Method, apparatus and computer program product for event sequencing

By using statistical methods and Bayes theorem to calculate the reliable probability of non-reliable edges in distributed file systems, accurately removing the most unreliable edges in event rings, solving the problems of low efficiency and high cost in traditional methods of event ring processing. , improves the accuracy and efficiency of event sorting.

CN112486940BActive Publication Date: 2024-12-03EMC IP HLDG CO LLC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201910866728.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2019-09-12
Publication Date
2024-12-03
Estimated Expiration
2039-09-12

AI Technical Summary

Technical Problem

Traditional methods are difficult to effectively solve the problem of event rings in distributed file systems, and rules-based methods require a lot of manual experience and are costly, so they cannot accurately remove non-reliable edges, which may lead to incorrect event sorting.

Method used

Using a statistical method-based event ring processing scheme, by calculating the reliability of non-reliable edges, using Bayes theorem to determine the reliability probability of each non-reliable edge, thereby accurately removing the most unreliable edges in the event ring and avoiding event rings. The occurrence of

Benefits of technology

It realizes the accurate removal of the most unreliable edges in the event ring in a distributed file system, improves the accuracy and efficiency of event sorting, and reduces manual intervention and costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112486940B_ABST
    Figure CN112486940B_ABST
Patent Text Reader

Abstract

Embodiments of the present disclosure relate to a method, device, and computer program product for event sorting in a distributed file system. The method includes generating an event ring indicating a causal relationship between multiple events based on collecting multiple events associated with an object from multiple nodes in the distributed file system, wherein the event ring includes one or more reliable edges and multiple unreliable edges. The method also includes determining the reliability of each unreliable edge in the event ring, and removing at least one unreliable edge from the event ring based on the determined reliability. Embodiments of the present disclosure use statistical methods to analyze the causal relationship between events in a distributed file system, and by calculating the reliability probability of each unreliable edge, the most unreliable edge in the event ring can be removed more accurately, thereby avoiding the occurrence of event rings.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of the present disclosure generally relate to the field of information technology, and more particularly to a method, device, and computer program product for event sequencing in a distributed file system. Background Art

[0002] Distributed file systems offer many significant advantages in building high-performance computing environments. One aspect is that they can be easily expanded even on a large scale. Another aspect is that they can use different protocols to store different types of data accessible to different types of clients. Distributed file systems can run on a cluster of nodes, allowing clients to connect to any node in the cluster to access and / or modify data stored on any node in the cluster.

[0003] Events within a file system can include creating a new file, changing permissions on a file, changing metadata for a file, and so on. These events can occur in parallel and can be processed on different nodes in a cluster of nodes. A distributed file system can have locking mechanisms in place that prevent access to a file while it is being modified by another user or file system process. When aggregating multiple events in a distributed file system, it is often desirable to know the causal order of these events. Summary of the invention

[0004] Embodiments of the present disclosure provide a method, apparatus, and computer program product for event sequencing in a distributed file system.

[0005] In one aspect of the present disclosure, a method for event sequencing is provided. The method includes: based on collecting multiple events associated with an object from multiple nodes in a distributed file system, generating an event ring indicating a causal relationship between the multiple events, wherein the event ring includes one or more reliable edges and multiple unreliable edges; determining multiple reliability levels of the multiple unreliable edges in the event ring; and removing at least one unreliable edge from the event ring based on the determined multiple reliability levels.

[0006] In another aspect of the present disclosure, a device for event sequencing is provided. The device includes a processing unit and a memory, wherein the memory is coupled to the processing unit and stores instructions. The instructions perform the following actions when executed by the processing unit: based on collecting multiple events associated with an object from multiple nodes in a distributed file system, an event ring indicating a causal relationship between multiple events is generated, wherein the event ring includes one or more reliable edges and multiple unreliable edges; multiple reliability levels of multiple unreliable edges in the event ring are determined; and based on the determined multiple reliability levels, at least one unreliable edge is removed from the event ring.

[0007] In another aspect of the present disclosure, a computer program product is provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, which, when executed, cause a computer to perform a method or process according to an embodiment of the present disclosure.

