CN105959412A - Cloud service resource allocation analysis method based on queue mining - Google Patents
Cloud service resource allocation analysis method based on queue mining Download PDFInfo
- Publication number
- CN105959412A CN105959412A CN201610504890.9A CN201610504890A CN105959412A CN 105959412 A CN105959412 A CN 105959412A CN 201610504890 A CN201610504890 A CN 201610504890A CN 105959412 A CN105959412 A CN 105959412A
- Authority
- CN
- China
- Prior art keywords
- resource
- queue
- service
- client
- resource allocation
- 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
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a cloud service resource allocation analysis method based on queue mining, which is applicable for solving the allocation of complicated resources in cloud service processes under the condition of interaction between customers and resources. The method is characterized in that in consideration of the interaction between the customers and the resources, customer service logs and resource service logs are established; the two types of service logs are firstly directly utilized to carry out resource matching through customer types and resource types, and to carry out optimization through the loss function L(d,pi(x)) to obtain a primary resource allocation plan pi(X); a queue mining method based on the resource service logs is then put forward based on the customer service logs, so as to obtain customer queue information, including customer queue length q(t), customer delay time h(t), and queue abandoning time (the symbol is as shown in the description), etc., and decision information for resource allocation, including service acceptance time of the customers (the symbol is as shown in the description) and resource service status F<STA>; on such a basis, a resource allocation algorithm based on service process attributes is put forward for resource re-allocation, and when the queue length and the delay time reach thresholds, a final cloud service resource allocation plan is obtained through optimized analysis.
Description
Technical field
The invention belongs under cloud computing environment Service Source scheduling field, relate to exist the business procedure of time delay excavate and
The distribution of related resource, is particularly well-suited under cloud computing environment, there is resource and the resource allocation problem under client's interaction scenario.
Background technology
Under cloud computing environment, service process is a kind of special business procedure, and it is to include cloud platform resource and client
Mutual business procedure.Along with society progress, cloud service process analysis by wider application in increasing field,
Be simultaneous for the research of cloud service process improving service quality, improve enterprises service efficiency and the aspect such as reduce service cost all
Play an important role.Current is to start with from event log about service process research method major part, is primarily upon client side
Be analyzed, pay close attention to the research in terms of the resource under cloud computing platform less, lack client and cloud resource mutual after analysis side
Method, causes service process to analyze comprehensive, and then can not obtain during cloud service reasonably Resource Allocation Formula.Also have comprehensive
The analysis that client and resource are carried out, but complexity is relatively big, and to hardly result under fairness and allocative efficiency are tested and assessed suitable simultaneously
Allocative decision.
Therefore, the research of the cloud service process interacted in the face of client and resource, it is necessary in existing process model mining technology
On the basis of, it is considered to the impact on cloud service process of the cloud platform resource side.Queue Mining Technology based on resource service daily record is proposed
Art, and applied in the research of resource distribution.Utilize and excavated the queuing message and decision information obtained by queue, to just
Sub-distribution scheme is supplemented, and obtains final allocative decision.
Summary of the invention
The technical problem to be solved is: provides a kind of domain constraint first passing through cloud service system, carries out cloud
The first sub-distribution of Service Source, it is ensured that system medium cloud resource distributional equity, then according to customer service in cloud service flow process
Daily record and resource service daily record carry out queue excavation, obtain the queuing message of service process: client's queue length q (t), Gu Keshi
H (t) and the decision information of resource distribution between time delay: client accepts service timeWith resource service state FSTA, utilize these teams
Column information and decision information carry out second time and distribute resource, improve the efficiency of resource distribution, and the most comprehensive two sub-distribution obtain
The method of last Resource Allocation Formula.
For solving above technical problem, the present invention adopts the following technical scheme that:
First, customer service daily record C.S-Log=(S, G, α in definition cloud service flow processC={ τ, η, ε }) and resource service
Daily record R.S-Log=(S, G, αR={ τ, η, σ, φ, δ }), wherein S represents the set of Service events (including client and resource)；G
Represent the set of service path；αCRepresent that the community set of client includes: timestamp attribute τ, client category attribute η and client's clothes
Business status attribute ε；αRRepresent resource community set include: timestamp attribute τ, client category attribute η, resource class attribute σ,
Resource status attribute φ and resource status converting attribute δ.The timestamp attribute of the client in resource service daily record and category attribute
Being different from client's daily record, it refers to accept the attribute of the client of resource service.
