CN109040214A - A kind of service arrangement method that reliability enhances under cloud environment - Google Patents

A kind of service arrangement method that reliability enhances under cloud environment Download PDF

Info

Publication number
CN109040214A
CN109040214A CN201810824345.7A CN201810824345A CN109040214A CN 109040214 A CN109040214 A CN 109040214A CN 201810824345 A CN201810824345 A CN 201810824345A CN 109040214 A CN109040214 A CN 109040214A
Authority
CN
China
Prior art keywords
services
node
sub
tail
head
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.)
Granted
Application number
CN201810824345.7A
Other languages
Chinese (zh)
Other versions
CN109040214B (en
Inventor
周傲
王尚广
孙其博
李静林
原媛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing University of Posts and Telecommunications
Original Assignee
Beijing University of Posts and Telecommunications
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing University of Posts and Telecommunications filed Critical Beijing University of Posts and Telecommunications
Priority to CN201810824345.7A priority Critical patent/CN109040214B/en
Publication of CN109040214A publication Critical patent/CN109040214A/en
Application granted granted Critical
Publication of CN109040214B publication Critical patent/CN109040214B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/142—Network analysis or design using statistical or mathematical methods
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
    • H04L41/5041—Network service management, e.g. ensuring proper service fulfilment according to agreements characterised by the time relationship between creation and deployment of a service
    • 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
    • H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
    • 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
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/1004—Server selection for load balancing
    • H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
    • 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
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/1004—Server selection for load balancing
    • H04L67/101—Server selection for load balancing based on network conditions

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Algebra (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Physics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Pure & Applied Mathematics (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Hardware Redundancy (AREA)

Abstract

This application discloses the service arrangement methods that reliability under a kind of cloud environment enhances, complex services are modeled as graph model first by this method, the host node of figure represents the sub-services of complex services, the side of figure represents message transmission rate between sub-services, it then is that each sub-services select corresponding Reliability Strategy according to the characteristic of each sub-services itself, and it is based on selected Reliability Strategy, increase key point memory node or redundant node for each host node in graph model, carry out the service arrangement of corresponding reliability enhancing, at the same time, the application has taken into account the otherness that each sub-services guaranteed reliability brings resource occupying again, it is occupied so as to minimize guaranteed reliability's bring additional network resources.

