CN111753411B - Cloud application reliability evaluation method considering edge cloud cooperation - Google Patents
Cloud application reliability evaluation method considering edge cloud cooperation Download PDFInfo
- Publication number
- CN111753411B CN111753411B CN202010532373.9A CN202010532373A CN111753411B CN 111753411 B CN111753411 B CN 111753411B CN 202010532373 A CN202010532373 A CN 202010532373A CN 111753411 B CN111753411 B CN 111753411B
- Authority
- CN
- China
- Prior art keywords
- state
- reliability
- component
- cloud
- simulation
- 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.)
- Expired - Fee Related
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 19
- 238000004088 simulation Methods 0.000 claims abstract description 45
- 238000000342 Monte Carlo simulation Methods 0.000 claims abstract description 12
- 238000009826 distribution Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 14
- 238000000034 method Methods 0.000 claims description 10
- 238000009827 uniform distribution Methods 0.000 claims description 3
- 230000001186 cumulative effect Effects 0.000 claims description 2
- 238000005315 distribution function Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 6
- 230000006872 improvement Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 3
- 238000013508 migration Methods 0.000 description 3
- 230000005012 migration Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 241000287196 Asthenes Species 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a cloud application reliability evaluation method considering edge cloud cooperation, which comprises the following steps: step S1: establishing a directed acyclic graph model based on the hierarchical dependency relationship among the components; step S2: calculating the reliability of each component based on the historical fault information of the component; and step S3: performing reliability simulation on the fault distribution based on the components by adopting Monte Carlo simulation; and step S4: determining the state of the cloud application according to the simulation state of each component in the directed acyclic graph model; step S5: and repeating the step S3 and the step S4 until the ending condition is met. According to the cloud application reliability evaluation method considering edge cloud cooperation, the reliability of the cloud application can be effectively evaluated through the setting of the steps S1 to S5.
Description
Technical Field
The invention relates to a cloud application reliability evaluation method, in particular to a Monte Carlo simulation-based reliability evaluation method considering edge cloud cooperation.
Background
The cloud computing is widely applied to various fields such as medical treatment, finance, education, scientific research and the like, and application programs with various responses run on a cloud platform and provide services for users in a remote access mode. The cloud application may be deployed in a variety of ways, including private, public, community, and hybrid clouds. In recent years, with the expansion of cloud application scale and the development of communication technologies such as 5G, edge clouds closer to terminals begin to emerge, so that the system architecture of cloud applications must consider edge cloud cooperation in order to utilize cloud resources more efficiently. Meanwhile, reliability is one of important indexes of cloud application service quality, and unreliable means a fault, so that not only user experience is influenced, but also economic loss is caused.
Reliability evaluation is the basis for improving the reliability of cloud applications, and the evaluation method usually requires the establishment of a system model. The cloud application is usually deployed on a cloud platform in a component-based manner, and each functional module is divided into independent components and deployed in a virtual machine or a container, so that a component-based model becomes a widely adopted cloud application modeling manner. Meanwhile, the component-based cloud application model makes it possible to simulate the reliability of a cloud application by simulating the reliability of each component.
The existing main reliability evaluation methods include: the patent document CN 105389434A proposes a reliability evaluation method for a multi-failure-mode cloud computing platform, which aims to establish a reliability analysis method for a multi-failure-mode cloud computing platform based on an extended form multivalued decision diagram of BDD, so as to allow multiple failure modes to be modeled and analyzed together, and improve accuracy and efficiency of reliability analysis; the patent document CN 108460177A proposes a reliability approximate calculation method for a large-scale multi-state serial-parallel system, which divides the connection structure between any father node and all the subordinate child nodes thereof into four types, then processes the four types in different ways, and finally obtains the reliability of the multi-state serial-parallel system through summary calculation, thereby realizing the balance between calculation precision and calculation efficiency; the patent document CN 107292019A provides a reliability analysis and calculation method for a standby system based on a polymorphic decision diagram, wherein the method establishes a polymorphic system model, establishes the polymorphic decision diagram according to the state transition possibility, and calculates the probability of each path in the diagram so as to calculate the reliability of the system; the patent document CN 106250251A proposes a cloud computing system reliability modeling method considering common cause and virtual machine fault migration, which considers fault-tolerant strategies of common cause faults of multiple virtual machines and virtual machine migration caused by server faults in a cloud computing system, and based on a state space model, solves the problem that other models do not consider the common cause faults and virtual machine fault migration well, simplifies the state space, and improves modeling efficiency.