Secondly, utilize two class serve log to carry out queue based on cloud service flow process to excavate.Obtain based on customer service daily record
Queuing message to service process: client's queue length
Client's time delayWhereinClient
Abandon Queue timeBased on resource service daily record obtain resource distribution decision information: resource state information
FSTA=φ [argmin (t-τ (S))], whereinResource is serve
TimeLast comprehensive queuing message and decision information provide the decision variable of resource distribution:
Finally, the way of resource distribution in cloud service flow process is provided based on queue digging technology.First in strict accordance with client
Category attribute with resource carries out the coupling of resource, and Resource Allocation Formula is optimized by recycling loss function L (d, π (x)),
Obtain the Resource Allocation Formula π (X) of angle based on domain constraint.Then queue is utilized to excavate the queuing message and decision-making obtained
Information carries out second time and distributes, and is the attribute intrinsic based on client in serve log including Part I: belong to according to client's classification
Property η (s) and resource service status attribute FSTACarrying out just sub-distribution, the distribution of Part II is to excavate, based on queue, the team obtained
Column information is carried out: client's decay timeWhen reaching threshold value, preferentially carry out resource distribution etc..Finally comprehensively obtain resource
Allocative decision.
Accompanying drawing explanation
Fig. 1 is the structure chart of the procedural model of the present invention.
Fig. 2 is the flow chart of the queue excavation of the present invention.
Fig. 3 is the flow chart of the resource distribution of the present invention.
Detailed description of the invention
The present invention proposes the cloud service process resources distribution method excavated based on queue.The most strictly enter according to consumer type
The coupling of row resource, and carry out inspection optimization with loss function L (d, π (x)), obtain the first sub-distribution about domain constraint angle
Scheme π (X), it may ensure that resource distributional equity；Secondly according to serve log, behavior profile digging technology is used to obtain
Queuing message q (t) of cloud service flow process, h (t) and decision information client accept service timeWith resource service state FSTA, fortune
Carry out second time by these information to distribute, it is considered to the impact that in cloud service flow process, resource is distributed by queuing message, improve resource and divide
The efficiency joined.Comprehensive two sub-distribution obtain final Resource Allocation Formula.
Below in conjunction with accompanying drawing, the present invention is further illustrated.
Fig. 1 is the structure chart of the procedural model of the present invention, carries out first including the constraint angle according to service system field
Secondary resource is distributed, then arranges logged sequence and obtain customer service daily record and resource service daily record, based on queue viewpoint to service thing
Part carries out queue excavation, obtains the queuing message in cloud service flow process and the decision information of resource distribution, distributes in first resource
On the basis of scheme, carry out second time resource distribution, obtain final distribution method.
Fig. 2 is the flow chart of the queue excavation of the present invention, includes specifically, according to service procedure duality definition cloud clothes
Client's event in business flow process and resource event, and arrange and obtain corresponding serve log, recycling queue digging technology is to thing
Part is analyzed obtaining queuing message: long delay h (t) that client's queue length q (t), client wait in queue and abandoning
Queue timeDecision information: resource service timeWith resource service state FSTA。
Fig. 3 is the flow chart of the resource allocation methods that the present invention excavates based on queue, and it comprises and is based respectively on customer service
The queue of daily record and resource service daily record is excavated, and sorts out queuing message and decision information, at the beginning of obtaining based on domain constraint
On the basis of sub-distribution scheme π (X), the long delay h waited in queue according to client's queue length q (t), client respectively
(t) and resource service state FSTACarry out secondary distribution.Finally obtain the Resource Allocation Formula of this cloud service flow process.
Claims (3)
1. the cloud service resource allocation methods excavated based on queue, including queue method for digging based on serve log, based on team
The resource allocation methods that row excavate, it is characterised in that: according to resource in cloud service flow process and client mutual in the case of, based on Gu
Visitor's serve log obtains about queue length, customer service time delay and the queuing message abandoning Queue time；Based on resource service
Daily record carries out queue excavation, obtains about the queue letter accepting service time and resource service state in service process about client
Breath.Resource allocation methods includes two steps, the first time distribution carried out based on domain constraint, it is ensured that distributional equity, further according to
The queuing message obtained is excavated in queue, carries out second time resource distribution.Final allocative decision is obtained in conjunction with two sub-distribution, it is ensured that point
Allocative efficiency is improved while joining fairness.
The queue of cloud service daily record the most according to claim 1 is excavated, it is characterised in that: consider client in cloud service flow process
The situation mutual with resource, serve log being arranged is customer service daily record and resource service daily record, then the serve log obtained
Carry out queue excavation, obtain client's queue length q (t), client's decay time h (t) and abandon Queue timeDistribute for resource
Queuing message is provided；Consider resource side's impact on client's queue in cloud service flow process, obtain accepting service time about clientWith resource service state FSTA, provide decision information for resource distribution.