Description

A kind of service arrangement method that reliability enhances under cloud environment
Technical field
This application involves computer application technology, in particular to the service arrangement of reliability enhancing under a kind of cloud environment Method.
Background technique
When disposing complex services in cloud computing platform, needs for each sub-services of complex services to be deployed to and be placed in object In virtual machine on reason machine, is cooperated by virtual machine and complete the execution of complex task.Since failure event leads to virtual resource Non-continuous occupancy so that reliability of service decline, it usually needs take key point store or virtual machine redundancy place etc. measures Improve reliability of service.Due to ignoring, sub-services calculate the otherness of feature to current reliability strategy and different sub-services exist Otherness in terms of reliability requirement, and be each sub-services using identical Reliability Strategy, accordingly, it is difficult to obtain it is optimal can Effect is ensured by property.
Summary of the invention
In view of this, the main purpose of the present invention is to provide service department's management sides that reliability under a kind of cloud environment enhances Method minimizes Reliability Strategy bring additional network resources expense while guaranteeing complex services reliability.
This application discloses the service arrangement methods that reliability under a kind of cloud environment enhances, comprising:
A: complex services are modeled as graph model, the host node of figure represents the sub-services of the complex services, the side generation of figure Message transmission rate between table sub-services；
B: traversal each sub-services of complex services, according to the characteristic of each sub-services of complex services for its select key point storage or Person's resource redundancy is as its Reliability Strategy；
C: being based on the selected Reliability Strategy, be extended to the graph model, and respectively each host node increases Key point memory node or redundant node；
D: the side of connection host nodes all in graph model is ranked up by the descending of message transmission rate, successively by side Node is mapped in the physical machine of cloud data center end to end, and distributes virtual machine on demand in physical machine, and node is corresponding Sub-services are deployed on virtual machine；
E: in graph model it is all connection host nodes and key point memory node side by message transmission rate descending into The tail node on side, is successively mapped in the physical machine of cloud data center by row sequence, and distributes in the physical machine for key point Memory space；
F: side remaining in graph model is ranked up by the descending of message transmission rate, tail node is successively mapped to cloud In the physical machine of data center, and virtual machine is distributed for redundancy services node in the physical machine.
Preferably, the b includes:
All sub-services for traversing the complex services, for the sub-services s currently traversedi, institute is executed according to without interruption Take time T_exe (si), the sub-services are carried out with T_cp (s the time required to the storage of key pointi) and progress key point storage Acceptable cycle T _ interval (si), ineffective time percentage perc is calculated based on following formula:
Then, perc is compared with acceptable degree threshold value delta, if perc is greater than delta, for the sub-services Select resource redundancy as its Reliability Strategy；If perc is less than or equal to perc, key point is selected to deposit for the sub-services Storage is used as its Reliability Strategy.
Preferably, a includes:
A1: an empty graph model G (V, E) is constructed for complex services, wherein node set V and line set E is empty；
A2: for complex services S={ s1,s2...si...smax(s)Each sub-services siIncrease a node of graph The type of node is host node, and the attribute of node includes demand of the sub-services to cpu resource, memory source and disk resource, And the node is added in the node set V of G；
A3: for there is the sub-services s of data interactioniAnd sj, it is its corresponding nodeWithIncrease a lineAnd the side is increased in the line set E of G, while type be main while, the attribute on side is sub-services siAnd sjIt Between message transmission rate.
Preferably, the c includes:
C1: traversal selects key point to store the sub-services as its Reliability Strategy, to the sub-services s currently traversediInstitute Corresponding host nodeIncrease corresponding key point memory nodeThe type of node is key point node, the attribute of node To store memory space required for crucial dot file
Then, it is calculated based on following formulaWithMessage transmission rate:
And increase a connectionWithSideWhile type be storage while, the attribute on side isWith Message transmission rate, and willIt is added in the node set V of the graph model, it willIt is added to the artwork In the line set E of type；
C2: traversal selects sub-services of the resource redundancy as its Reliability Strategy, to the sub-services s currently traversediInstitute is right The host node answeredIncrease corresponding redundant nodeAlso, for any host node in set VIf its withIt is connected directly, increases a lineThe attribute on side isWithMessage transmission rate, the type on side is superfluous Yu Bian, willIt is added in set E.
Preferably, the d includes:
D1: main side all in the line set E of the graph model is added in list Ranked_EM [], to list Ranked_EM [] sorts from high to low by message transmission rate；
D2: traversal of lists Ranked_EM [], to the side e currently traversedi, obtain eiHead node head (ei) and tail node tail(ei)；
D3: if head (ei) and tail (ei) corresponding sub-services not yet dispose, then the subnet of data center is searched for, such as There are two physical machines for fruitWithSubnet, meetSurplus resources be greater than head (ei) resource requirement, Surplus resources be greater than tail (ei) resource requirement, then existIt is upper to press head (ei) resource requirement distribute virtual machine, will head(ei) corresponding sub-services are deployed on the virtual machine,It is upper to press tail (ei) resource requirement distribute virtual machine, will tail(ei) corresponding sub-services are deployed on the virtual machine, return step d2；
