CN106453594A - A global logical clock synchronization distributed method - Google Patents
A global logical clock synchronization distributed method Download PDFInfo
- Publication number
- CN106453594A CN106453594A CN201610946525.3A CN201610946525A CN106453594A CN 106453594 A CN106453594 A CN 106453594A CN 201610946525 A CN201610946525 A CN 201610946525A CN 106453594 A CN106453594 A CN 106453594A
- Authority
- CN
- China
- Prior art keywords
- clock
- distributed method
- global
- event
- synchronization distributed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J3/00—Time-division multiplex systems
- H04J3/02—Details
- H04J3/06—Synchronising arrangements
- H04J3/0635—Clock or time synchronisation in a network
- H04J3/0638—Clock or time synchronisation among nodes; Internode synchronisation
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer And Data Communications (AREA)
Abstract
The invention provides a global logical clock synchronization distributed method. The method comprises the steps of 1, when a connection client executes an event m related to the global, an encoding A is requested to a clock system; and 2, the order in which the system performs operations is executed in accordance with a given logical clock number, and a process pi only brings forward a clock application of an next instruction after executing an instruction. Solving of the main problems in the coherence of data is explained. A problem that only one motion can be passes and a problem that the time for passing of the motion is influenced by the length of the motion is studied. The algorithm is improved under the premise that more than half of the motion is stored by a server cluster.
Description
Technical field
The invention belongs to data processing field, more particularly to a kind of global logic clock synchronization distributed method.
Background technology
Data Consistency is the classical problem of field of distributed type, while being also difficulties.Centered on server
Service model under, Single Point of Faliure can cause server externally provide service, it is therefore desirable to introduce multiple servers.So
Data Consistency will be brought, it is therefore necessary to have replica control protocols to realize the concordance of data.Server
Service ability is restricted by the hardware condition of its own network.The service ability that single server externally can be provided is limited road
Number, affected by bandwidth, network interface card disposal ability, and this is typical single-point problem.When single server reaches service
During the upper limit of ability, then higher service ability is externally provided using with multiple servers.Now need to solve the one of data
Cause sex chromosome mosaicism.The solution of single-point problem is that multiserver means polycentric using backup and multiserver mostly
Occur, how multiple centers ensure data consistency, and externally to provide faster service be the problem that primarily need to solve.Especially
Various transaction platforms (if multicenter unanimously can not be so that business receives impact) in Internet era which is, reach an agreement
Time used by property will affect the performance of Consumer's Experience and platform.Thus, the consistency algorithm of response time reasonable in design
Meaningful for actual application.
Content of the invention
In order to solve the above problems, the present invention provides a kind of global logic clock synchronization distributed method, to solving data
Subject matter present in concordance is set forth, and the time that can only pass through a proposal problem and pass a bill receives proposal
Effect length problem is studied, and in conjunction with the premise of the more than half storages of proposal being serviced device cluster, algorithm is improved.
A kind of global logic clock synchronization distributed method of the present invention is achieved through the following technical solutions：
A kind of global logic clock synchronization distributed method, including following step：
Step 1：When one connection client executing is to global related event m, to clock system application, one encodes A；
Step 2：System executes the order of operation to be executed according to given logical timer numbering, and process pi only has
An instruction is executed, the clock application of next instruction could have been proposed.
The clock system includes multiple time service centers, and ensures the consistent of multiple time service centers using paxos algorithm
Property.
The clock system time service process is as follows：
If a () event a represents the event that process Pi sends message m, then time stamp T m=Ci that message m includes
(a)；
B (), after message m is received, clock system can arrange the value of Cj so that it is more than or equal to its currency and is more than
Tm.
Clock system is additionally provided with autonomous learning element, and autonomous learning element can cause numbering monotonic increase.
The device have the advantages that：
Caused using system of the present invention in hgher efficiency when there is a large amount of requests；And can be by changing configuration file
To build larger range of application.