[0008] This summary is provided to introduce a selection of concepts in a simplified form, which are further described in the detailed description below. This summary is not intended to identify key features or essential features of the disclosure, nor is it intended to limit the scope of the various embodiments of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other objects, features and advantages of the present disclosure will become more apparent through a more detailed description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, wherein like reference numerals generally represent like elements throughout the exemplary embodiments of the present disclosure.

[0010] Figure 1 A schematic diagram of a distributed file system according to an embodiment of the present disclosure is shown;

[0011] Figure 2 A flowchart of a method for event sequencing according to an embodiment of the present disclosure is shown;

[0012] Figure 3 A schematic diagram for splitting an event chain according to an embodiment of the present disclosure is shown;

[0013] Figure 4A A schematic diagram for updating a strong edge table according to an embodiment of the present disclosure is shown;

[0014] Figure 4B A schematic diagram for updating a weak edge table according to an embodiment of the present disclosure is shown;

[0015] Figure 5 A schematic diagram showing a method for removing weak edges from an event ring according to an embodiment of the present disclosure; and

[0016] Figure 6 A schematic block diagram of a device that can be used to implement an embodiment of the present disclosure is shown. Specific embodiments

[0017] The preferred embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some specific embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0018] As used herein, the term "including" and its variations mean open inclusion, i.e., "including but not limited to". Unless otherwise stated, the term "or" means "and / or". The term "based on" means "based at least in part on". The terms "an example embodiment" and "an embodiment" mean "at least one example embodiment". The term "another embodiment" means "at least one additional embodiment". The terms "first", "second", etc. may refer to different or the same objects, unless explicitly indicated to be different.

[0019] Typically, multiple events collected from various nodes in a distributed file system need to be causally sorted, for example, to build a system log. Multiple events for the same or related objects in each node can be collected together, and then the causal relationship between the events can be constructed through version numbers. Generally speaking, low-version events usually occur before high-version events. Therefore, events with causal relationships can be linked together through directed edges to build an event chain or event graph. Each edge can also be divided into reliable edges and unreliable edges according to its reliability, where a reliable edge indicates that the two events between the edges must have a causal relationship, while an unreliable edge indicates that the two events between the edges may or may not have a causal relationship.

[0020] In some scenarios, the event chain generated by the version number may have an event loop (caused by some edges being unreliable edges), which is inconsistent with the actual scenario. Therefore, it is necessary to remove unreliable edges in the event loop in order to avoid the occurrence of event loops. One method to remove unreliable edges is to construct a series of rules, which is a deterministic method to eliminate event loops. However, the rule-based method cannot guarantee that all event loop problems will be solved, and in some scenarios it may not be possible or unknown to remove a specific unreliable edge among multiple unreliable edges, and incorrectly removing unreliable edges may cause errors in the entire event sorting process. Therefore, the traditional method cannot effectively solve all event loop problems in event sorting in distributed file systems. In addition, the traditional rule-based method requires a lot of manual experience, which has high labor costs and is difficult to implement.

[0021] To this end, the embodiments of the present disclosure propose a new solution for removing unreliable edges in an event ring based on statistics. The embodiments of the present disclosure use statistical methods to analyze the causal relationship of events in a distributed file system, and by calculating the reliability of each unreliable edge, the most unreliable edge in the event ring can be removed, thereby avoiding the occurrence of event rings.

[0022] The following references Figures 1 to 6It should be understood that these exemplary embodiments are provided only to enable those skilled in the art to better understand and implement the embodiments of the present disclosure, and are not intended to limit the scope of the present disclosure in any way.

[0023] Figure 1 FIG. 1 shows a schematic diagram of a distributed file system 100 according to an embodiment of the present disclosure. Figure 1 As shown, the distributed file system 100 may include multiple nodes 110, 120, 130, wherein each node may be a computing device, such as a server, a computer, etc. In an embodiment of the present disclosure, each node may include a recording module for recording events, wherein the events may include but are not limited to creating a file, modifying a file, deleting a file, renaming a file, etc. Figure 1 , node 110 includes a recording module 115, node 120 includes a recording module 125, and node 130 includes a recording module 135. Since the clocks of the nodes in the distributed file system 100 are not necessarily synchronized and there are transmission or synchronization delays, the distributed file system 100 does not use timestamps to indicate the causal relationship of each event; instead, the distributed file system uses the version number of the object (such as a file) to indicate the causal relationship of the event, for example, operations on the old version are usually performed before operations on the new version.