The resource allocation methods excavated based on queue the most according to claim 1, it is characterised in that: it is primarily based on field about
Bundle condition, carries out resource matched according to the classification of client and resource, and recycling loss function L (d, π (x)) is optimized and is provided
Allocative decision π (X) that source is first；Excavate the queuing message obtained further according to queue, carry out second time and distribute: utilize queue length
The queuing messages such as the threshold value with customer service time delay, are allocated in conjunction with decision informations such as resource service states, obtain optimization
Cloud service Resource Allocation Formula.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610504890.9A CN105959412A (en) | 2016-06-29 | 2016-06-29 | Cloud service resource allocation analysis method based on queue mining |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610504890.9A CN105959412A (en) | 2016-06-29 | 2016-06-29 | Cloud service resource allocation analysis method based on queue mining |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105959412A true CN105959412A (en) | 2016-09-21 |
Family
ID=56902723
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610504890.9A Pending CN105959412A (en) | 2016-06-29 | 2016-06-29 | Cloud service resource allocation analysis method based on queue mining |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105959412A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102682396A (en) * | 2007-11-07 | 2012-09-19 | 谷歌公司 | Method and systme used for resouce distribution |
| CN102917077A (en) * | 2012-11-20 | 2013-02-06 | 无锡城市云计算中心有限公司 | Resource allocation method in cloud computing system |
| WO2013087610A1 (en) * | 2011-12-13 | 2013-06-20 | Siemens Aktiengesellschaft | Device and method for the dynamic load management of cloud services |
| CN105138445A (en) * | 2015-08-17 | 2015-12-09 | 安徽理工大学 | New method for mining invisible tasks in service process based on probability behavior relationship |
-
2016
- 2016-06-29 CN CN201610504890.9A patent/CN105959412A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102682396A (en) * | 2007-11-07 | 2012-09-19 | 谷歌公司 | Method and systme used for resouce distribution |
| WO2013087610A1 (en) * | 2011-12-13 | 2013-06-20 | Siemens Aktiengesellschaft | Device and method for the dynamic load management of cloud services |
| CN102917077A (en) * | 2012-11-20 | 2013-02-06 | 无锡城市云计算中心有限公司 | Resource allocation method in cloud computing system |
| CN105138445A (en) * | 2015-08-17 | 2015-12-09 | 安徽理工大学 | New method for mining invisible tasks in service process based on probability behavior relationship |
Non-Patent Citations (2)
| Title |
|---|
| ARIK SENDEROVICH: "Mining Resource Scheduling Protocols", 《INTEMATIONALCONFERENCE ON BUSINESS PROCESS MANAGEMENT，SPRINGER INTERNATIONAL PUBLISHING》 * |
| ARIK SENDEROVICH: "Queue Mining for Delay Prediction in Multi-Class Service Processes", 《INFORMATION SYSTEMS》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Curi et al. | The Italian airport industry in transition: a performance analysis | |
| US20140257909A1 (en) | Estimating project cost | |
| CN101540736A (en) | Method for distributing task by using instant communication device | |
| CN114611008B (en) | User service strategy determination method and device based on federal learning and electronic equipment | |
| CN106126519A (en) | The methods of exhibiting of media information and server | |
| CN110533477A (en) | A kind of intelligent analysis method and system based on big data | |
| Lee et al. | Escaping the Middle‐Income Trap in S outheast A sia: Micro Evidence on Innovation, Productivity, and Globalization | |
| Iwan et al. | The integrated approach to adaptation of good practices in urban logistics based on the Szczecin example | |
| US20150278718A1 (en) | Systems and methods for communication sharing in a relationship management system | |
| CN101094202A (en) | Method and system for receiving events in management system through instant communication tool | |
| Li et al. | Owner-dominated building information modeling and lean construction in a megaproject | |
| Becker et al. | Shared service center vs. shared service network: A multiple case study analysis of factors impacting on shared service configurations | |
| CN101840423B (en) | Bill accuracy auditing system based on pair trading principle and data mining technology | |
| CN105959412A (en) | Cloud service resource allocation analysis method based on queue mining | |
| CN103093294A (en) | Enterprise evaluation service system and method based on cloud computing | |
| CN112435017A (en) | Modeling and construction method for cross-organization business process interaction | |
| CN101471846A (en) | Network event analysis and management system | |
| CN114091057B (en) | Federal learning safety audit method and device based on model simulation | |
| Zhang et al. | A correlation context‐aware approach for composite service selection | |
| Marpaung et al. | Quality of Service from a Network when Using Youtube Application | |
| Li et al. | A tree-network model for mining short message services seed users and its empirical analysis | |
| Sawatani et al. | Service research model for value co-creation | |
| El Hamali et al. | Towards a reeingeniering of business process management lifecycle by the use of social networks | |
| Niehaves et al. | Market, network, hierarchy: emerging mechanisms of governance in business process management | |
| Iqbal et al. | Frequently occurring risks for IT outsourcing projects |
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: 20160921 |