Specific embodiment
In distributed system, global clock needs can be to occurring event to be in systems ranked up.Energy wanted by one clock
There is sequencing, and its condition to be met for causal ordering for portraying event：For any one occurrence a, b：If a->B,
So Cfunction (a)<Cfunction(b).
Cfunction is defined as a function, any one occurrence a distribution numbering Cfunction (a) as in process.I.e.：
(1) each process Pi can increase the value of Ci between arbitrarily continuous two events；
(2) if event a represents the event that process Pi sends message m, then time stamp T m=Ci that message m includes
(a)；
(3) after message m is received, process Pj can arrange the value of Cj and cause it more than or equal to its currency and more than Tm.
A kind of global logic clock synchronization distributed method, including following step：
Step 1：When one connection client executing is to global related event m, to clock system application, one encodes A；It
After could execute, ensure that the operation of event m can be received by other clients；
Step 2：System executes the order of operation to be executed according to given logical timer numbering, i.e. each node
The event1 that C (event1) is 1 is first carried out, and one has only been executed event2 ... process pi that C (event2) is 2 is executed
Instruction, could propose the clock application of next instruction.
The clock system includes multiple time service centers；But the clock at multiple time service centers is the moment when being difficult to ensure that again
Carve and unanimously can only be consistent in certain accuracy rating.Therefore in order to complete to award any section in the case of multinode
Point identical clock, can run consistency algorithm paxos in these distributed time services in the heart and carry out the execution for event system
Clock cooperative signal is provided.So that the event that process in each node occurs in transaction system can be registered uniquely
Numbering, as long as which is to propose application through our clock system.
The clock system time service process is as follows：
If a () event a represents the event that process Pi sends message m, then time stamp T m=Ci that message m includes
(a)；
B (), after message m is received, clock system can arrange the value of Cj so that it is more than or equal to its currency and is more than
Tm.
Clock system is additionally provided with autonomous learning element, and autonomous learning element can cause numbering monotonic increase.
The lifted embodiment of the present invention or embodiment have been carried out to the object, technical solutions and advantages of the present invention further
Describe in detail, should be understood that embodiment provided above or embodiment are only the preferred embodiment of the present invention,
Not in order to limit the present invention, all any modifications made for the present invention within the spirit and principles in the present invention, equivalent,
Improve etc., should be included within the scope of the present invention.
Claims (4)
1. a kind of global logic clock synchronization distributed method, it is characterised in that including following step：
Step 1：When one connection client executing is to global related event m, to clock system application, one encodes A；
Step 2：System executes the order of operation to be executed according to given logical timer numbering, and process pi only has execution
A complete instruction, could propose the clock application of next instruction.
2. global logic clock according to claim 1 synchronization distributed method, it is characterised in that：The clock system bag
Multiple time service centers are included, and ensures the concordance at multiple time service centers using paxos algorithm.
3. global logic clock according to claim 1 synchronization distributed method, it is characterised in that：The clock system is awarded
When process as follows：
If a () event a represents the event that process Pi sends message m, then time stamp T m=Ci (a) that message m includes；
B (), after message m is received, clock system can arrange the value of Cj and cause it more than or equal to its currency and more than Tm.