[0024] Continue to refer Figure 1 One or more nodes in the distributed file system 100 may also include an aggregation module for aggregating events on each node, for example, node 120 includes an aggregation module 128. Aggregation module 128 can collect events recorded by recording module 115, recording module 125, and recording module 135, and sort all events for the same or related objects, such as multiple events for the same object. In some cases, the event chain generated by event sorting may have an event loop. The embodiment of the present disclosure can accurately remove the most unreliable edge in the event loop through a statistical method.

[0025] Figure 2 FIG. 2 is a flowchart of a method 200 for event sequencing according to an embodiment of the present disclosure. The method 200 may be, for example, Figure 1 The described one or more nodes are executed, such as node 120.

[0026] At 202, based on collecting multiple events associated with one or more objects from multiple nodes in a distributed file system, an event ring indicating a causal relationship between the multiple events is generated, wherein the event ring includes one or more reliable edges and multiple unreliable edges. The embodiment method of the present disclosure is directed to the case where the event ring includes at least two unreliable edges, and the most unreliable unreliable edge is removed from the event ring. For the case where there is only one unreliable edge in the event ring, the unreliable edge can be directly removed.

[0027] In the context of the present disclosure, a reliable edge refers to a causal relationship between two events that can be guaranteed, also known as a "strong edge"; and an unreliable edge refers to a causal relationship between two events that cannot be guaranteed, but is only possible, also known as a "weak edge", and the reliability of an unreliable edge can be characterized by its reliability probability. In some embodiments, if the causal relationship is based on a pair of modified inode numbers and inode version numbers, the edge can be considered a strong edge; and if the causal relationship is based on a pair of viewed inode numbers and inode version numbers, the edge can be considered a weak edge, where inode can represent an in-memory representation or an on-disk data structure, which can store information or metadata about files and directories, such as file size, file ownership, access mode (read, write, execute permissions), creation and modification time and date, file type, data protection process information such as encryption and / or compression information, snapshot information, hash values ​​associated with file locations, mappings to cloud data objects, pointers to cloud metadata objects, and the like. It should be understood that any existing or future developed reliable and unreliable edge determination methods can be used in conjunction with the embodiments of the present disclosure.

[0028] Usually, there is a causal relationship between events in a distributed file system. For example, a file or directory needs to be created first before it can be modified or deleted, and a non-existent file cannot be deleted. If you try to delete a non-existent file, the distributed file system will throw an error and fail, and it will not generate an event. For example, for a file or directory, in many cases, it will go through the process of creation → modification → deletion or creation → rename → deletion or other actions.

[0029] At 204, multiple reliability levels of multiple unreliable edges in the event ring are determined. In some embodiments, the reliability probability of each unreliable edge in the multiple unreliable edges in the corresponding sub-event chain can be determined by Bayes' theorem based on a pre-built reliable edge table (i.e., strong edge table) and an unreliable edge table (i.e., weak edge table). In other words, in the case of a certain sub-event chain, the reliability probability of the unreliable edge (i.e., weak edge) is determined.

[0030] At 206, at least one unreliable edge is removed from the event ring based on the determined multiple reliability levels. For example, an unreliable edge with the lowest reliability level can be removed from the event ring, and then the events in the event ring are reordered into a loop-free event chain. When trying to remove one or more weak edges in the event ring, it is actually deciding which one or more weak edges are more unreliable, which can be determined, for example, by the reliability probability of each weak edge.

[0031] Therefore, the method 200 of the embodiment of the present disclosure uses a statistical method to analyze the causal relationship of events in a distributed file system. By calculating the reliability of each weak edge, the weakest edge in the event loop can be removed more accurately, thereby avoiding the occurrence of the event loop.

[0032] Some embodiments of the present disclosure may include two phases. In the first phase, historical data is collected, and during the historical successful event sorting, the event data is processed and useful information is stored. In the second phase, the reliable probability of each weak edge in the event loop is calculated in the actual scenario, which uses the information stored in the first phase to decide which weak edge to remove when an event loop occurs. Figure 3 and Figure 4A-4B Describes an example implementation of the first phase, see Figure 5 An example implementation of the second phase is described.