The above patents all propose reliability evaluation methods for cloud platforms or systems from different angles, but still have certain defects: if the scene is single, only the reliability of the traditional cloud platform is considered, and the application and the edge cloud are not considered; the failure mode is single, most of the failure modes only consider hardware failures and do not consider software failures; the modeling efficiency and the calculation accuracy are not high.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a reliability evaluation method based on Monte Carlo simulation, which simultaneously models an application and a cloud platform and considers edge cloud cooperation, so as to solve the problem of accurately evaluating the reliability of the cloud application in an edge cloud cooperation scene.
In order to achieve the purpose, the invention provides the following technical scheme: a cloud application reliability evaluation method considering edge cloud cooperation comprises the following steps:
step S1: establishing a directed acyclic graph model based on the hierarchical dependency relationship among the components;
step S2: calculating the reliability of each component based on the historical fault information of the component;
and step S3: performing reliability simulation on the fault distribution based on the components by adopting Monte Carlo simulation;
and step S4: determining the state of the cloud application according to the simulation state of each component in the directed acyclic graph model;
step S5: and repeating the step S3 and the step S4 until the ending condition is met.
As a further improvement of the present invention, the step S1 includes the steps of:
step S1.1, analyzing related components in the cloud application, decomposing the cloud application into a plurality of sub-modules, wherein each sub-module represents a service;
step S1.2, determining the dependency relationship among the components in the step S1.1 on the basis that one service needs a calculation result based on another service to complete a task and the service depends on the service providing the calculation result;
and S1.3, establishing a directed acyclic graph model according to the components and the dependency relationship thereof.
As a further improvement of the present invention, the specific steps of calculating the reliability in step S2 are as follows:
step S2.1, taking the unit time as one day, and if a component has a failures in one day on average, the failure rate per unit time is λ = a;
step S2.1, according to the formula r = e -λ ＝e -a And calculating the reliability r of the unit time.
As a further improvement of the present invention, the specific steps of performing the simulation in step S3 are as follows:
s3.1, based on the reliability of each component, generating a random number representing the reliability of each component according to the uniform distribution U [0, 1);
and S3.2, recording the reliability value of the ith component as di, and recording the state as Si ∈ [0,1], wherein the method comprises the following steps:
wherein λ is i For the failure rate per unit time of the ith component, si =0 indicates that the state of the ith component in the unit time is a failure state, and correspondingly, si =1 indicates that the state of the ith component in the unit time is a normal state.
As a further improvement of the present invention, the specific steps of determining the state of the cloud application in step S4 are as follows:
s4.1, judging the states of all service instances according to rules;
s4.2, judging the state of each service according to the state of the service instance;
and S4.3, after each service state is determined, determining the application state according to the dependency relationship among the services.
As a further improvement of the present invention, the meeting of the ending condition in step S5 is that the simulation structure simultaneously meets three constraints, which are respectively a confidence interval constraint, an effective number constraint and a fault number constraint of the simulation result.
Compared with the prior art, the method has the advantages that the modeling is applied to the cloud platform, both hardware faults and software faults are considered, the reliability of the cloud application is accurately evaluated based on Monte Carlo simulation, the application and the edge cloud are considered, and the modeling efficiency and the computing accuracy are improved.
Drawings
FIG. 1 is a diagram of all component relationships relied upon by a cloud application;
FIG. 2 is a service dependency diagram.
Detailed Description
The invention will be further described in detail with reference to the following examples, which are given in the accompanying drawings.
The reliability evaluation method based on Monte Carlo simulation and capable of modeling the application and the cloud platform simultaneously and considering edge cloud cooperation is used for solving the problem of accurately evaluating the reliability of the cloud application in an edge cloud cooperation scene.