4. global logic clock according to claim 1 synchronization distributed method, it is characterised in that：Clock system also sets up
There is autonomous learning element, autonomous learning element can cause numbering monotonic increase.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610946525.3A CN106453594A (en) | 2016-10-26 | 2016-10-26 | A global logical clock synchronization distributed method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610946525.3A CN106453594A (en) | 2016-10-26 | 2016-10-26 | A global logical clock synchronization distributed method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106453594A true CN106453594A (en) | 2017-02-22 |
Family
ID=58178394
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610946525.3A Pending CN106453594A (en) | 2016-10-26 | 2016-10-26 | A global logical clock synchronization distributed method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106453594A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111294310A (en) * | 2018-12-06 | 2020-06-16 | 上海西门子医疗器械有限公司 | Method for automatically extracting workflow and method for detecting intrusion in distributed system |
| CN113946448A (en) * | 2021-10-28 | 2022-01-18 | 苏州浪潮智能科技有限公司 | A method, device and electronic device for time sequence management of server cluster |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101520740A (en) * | 2009-04-03 | 2009-09-02 | 北京航空航天大学 | Method for realizing event consistency based on time mapping |
| CN102419764A (en) * | 2010-10-20 | 2012-04-18 | 微软公司 | Distributed transaction management for database systems with multiversioning |
| CN105900077A (en) * | 2013-11-05 | 2016-08-24 | 美国国家仪器有限公司 | Lossless time based data acquisition and control in a distributed system |
-
2016
- 2016-10-26 CN CN201610946525.3A patent/CN106453594A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101520740A (en) * | 2009-04-03 | 2009-09-02 | 北京航空航天大学 | Method for realizing event consistency based on time mapping |
| CN102419764A (en) * | 2010-10-20 | 2012-04-18 | 微软公司 | Distributed transaction management for database systems with multiversioning |
| CN105900077A (en) * | 2013-11-05 | 2016-08-24 | 美国国家仪器有限公司 | Lossless time based data acquisition and control in a distributed system |
Non-Patent Citations (1)
| Title |
|---|
| 王丽颖; 王红; 刘春涨: "基于分布式同步时钟的paxos算法改进", 《中国新通信》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111294310A (en) * | 2018-12-06 | 2020-06-16 | 上海西门子医疗器械有限公司 | Method for automatically extracting workflow and method for detecting intrusion in distributed system |
| CN113946448A (en) * | 2021-10-28 | 2022-01-18 | 苏州浪潮智能科技有限公司 | A method, device and electronic device for time sequence management of server cluster |
| CN113946448B (en) * | 2021-10-28 | 2024-05-28 | 苏州浪潮智能科技有限公司 | A server cluster timing management method, device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109361532B (en) | High availability system and method for network data analysis and computer readable storage medium | |
| EP2834755B1 (en) | Platform for continuous graph update and computation | |
| JP6692000B2 (en) | Risk identification method, risk identification device, cloud risk identification device and system | |
| US20130124616A1 (en) | Methods for dynamically generating application interfaces for modeled entities and devices thereof | |
| Zaharia et al. | Discretized streams: A fault-tolerant model for scalable stream processing | |
| CN103618652A (en) | Audit and depth analysis system and audit and depth analysis method of business data | |
| CN115297008B (en) | Collaborative training method, device, terminal and storage medium based on intelligent computing network | |
| CN115249073A (en) | A federated learning method and device | |
| CN113033082A (en) | Decentralized federated learning framework based on heterogeneous computational power perception and modeling method | |
| CN106033476A (en) | An Incremental Graph Computing Method in Distributed Computing Mode in Cloud Computing Environment | |
| CN103701783A (en) | Preprocessing unit, data processing system consisting of same, and processing method | |
| CN109614227A (en) | Task resource allocation method, apparatus, electronic device, and computer-readable medium | |
| CN106775936A (en) | The management method and device of a kind of virtual machine | |
| CN106209974B (en) | A data synchronization method, device and system | |
| CN108390771A (en) | A kind of network topology method for reconstructing and device | |
| CN103166964B (en) | Front end and back end real time communication method for super computer system monitoring | |
| Filip et al. | Data capsule: Representation of heterogeneous data in cloud-edge computing | |
| WO2024037368A1 (en) | Scheduling optimization method of scheduling apparatus, scheduling apparatus and storage medium | |
| CN110580307A (en) | A fast statistical processing method and device | |
| CN106453594A (en) | A global logical clock synchronization distributed method | |
| CN107845013A (en) | A kind of electric power transaction platform distributed application services calling processing method | |
| CN113515524B (en) | A method and device for automatically and dynamically allocating nodes in a distributed cache access layer | |
| Chemodanov et al. | A near optimal reliable orchestration approach for geo-distributed latency-sensitive SFCs | |
| CN109002666A (en) | Emulated computation method based on DR second order algorithm and DDS-QOS | |
| CN103269327A (en) | Method for realizing general file data service based on agent mechanism |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170222 |