[0033] Figure 3 FIG300 shows a diagram of splitting an event chain according to an embodiment of the present disclosure. In the historical successful event sorting, a directed acyclic graph of events, such as various acyclic event chains, can be obtained. For a longer event chain, each event chain in a plurality of acyclic event chains is first split into a plurality of sub-event chains based on a predetermined splitting rule. Figure 3 As shown, for the acyclic event chain 310, it can be split into multiple sub-event chains 320. The splitting rule can be, for example: each sub-event chain includes at most 5 events, wherein there can be at most 2 events before the edge to be evaluated, and at most 1 event after it. It should be understood that the splitting rule is a hyperparameter, which can be adjusted according to different workflows.

[0034] Continue to refer Figure 3According to the above rules, the acyclic event chain 310A→B→C→D→E→F can be split into the following sub-event chains: sub-event chain 321A→B→C, which is used to evaluate the edge A→B; sub-event chain 322A→B→C→D, which is used to evaluate the edge B→C; sub-event chain 323A→B→C→D→E, which is used to evaluate the edge C→D; sub-event chain 324B→C→D→E→F, which is used to evaluate the edge D→E; sub-event chain 25C→D→E→F, which is used to evaluate the edge E→F. According to an embodiment of the present disclosure, only a limited number of events before and after the edge to be evaluated are used to calculate the reliable probability. In other words, the reason for splitting the acyclic event chain is that the impact of events that are too far away on the edge is usually small, so it can be ignored.

[0035] Next, the strong edge table and the weak edge table according to the embodiment of the present disclosure are maintained according to the split sub-event chain. For a given edge in a given sub-event chain, first determine whether the given edge is a strong edge or a weak edge; if it is a strong edge, update the strong edge table, such as Figure 4A As shown; otherwise, if it is a weak edge, update the weak edge table, such as Figure 4B shown.

[0036] In an embodiment of the present disclosure, an event chain or a sub-event chain may represent a chain of multiple events with a causal relationship, such as create → modify → delete; wherein Es represents an event where edge E is a strong edge, Ew represents an event where edge E is a weak edge, and Ch represents an event where sub-event chain Ch occurs. In addition, #Ct represents the total number of sub-event chains, #Ch represents the number of sub-event chains Ch, #Et represents the total number of sub-event chains that evaluate edge E, wherein edge E is a strong edge or a weak edge, #Es represents the number of sub-event chains where E is a strong edge among all sub-event chains that evaluate edge E, #Ew represents the number of sub-event chains where E is a weak edge among all sub-event chains that evaluate edge E, #Ch(Es) represents the number of sub-event chains Ch with a strong edge E, #Ch(Ew) represents the number of sub-event chains Ch with a weak edge E, and #Ct(Es) represents the total number of sub-event chains with a strong edge E.

[0037] Figure 4A FIG400 shows a diagram for updating a strong edge table according to an embodiment of the present disclosure. Figure 4AAs shown, the embodiments of the present disclosure propose a strong edge table and a weak edge table. In the strong edge table, each row of entries includes a strong edge Es, a sub-event chain Ch, and the number #Ch(Es) of sub-event chains Ch with a strong edge E, that is, each row of entries stores the frequency of occurrence of sub-event chains Ch, where sub-event chain Ch evaluates edge E and E is a strong edge. In the weak edge table, each row of entries includes a weak edge Ew, a sub-event chain Ch, and the number #Ch(Ew) of sub-event chains Ch with a weak edge E, that is, each row of entries stores the frequency of occurrence of sub-event chains Ch, where sub-event chain Ch evaluates edge E and E is a weak edge. For strong edges or weak edges that do not exist in the strong edge table and the weak edge table, new entries can be inserted; and for strong edges and weak edges that already exist in the strong edge table and the weak edge table, their chain number information can be updated.

[0038] Continue to refer Figure 4A Table 410 shows the state before processing the sub-event chain 322A→B→C→D described above. For the sake of brevity, some sub-event chains are merged in Table 410 and represented by "other sub-event chains". Before updating the strong edge table or the weak edge table, first determine whether the given edge is a strong edge or a weak edge. If the given edge B→C is determined to be a strong edge, the strong edge table is updated in Table 410, and 1 is added to the corresponding entry in the strong edge table (for example, in the row with sequence number 1 in the strong edge table). The updated strong edge table is shown in Table 420.