In order to achieve the purpose, the invention provides the following technical scheme:
step S1: and establishing a directed acyclic graph model based on the hierarchical dependency relationship among the components.
Step S2: the reliability of each component is calculated based on its historical failure information.
And step S3: and performing reliability simulation on the fault distribution based on the components by adopting Monte Carlo simulation.
And step S4: and determining the state of the cloud application according to the simulation state of each component in the directed acyclic graph model.
Step S5: and repeating S3 and S4 until the ending condition is met.
The step S1 specifically includes: establishing a directed acyclic graph model based on the hierarchical dependency relationship among the components, and specifically comprising the following steps:
step S1.1: cloud application related components are analyzed. The cloud application is decomposed into a plurality of sub-modules, each sub-module representing a service. The number of instances needing to be deployed is determined by each service according to different fault tolerance requirements, each service instance is deployed in different virtual machines or containers, the containers can be deployed in the virtual machines or directly deployed on physical servers, and the virtual machines are directly deployed on the physical servers. According to the nature of the functions completed by the virtual machines or containers, the virtual machines or containers are deployed on physical servers in different positions, for example, the containers or virtual machines containing user privacy information are deployed in a private cloud, the containers or virtual machines containing common tasks are deployed in a public cloud, and the containers or virtual machines containing real-time rapid processing tasks are deployed in an edge cloud. All components on which the service, the container, the virtual machine and the physical server, namely the cloud application depend are shown in the figure 1
Step S1.2: further, the dependency relationship between the above components is determined. The invention defines the dependency relationship as the relationship that one component needs another component to complete the function of the component, for example, if one service needs the calculation result based on another service to complete the task, the service depends on the service providing the calculation result; if the virtual machine needs to normally operate by the physical server, the virtual machine component depends on the physical server component. As shown in fig. 2 below, service S0 depends on services S1, S2, and S3, and services S4 and S5 both depend on service S6.
Step S1.3: further, a directed acyclic graph model is built according to the components and the dependency relationships thereof. From bottom to top, the directed acyclic graph can be four layers: the system comprises a physical server layer, a virtual machine layer, a container layer and an application layer. The physical servers comprise servers in private cloud, public cloud and edge cloud, and each server can host a plurality of virtual machines and a plurality of containers simultaneously; several services may be deployed on a virtual machine, or several containers hosted thereon; one service is deployed in one container. According to the four-layer dependency relationship, a directed acyclic graph G (V, E) between the components is established, wherein V = { V1, V2, \8230, vm } is a node set in the directed acyclic graph, m is the number of nodes, namely the number of the components, E = { E1, E2, \8230, en } is a set of edges in the directed acyclic graph, and n is the number of the edges, namely the number of the dependency relationship between the components.
The step S2 specifically includes: and analyzing the failure rate lambda and the reliability r of the component per unit time based on the historical failure information of the component, wherein the unit time is one day. For example, if a component fails a times a day on average, the failure rate per unit time is λ = a. Assuming that the occurrence of all faults follows an exponential distribution, the unit time reliability of the above-mentioned components is r = e -λ ＝e -a . Thus, given a failure rate λ per unit time for a component, its probability of being in a normal state per unit time is e -λ Accordingly, the probability of being in the fault state is 1-e -λ 。
The failure rates per unit time are not exactly the same for different types of components: the failure rate per unit time of the software component can be obtained by analyzing an operation log file in a period of time, wherein the software component refers to a component except a physical server in the components; the Failure rate per unit Time of a hardware component is usually much lower than that of a software component, and the Failure rate per unit Time can be obtained by analyzing the running log of the hardware component and can also be calculated according To the theoretical MTTF (Mean Time To Failure) of the hardware component, wherein the calculation formula is
Further, the reliability r of each component is calculated from its failure rate λ per unit time.
Step S3 specifically includes: the reliability of each component is simulated using Monte Carlo simulation based on the reliability of all components, and specifically, a random number representing the reliability of each component is generated from a uniform distribution U [0, 1) based on the reliability of each component. Specifically, the reliability value of the ith component is recorded as di, and the state is recorded as Si ∈ [0,1], so that:
wherein λ is i For the failure rate of the ith component per unit time, si =0 indicates that the state of the ith component in the unit time is a failure state, and correspondingly, si =1 indicates that the state of the ith component in the unit time is a normal state.
