Random Access Memory

121T: Management table

130~134: Storage device

RAID: Fault Tolerant Disk Array

DR: data area

TR: table area

LR: log area

Claims (7)

A method for managing mapping information about a fault-tolerant disk array (Redundant Array of Independent Disks, RAID), the method comprising: writing data into a data of the fault-tolerant disk array by means of write redirection area, so that the data to be written in the writing process is written to at least one access unit of a plurality of access units in the data area, and records which are respectively written in the data mapping information between logical addresses of data to the plurality of access units and respective access-unit addresses of the plurality of access units to a table area of the fault-tolerant disk array a logical-address-to-access-unit-address table in a (table region) where any one of the plurality of access units accesses the A logical access unit in a fault-tolerant disk array, and includes respective local storage areas of a plurality of storage devices used to form the fault-tolerant disk array, and the data in the data are written to the plurality of The data of the access units are respectively protected in the plurality of access units by the check codes corresponding to the data respectively written to the plurality of access units; when the data are respectively written to the plurality of access units Any data in the data of each access unit is updated, maintaining a set of updated logical address pair access unit address table entries respectively corresponding to the updated data and additionally including auxiliary corresponding to the updated data. An update list of information is stored in a volatile memory, and a recovery log containing the update data is maintained in a log area of the fault-tolerant disk array for power failure recovery, wherein the recovery log includes a plurality of recovery log entries, each recovery log entry position corresponds to an access unit in the data area, and a value of the auxiliary information indicates a particular recovery log entry position; and a log entry having the same value when stored in the update list The auxiliary information reaches a predetermined threshold value, and the predetermined threshold value is at least two, it is determined that the set of updated logical address pair access unit address table entries corresponds to a logical address pair access unit address table corresponding to the same recovery log entry position the number of entries reaches the predetermined threshold, using the auxiliary information with the same value to access the logical address pair access unit bit in the update list corresponding to the same recovery log entry location address table entry, and update the logical address pair access unit address table with the logical address pair access unit address table entry in the update list corresponding to the same recovery log entry location, where the corresponding The logical address pair access unit address table entry at the same recovery log entry location represents the update data that should be stored to the same access unit in the table area; wherein the predetermined threshold value is irrelevant The capacity of the update list, regardless of the total amount of data in the update list. The method of claim 1, wherein the update list includes a set of pointers respectively corresponding to the set of updated logical address pair access unit address table entries, and the set of pointers respectively point to the restoration A set of restore log entries in the log. The method of claim 2, wherein the logical address pair access unit address table includes at least one sub-table that should be updated by the set of updated logical address pair access unit address table entries; And the position of the at least one sub-table in the table area corresponds to a group of access units in the table area. The method of claim 1, wherein the logical address pair access unit address table contains logical address pair access unit address table entries that should be updated by the set of updated At least one subtable. The method of claim 1, wherein the update list includes at least two pointers respectively corresponding to the logical address pair access unit address table entries, the at least two pointers respectively point to the restoration The same recovery log entry in the log, and the same recovery log entry points to the same access unit in the table area. A storage system, comprising: a fault-tolerant disk array (Redundant Array of Independent Disks, RAID); and a processing circuit for controlling the operation of the storage system and managing mapping information about the fault-tolerant disk array, The operations of the storage system include: writing data into a data area of the fault-tolerant disk array by means of write redirection, so that the data written in the writing process is written to the data At least one access unit of a plurality of access units in the area, and records the logical addresses of the data in the data respectively written to the plurality of access units and the data of the plurality of access units Mapping information between respective access-unit addresses to a logical-address-to-access-unit address table in a table region of the fault-tolerant disk array -to-access-unit-address table), wherein any access unit of the plurality of access units is a logical access unit when accessing the fault-tolerant disk array, and includes elements used to form the fault-tolerant disk array The respective local storage areas of the plurality of storage devices of the disk array, and the data respectively written to the plurality of access units in the data are respectively corresponding to the respective data in the plurality of access units where the check codes of the data written to the plurality of access units protection; when any of the data respectively written to the plurality of access units is updated, maintain an access unit address table entry containing a set of updated logical addresses respectively corresponding to the updated data ) and further includes an update list of auxiliary information corresponding to the update data in a volatile memory, and maintains a recovery log including the update data in a log area of the fault-tolerant disk array for power failure use of recovery, wherein the recovery log contains a plurality of recovery log entries, each recovery log entry location corresponds to an access unit in the data area, and a value of the auxiliary information indicates a particular recovery log entry location; and when Auxiliary information with the same value stored in the update list reaches a predetermined threshold value, and the predetermined threshold value is at least two, then it is determined that the set of updated logical addresses corresponds to the access unit address table entry. The number of logical address pair access unit address table entries of the same recovery log entry location reaches the predetermined threshold, using the auxiliary information with the same value to access the corresponding address in the update list The logical address of the same recovery log entry location is paired with the access unit address table entry, and the access unit address is paired with the logical address in the update list corresponding to the same recovery log entry location table entry to update the logical address pair access unit address table, wherein the logical address pair access unit address table entry corresponding to the same recovery log entry location represents the The update data of the same access unit; wherein, the predetermined threshold value is not related to the capacity of the update list, and is not related to the total data volume of the update list. A device for managing mapping information about a fault-tolerant disk array (Redundant Array of Independent Disks, RAID), the device comprises: a processing circuit located in a storage system for controlling the operation of the storage system, The operations of the storage system include: writing data into a data area of the fault-tolerant disk array by means of write redirection, so that the data written in the writing process is written to the data At least one access unit of a plurality of access units in the area, and records the logical addresses of the data in the data respectively written to the plurality of access units and the data of the plurality of access units Mapping information between respective access-unit addresses to a logical-address-to-access-unit address table in a table region of the fault-tolerant disk array -to-access-unit-address table), wherein any access unit of the plurality of access units is a logical access unit when accessing the fault-tolerant disk array, and includes elements used to form the fault-tolerant disk array The respective local storage areas of the plurality of storage devices of the disk array, and the data respectively written to the plurality of access units in the data are respectively corresponding to the respective data in the plurality of access units is protected by the check code of the data written to the plurality of access units; when any data in the data respectively written to the plurality of access units is updated, the maintenance includes the data corresponding to the updated data respectively. A set of updated logical address pair access unit address table entries and further includes an update list of auxiliary information respectively corresponding to the update data in a volatile memory, and maintains an update list including the update data The recovery log is stored in a log area of the fault-tolerant disk array for power failure recovery, wherein the recovery log includes a plurality of recovery log entries, each of which is repeated The original log entry position corresponds to an access unit in the data area, and a value of the auxiliary information indicates a specific recovery log entry position; and when auxiliary information with the same value stored in the update list reaches a predetermined value threshold value, and the predetermined threshold value is at least two, it is determined that the logical address pair access unit address corresponding to a same recovery log entry position in the set of updated logical address pair access unit address table entries The number of table entries reaches the predetermined threshold, the auxiliary information having the same value is used to access the logical address pair access unit in the update list corresponding to the same recovery log entry location address table entry, and update the logical address pair access unit address table with the logical address pair access unit address table entry in the update list corresponding to the same recovery log entry location, wherein The logical address pair access unit address table entry corresponding to the same recovery log entry location represents update data that should be stored to the same access unit in the table area; wherein the predetermined threshold is none About the capacity of the update list, and not about the total data volume of the update list.

Images