[0039] Figure 4B A diagram 450 for updating a weak edge table according to an embodiment of the present disclosure is shown. If a given edge B→C is determined to be a weak edge, the weak edge table is updated in table 410, and 1 is added to the corresponding entry in the weak edge table (e.g., in the row with sequence number 1 in the weak edge table), and the updated weak edge table is shown in table 460.

[0040] Next, after completing the processing of the historical sub-event chains, the strong edge table and the weak edge table are constructed, and then based on the strong edge table and the weak edge table, the Bayesian theorem can be used to calculate the reliable probability of each weak edge in each sub-event chain.

[0041] Bayes' theorem is a theorem about the conditional probability (or marginal probability) of random events A and B, where P(A|B) is the probability of A occurring when B occurs. Bayes' theorem can be expressed by the following formula (1):

[0042] P(A|B)=P(B|A)*P(A) / P(B) (1)

[0043] Among them, A and B are events, and P(B) is not equal to 0; P(A|B) is the conditional probability of A occurring when B is known to have occurred, P(B|A) is the conditional probability of B occurring when A is known to have occurred, P(A) is the prior probability or marginal probability of A, and P(B) is the prior probability or marginal probability of B.

[0044] The following formula (2) shows the use of Bayes' theorem to calculate the reliable probability of a weak edge when a given sub-event chain occurs:

[0045] P(Es|Ch)=P(Ch|Es)*P(Es) / P(Ch) (2)

[0046] Where P(Es|Ch) represents the conditional probability of event Es occurring given the occurrence of sub-event chain Ch, or the reliable probability that edge E in sub-event chain Ch is a strong edge; P(Ch|Es) represents the conditional probability of sub-event chain Ch occurring given edge E is a strong edge; P(Es) is the marginal probability of observing the strong edge E; P(Ch) is the marginal probability of observing sub-event chain Ch.

[0047] In order to calculate the conditional probability P(Es|Ch) of the occurrence of event Es when a given sub-event chain Ch occurs, P(Ch|Es), P(Es), and P(Ch) need to be calculated. P(Ch|Es), P(Es), and P(Ch) can be calculated based on the constructed strong edge table and weak edge table. Specifically, P(Ch|Es), P(Es), and P(Ch) can be calculated using the following equations (3)-(5):

[0048] P(Ch|Es)=#Ch(Es) / #Ct(Es) (3)

[0049] P(Es)=#Es / #Et (4)

[0050] P(Ch)=#Ch / #Ct (5)

[0051] Therefore, based on the above formulas (3)-(5), for a given sub-event chain A→B→C→D and a given edge B→C, based on Figure 4A From Table 410 in, we can calculate P(Ch|Es)=P(sub-event chain A→B→C→D|edge B→C is a strong edge)=10 / (10+190)=0.05, P(Es)=P(edge ​​B→C is a strong edge)=200 / 400=0.5, P(Ch)=P(sub-event chain is A→B→C→D)=30 / 750=0.04. In this way, the probability P(Es|Ch) of event Es occurring when a given sub-event chain Ch occurs can be calculated according to the above formula (2).

[0052] Figure 5 FIG. 5 shows an example diagram 500 for removing a weak edge from an event ring according to an embodiment of the present disclosure. For example, when a contradictory event ring appears in the actual event ordering, the weak edge can be removed by using Figure 4A and 4B The data in the strong edge table and weak edge table in the image are used to eliminate event loops. Figure 5As shown, assuming that there is an event loop 510: A→B→C→D→G→H→B→C→D→E→F→A, first identify the weak edges therein (such as the weak edges B→C and D→E represented by the dotted lines in the event loop 510), and then split the event loop 510 into multiple sub-event chains 520 according to the weak edges, which include sub-event chains 521 (A→B→C→D) and 522 (B→C→D→E→F), so that each of the split sub-event chains includes a weak edge. It should be understood that in each sub-event chain, except for the weak edge to be evaluated, it is not important whether other edges are strong edges, and only the event order of other events needs to be paid attention to. In addition, as Figure 5 As shown, some events can appear multiple times in an event chain.