Step S4 specifically includes: and determining the state of the cloud application according to the simulation state of each component in the directed acyclic graph model. From bottom to top, if a physical server component is in a failure state, all virtual machines and container components thereon are in a failure state; if the virtual machine component is in a fault state, all container components or service instance components on the virtual machine component are in the fault state; if the container component is in a failed state, the service instance deployed thereon is in a failed state. According to the above rules, the status of all service instances can be finally determined. Further, the status of each service is determined based on the service instance status. Specifically, each service corresponds to a fault tolerance requirement, which is denoted as k-out-of-n, and indicates that one service has n instances, and if at least k instances are required to be in a normal state, the service is in a normal state, otherwise, the service is in a fault state. For the jth service, the fault tolerance requirement is recorded as kj-out-of-nj, the state is recorded as Sj, and Sj belongs to {0,1}, and if the number of instances in the normal state is gj, the following steps are provided:
and after each service state is determined, determining the state of the application according to the dependency relationship among the services. As shown in FIG. 2, there is S according to the dependency relationship between the services 5 ＝S 5 ∧S 6 That is, the state of the 5 th service is determined by the state of the 5 th service and the state of the 6 th service, and the same principle is as follows:
S 4 ＝S 4 ∧S 6
S 3 ＝S 3 ∧S 5 ∧S 6
S 2 ＝S 2 ∧S 4 ∧S 6
S 1 ＝S 1 ∧S 4 ∧S 6
S 0 ＝S 0 ∧S 1 ∧S 2 ∧S 3 ∧S 4 ∧S 5 ∧S 6
thus, the state S of the application app ＝S 0 ＝S 0 ∧S 1 ∧S 2 ∧S 3 ∧S 4 ∧S 5 ∧S 6 That is, in any round of simulation, if any one service is in a failure state, the application is in a failure state, and if all the services are in a normal state, the application is in a normal state.
Step S5 specifically includes: and repeating S3 and S4 until the end condition is met. The invention combines the steps S3 and S4 into a simulation turn, and records the reliability result corresponding to the application state obtained in the step S4 in the ith turn as
The application reliability evaluated after n rounds of simulation is recorded as
Then there are:
namely, the average value of the application state values in n-round simulation is used as the simulation result of the application reliability.
The invention adopts three kinds of constraints as the stopping rules of the simulation, the corresponding constraint value is updated after each round of simulation is finished, when the simulation result simultaneously meets the three kinds of constraints, the simulation is finished, and the three kinds of constraints are respectively:
1. and (5) constraining the confidence interval of the simulation result. The constraints ensure the credibility of the simulation results. Noting the true reliability of the simulated application as R real Setting of simulation resultsThe confidence interval is α, which may be a common confidence level, e.g., 95%, 99%, etc. For an n-round monte carlo simulation, α satisfies:
wherein phi -1 Which is the inverse of the cumulative distribution function of a standard normal distribution,
is the standard deviation of the simulation results, wherein
Is the current round simulation result, the value of which is 0 or 1,
for the results of the n-time simulation,
under the constraint, each round of simulation needs to be calculated
Wherein
Is a fixed value, n is a round, so that s is calculated for each round, i.e. the calculation is repeated
To reduce the amount of computation, the invention defines
s 2 ＝S n V (n-1), at the start of the simulation,
and S 1 =0, starting from the second round, the i-th round updates the following:
2. a significant number constraint. This constraint ensures the accuracy of the simulation results. And recording the effective digit number of the decimal point of the simulation result as beta, wherein for one n-round Monte Carlo simulation, the beta meets the following conditions:
3. the number of failures constraints. The constraint ensures the validity of the simulation result. The reliability per unit time of some highly reliable software and hardware may reach 5 to 9, i.e. 99.999%, which would result in that
The value of (c) is extremely small, so that the simulation result can meet the first and second constraint conditions under the condition of few rounds, and the simulation is ended under the condition of insufficient rounds, which is contrary to the basic idea that the Monte Carlo simulation approaches the true probability through a large number of random tests. The fault quantity constraint of the ith component is recorded as Fi, i is more than or equal to 1 and less than or equal to m, m is the component quantity, the Fi can adopt an empirical value, each simulation round updates the accumulated fault quantity of the component according to the state of the component after the simulation round, the accumulated fault quantity is recorded as Fi, and the Fi needs to meet the requirements when the simulation is finished:
f i ≥F i
combining the above three constraints, the simulation of the present invention will end when the following conditions are simultaneously satisfied:
the above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and adaptations to those skilled in the art without departing from the principles of the present invention should also be considered as within the scope of the present invention.