[0053] After obtaining multiple sub-event chains 520 including weak edges to be evaluated, Bayes' theorem is used to determine the reliable probability of each weak edge in the corresponding sub-event chain based on the pre-constructed strong edge table and weak edge table. For example, the reliable probability of edge B→C in the sub-event chain A→B→C→D can be determined as P(B→C is a strong edge|sub-event chain is A→B→C→D)=P(sub-event chain A→B→C→D|edge B→C is a strong edge)*P(edge ​​B→C is a strong edge) / P(sub-event chain is A→B→C→D)=0.05*0.5 / 0.04=0.625. Based on a similar method, P(D→E is a strong edge|sub-event chain is B→C→D→E→F)=0.857 can be calculated.

[0054] By comparison, the reliability probability of P(B→C is a strong edge | the sub-event chain is A→B→C→D) is less than P(D→E is a strong edge | the sub-event chain is B→C→D→E→F), that is, the edge D→E is more likely to be a strong edge, so the edge B→C can be removed from the event ring A→B→C→D→G→H→B→C→D→E→F→A to obtain a loop-free event chain 530, that is, C→D→G→H→B→C→D→E→F→A. For example, remove a weak edge with the lowest or relatively low reliability.

[0055] Therefore, according to some embodiments of the present disclosure, a statistical method based on Bayes' theorem is used to determine the reliable probability of a weak edge in a corresponding sub-event chain, thereby determining which weak edge should be removed from a contradictory event ring. Therefore, the embodiments of the present disclosure can more effectively remove the ring in the event ring and solve the problem of the event ring.

[0056] Figure 6 1 shows a schematic block diagram of a device 600 that can be used to implement an embodiment of the present disclosure. The device 600 may be a device or apparatus described in an embodiment of the present disclosure. Figure 6As shown, the device 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 602 or loaded from a storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 can also be stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0057] A number of components in the device 600 are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0058] The various methods or processes described above may be performed by the processing unit 601. For example, in some embodiments, the methods may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the CPU 601, one or more steps or actions in the methods or processes described above may be performed.

[0059] In some embodiments, the methods and processes described above may be implemented as a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present disclosure.

[0060] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0061] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0062] The computer program instructions for performing the disclosed operation may be assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, and conventional procedural programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, executed as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, 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., utilizing an Internet service provider to connect via the Internet). In certain embodiments, by utilizing the state information of a computer-readable program instruction to customize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit may execute a computer-readable program instruction, thereby realizing various aspects of the present disclosure.

[0063] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0064] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0065] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the equipment, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each block in the flow chart or block diagram can represent a part of a module, a program segment or an instruction, and a part of the module, a program segment or an instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the blocks can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous blocks can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flow chart, and the combination of the blocks in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0066] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for event sequencing, comprising: Based on collecting a plurality of events associated with an object from a plurality of nodes in a distributed file system, an event ring indicating a causal relationship between the plurality of events is generated, the event ring including one or more reliable edges each indicating that a causal relationship between two events is guaranteed and a plurality of unreliable edges each indicating that a causal relationship between two events is not guaranteed, wherein the generating includes: identifying the plurality of unreliable edges in the event ring; and Based on the identified multiple unreliable edges, split the event ring into multiple sub-event chains, so that each of the multiple sub-event chains includes an unreliable edge; Determining multiple reliability levels of the multiple unreliable edges in the event ring, wherein the determining comprises: determining a reliability probability of each of the multiple unreliable edges when in a corresponding sub-event chain in the multiple sub-event chains based on a pre-constructed reliable edge table and an unreliable edge table, the reliable edge table and the unreliable edge table being maintained based on the split multiple sub-event chains; and At least one unreliable edge among the plurality of unreliable edges is removed from the event ring based on the plurality of reliability levels.

2. The method according to claim 1, further comprising: Collect multiple acyclic event chains that have been successfully sequenced historically; as well as processing the plurality of acyclic event chains; as well as Stores historical statistics about reliable and unreliable edges.

3. The method of claim 2, wherein the processing of the plurality of acyclic event chains comprises: Based on a predetermined splitting rule, each event chain in the plurality of acyclic event chains is split into a plurality of sub-event chains.

4. The method according to claim 3, wherein the storing of the historical statistical data about reliable edges and unreliable edges comprises: For a given edge in a given sub-event chain, determining whether the given edge is a reliable edge or an unreliable edge; According to determining that the given edge is a reliable edge, updating first information related to the reliable edge and the given sub-event chain in a reliable edge table; as well as According to determining that the given edge is an unreliable edge, second information related to the unreliable edge and the given sub-event chain is updated in an unreliable edge table.