Claims (6)
1. A cloud application reliability assessment method considering edge cloud cooperation is characterized by comprising the following steps: the method comprises the following steps:
step S1: establishing a directed acyclic graph model based on the hierarchical dependency relationship among the components;
step S2: calculating the reliability of each component based on the historical fault information of the component;
and step S3: performing reliability simulation on the fault distribution based on the components by adopting Monte Carlo simulation;
and step S4: determining the state of the cloud application according to the simulation state of each component in the directed acyclic graph model;
step S5: repeating the step S3 and the step S4 until the ending condition is met;
wherein the ending condition is as follows:
in the formula phi -1 The method is characterized in that the method is an inverse function of a standard normal distribution cumulative distribution function, alpha is a confidence interval of a simulation result, s is a standard deviation of the simulation result, beta is an effective digital digit after a decimal point of the simulation result, fi is the accumulated fault quantity updated according to the state of the component after each simulation, and Fi is the fault quantity constraint of the ith component.
2. The cloud application reliability evaluation method considering edge-cloud cooperation according to claim 1, characterized in that: the step S1 includes the steps of:
s1.1, analyzing related components in the cloud application, and decomposing the cloud application into a plurality of sub-modules, wherein each sub-module represents a service;
s1.2, determining the dependency relationship among the components in the step S1.1 on the basis that one service needs a calculation result based on another service to complete a task and the service depends on the service providing the calculation result;
and S1.3, establishing a directed acyclic graph model according to the components and the dependency relationship thereof.
3. The cloud application reliability evaluation method considering edge cloud coordination according to claim 1 or 2, characterized in that: the specific steps of calculating the reliability in step S2 are as follows:
step S2.1, taking the unit time as one day, and if a component has a faults averagely in one day, the fault rate per unit time is λ = a;
step S2.1, according to the formula r = e -λ ＝e -a The unit time reliability r is calculated.
4. The cloud application reliability evaluation method considering edge cloud coordination according to claim 1 or 2, characterized in that: the specific steps of the simulation in the step S3 are as follows:
s3.1, based on the reliability of each component, generating a random number representing the reliability of each component according to the uniform distribution U [0, 1);
step S3.2, recording the reliability value of the ith component as di, and recording the state as Si epsilon [0,1], wherein the method comprises the following steps:
wherein λ is i For the failure rate per unit time of the ith component, si =0 indicates that the state of the ith component in the unit time is a failure state, and correspondingly, si =1 indicates that the state of the ith component in the unit time is a normal state.
5. The cloud application reliability evaluation method considering edge cloud cooperation according to claim 1 or 2, characterized in that: the specific steps of determining the state of the cloud application in step S4 are as follows:
s4.1, judging the states of all service instances according to rules;
s4.2, judging the state of each service according to the state of the service instance;
and S4.3, after each service state is determined, determining the application state according to the dependency relationship among the services.
6. The cloud application reliability evaluation method considering edge cloud coordination according to claim 1 or 2, characterized in that: the condition of meeting the end in the step S5 is that the simulation structure meets three kinds of constraints simultaneously, and the three kinds of constraints are respectively a confidence interval constraint, an effective number constraint and a fault number constraint of the simulation result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010532373.9A CN111753411B (en) | 2020-06-12 | 2020-06-12 | Cloud application reliability evaluation method considering edge cloud cooperation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010532373.9A CN111753411B (en) | 2020-06-12 | 2020-06-12 | Cloud application reliability evaluation method considering edge cloud cooperation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111753411A CN111753411A (en) | 2020-10-09 |
| CN111753411B true CN111753411B (en) | 2022-12-20 |
Family
ID=72675112
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010532373.9A Expired - Fee Related CN111753411B (en) | 2020-06-12 | 2020-06-12 | Cloud application reliability evaluation method considering edge cloud cooperation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111753411B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104977022A (en) * | 2014-04-04 | 2015-10-14 | 西北工业大学 | Multi-target track system performance evaluation simulation method |
| CN107358352A (en) * | 2017-07-05 | 2017-11-17 | 国网山东省电力公司电力科学研究院 | Model in Reliability Evaluation of Power Systems system and method based on Monte Carlo simulation |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9577910B2 (en) * | 2013-10-09 | 2017-02-21 | Verisign, Inc. | Systems and methods for configuring a probe server network using a reliability model |
-
2020
- 2020-06-12 CN CN202010532373.9A patent/CN111753411B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104977022A (en) * | 2014-04-04 | 2015-10-14 | 西北工业大学 | Multi-target track system performance evaluation simulation method |
| CN107358352A (en) * | 2017-07-05 | 2017-11-17 | 国网山东省电力公司电力科学研究院 | Model in Reliability Evaluation of Power Systems system and method based on Monte Carlo simulation |
Non-Patent Citations (3)
| Title |
|---|
| Evaluation and design of highly reliable and highly utilized cloud computing systems;B. Snyder 等;《 Journal of Cloud Computing》;20150528;1-30页 * |
| On the Relatively Small Impact of Deep Dependencies on Cloud Application Reliability;X. Wang 等;《2017 IEEE 10th International Conference on Cloud Computing》;20170625;第528-535页 * |
| 基于Gibbs采样与概率分布估计的移动云数据存储;李又玲等;《计算机工程》;20170115(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111753411A (en) | 2020-10-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112650762B (en) | Data quality monitoring method and device, electronic equipment and storage medium | |
| US7702788B2 (en) | Method and apparatus for performance and policy analysis in distributed computing systems | |
| US20040010733A1 (en) | System and method for fault identification in an electronic system based on context-based alarm analysis | |
| CN113010393A (en) | Fault drilling method and device based on chaotic engineering | |
| CN117078048B (en) | Digital twinning-based intelligent city resource management method and system | |
| CN109684320B (en) | Method and equipment for online cleaning of monitoring data | |
| CN110083518B (en) | AdaBoost-Elman-based virtual machine software aging prediction method | |
| Nazari Cheraghlou et al. | New fuzzy-based fault tolerance evaluation framework for cloud computing | |
| CN108415819B (en) | Hard disk fault tracking method and device | |
| US20190005169A1 (en) | Dynamic Design of Complex System-of-Systems for Planning and Adaptation to Unplanned Scenarios | |
| CN114282011A (en) | Knowledge graph construction method and device, and graph calculation method and device | |
| CN112506802B (en) | Test data management method and system | |
| CN114172819A (en) | Demand resource prediction method, system, electronic device and storage medium for NFV network element | |
| CN111753411B (en) | Cloud application reliability evaluation method considering edge cloud cooperation | |
| Bosse et al. | Predicting availability and response times of IT services | |
| CN113541986B (en) | Fault prediction method and device for 5G slice and computing equipment | |
| Rabah et al. | Performability evaluation of multipurpose multiprocessor systems: the" separation of concerns" approach | |
| CN117580046A (en) | Deep learning-based 5G network dynamic security capability scheduling method | |
| Ivanchenko et al. | Availability assessment of a cloud server system: comparing Markov and Semi-Markov models | |
| CN116108622A (en) | Rail transit signal system reliability analysis method and device and electronic equipment | |
| CN113469377B (en) | Federal learning auditing method and device | |
| Gomes et al. | Performability model for assessing nosql dbms consistency | |
| CN115348293A (en) | Intelligent control remote operation and maintenance method and platform for industrial internet equipment | |
| CN114860615A (en) | Rule automatic testing method and device, electronic equipment and storage medium | |
| CN114358655A (en) | Method and device for generating recommendation scheme |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20221220 |