5. The method according to claim 1, wherein determining the reliable probability of each of the multiple unreliable edges when it is in the corresponding sub-event chain in the multiple sub-event chains based on the pre-constructed reliable edge table and the unreliable edge table is based on Bayes' theorem.

6. The method according to claim 5, wherein the determination of the reliability probability of each unreliable edge of the plurality of unreliable edges when in the corresponding sub-event chain based on Bayes' theorem comprises: For a given unreliable edge in a given sub-event chain, Determine a first probability that the given sub-event chain occurs when the given unreliable edge is reliable; determining a second probability that the given unreliable edge is reliable; determining a third probability of occurrence of the given sub-event chain; Based on the first probability, the second probability and the third probability, a given reliable probability of the given unreliable edge in the given sub-event chain is determined.

7. The method of claim 6, wherein the removing of the at least one unreliable edge from the event ring comprises: Remove an unreliable edge with the lowest reliability from the event ring; as well as The events in the event ring are reordered into an acyclic event chain.

8. A device for event sequencing, comprising: Processing unit; as well as A memory coupled to the processing unit and storing instructions that, when executed by the processing unit, perform the following actions: Based on collecting a plurality of events associated with an object from a plurality of nodes in a distributed file system, an event ring indicating a causal relationship between the plurality of events is generated, wherein the event ring includes one or more reliable edges each indicating that a causal relationship between two events is guaranteed and a plurality of unreliable edges each indicating that a causal relationship between two events is not guaranteed, wherein the generating includes: identifying the plurality of unreliable edges in the event ring; and Based on the identified multiple unreliable edges, split the event ring into multiple sub-event chains, so that each of the multiple sub-event chains includes an unreliable edge; Determining multiple reliability levels of the multiple unreliable edges in the event ring to generate multiple determined reliability levels, wherein the determining comprises: determining a reliability probability of each of the multiple unreliable edges when in a corresponding sub-event chain in the multiple sub-event chains based on a pre-constructed reliable edge table and an unreliable edge table, the reliable edge table and the unreliable edge table being maintained based on the split multiple sub-event chains; and At least one unreliable edge is removed from the event ring based on the determined multiple reliability levels.

9. The device of claim 8, the actions further comprising: Collect multiple acyclic event chains that have been successfully sequenced historically; as well as processing the plurality of acyclic event chains; as well as Stores historical statistics about reliable and unreliable edges.

10. The apparatus of claim 9, wherein the processing of the plurality of acyclic event chains comprises: Based on a predetermined splitting rule, each event chain in the plurality of acyclic event chains is split into a plurality of sub-event chains.

11. The apparatus of claim 10, wherein the storing of the historical statistical data about reliable edges and unreliable edges comprises: For a given edge in a given sub-event chain, determining whether the given edge is a reliable edge or an unreliable edge; Based on the determination that the given edge is a reliable edge, updating first information related to the reliable edge and the given sub-event chain in a reliable edge table; as well as According to the determination that the given edge is an unreliable edge, second information related to the unreliable edge and the given sub-event chain is updated in an unreliable edge table.

12. The device according to claim 8, determines the reliable probability of each of the multiple unreliable edges when it is in the corresponding sub-event chain in the multiple sub-event chains based on the pre-constructed reliable edge table and the unreliable edge table based on Bayes' theorem.

13. The apparatus of claim 12, wherein the determining of the reliability probability of each of the plurality of unreliable edges when in the corresponding sub-event chain based on Bayes' theorem comprises: For a given unreliable edge in a given sub-event chain, Determine a first probability that the given sub-event chain occurs when the given unreliable edge is reliable; determining a second probability that the given unreliable edge is reliable; determining a third probability of occurrence of the given sub-event chain; Based on the first probability, the second probability and the third probability, a given reliable probability of the given unreliable edge in the given sub-event chain is determined.

14. The apparatus of claim 13, wherein the removing of the at least one unreliable edge from the event ring comprises: Remove an unreliable edge with the lowest reliability from the event ring; as well as The events in the event ring are reordered into an acyclic event chain.

15. A computer program product tangibly stored on a non-transitory computer readable medium and comprising computer executable instructions which, when executed, cause a computer to perform the method according to any one of claims 1 to 7.