D4: if not finding the subnet of the condition of satisfaction in step d3, searching for the cluster of data center, if there is having Two physical machinesWithCluster, meetSurplus resources be greater than head (ei) resource requirement,Remaining money Source is greater than tail (ei) resource requirement, then existIt is upper to press head (ei) resource requirement distribute virtual machine, by head (ei) right The sub-services answered are deployed on the virtual machine,It is upper to press tail (ei) resource requirement distribute virtual machine, by tail (ei) right The sub-services answered are deployed on the virtual machine, return step d2；
D5: if not meeting the cluster of condition in step d4, two physical machines are found from entire data centerWithMeetSurplus resources be greater than head (ei) resource requirement,Surplus resources be greater than tail (ei) resource need It asks, by head (ei) corresponding sub-services are deployed totail(ei) corresponding sub-services are deployed toReturn step d2；
D6: if only head (ei) not yet dispose, to dispose tail (ei) physical machine centered on, from closely to remote search Surplus resources are greater than head (ei) resource requirement physical machine, in the physical machine press head (ei) to the demand assignment of resource Virtual machine, by head (ei) corresponding sub-services are deployed on the virtual machine, return step d2；
D7: if only tail (ei) not yet dispose, to dispose head (ei) physical machine centered on, from closely to remote search Surplus resources are greater than tail (ei) resource requirement physical machine, in the physical machine press tail (ei) to the demand assignment of resource Virtual machine, by tail (ei) corresponding sub-services are deployed on the virtual machine, return step d2.
Preferably, the e includes:
E1: storage side all in the line set E of the graph model is added in list Ranked_EC [], right Ranked_EC [] sorts from high to low by message transmission rate；
E2: traversal Ranked_EC [], to the side e currently traversedi, obtain head node head (ei) and tail node tail (ei)；
E3: with head (ei) deployment physical machine centered on, from being closely greater than tail (e to remote search surplus resourcesi) resource The physical machine is assigned as head (e by the physical machine of demandi) key point store physical machine.
Preferably, the f includes:
F1: redundancy sides all in the line set E of the graph model are added in list Ranked_ER [], to Ranked_ ER [] sorts from high to low by message transmission rate；
F2: traversal Ranked_ER [], to the side e currently traversedi, obtain head node head (ei) and tail node tail (ei)；
F3: with head (ei) deployment physical machine centered on, from being closely greater than tail (e to remote search surplus resourcesi) resource need The physical machine asked presses tail (e in the physical machinei) to the demand assignment virtual machine of resource, by tail (ei) corresponding sub-services It is deployed on the virtual machine.
As seen from the above technical solution, the application propose cloud environment under reliability enhance service arrangement method pass through by Complex services are modeled as graph model, and the host node of figure is enabled to represent the sub-services of complex services, and the side of figure represents number between sub-services It is then that each sub-services select corresponding Reliability Strategy, and are based on institute according to the characteristic of each sub-services itself according to transmission rate The Reliability Strategy of selection is that each host node in graph model increases key point memory node or redundant node, carries out corresponding Reliability enhancing service arrangement, at the same time, the application has taken into account each sub-services guaranteed reliability again and has brought Internet resources The otherness of occupancy occupies so as to minimize guaranteed reliability's bring additional network resources.
Detailed description of the invention
The flow diagram for the service arrangement method that Fig. 1 enhances for reliability under the cloud environment of the embodiment of the present invention one.
Specific embodiment
It is right hereinafter, referring to the drawings and the embodiments, for the objects, technical solutions and advantages of the application are more clearly understood The application is described in further detail.
To solve the problems of prior art, the application proposes the service arrangement that reliability enhances under a kind of cloud environment Method, this method are that each sub-services select corresponding Reliability Strategy according to the characteristic of each sub-services itself, carry out reliability increasing Strong service arrangement takes into account the otherness that each sub-services guaranteed reliability brings resource occupying again, to minimize reliability guarantor Hinder bring additional network resources to occupy.
This application discloses the service arrangement methods that reliability under a kind of cloud environment enhances, method includes the following steps:
Step 1: complex services being modeled as graph model, the host node of figure represents the sub-services of the complex services, figure While representing message transmission rate between sub-services.
Step 2: traversal each sub-services of complex services select key point to deposit according to the characteristic of each sub-services of complex services for it Storage or resource redundancy are as its Reliability Strategy.
Step 3: the Reliability Strategy selected based on step 2 is extended the graph model of step 1, respectively each master Node increases key point memory node or redundant node.
Step 4: the side of connection host nodes all in graph model being ranked up by the descending of message transmission rate, successively will The node end to end on side is mapped in the physical machine of cloud data center, and distributes virtual machine on demand in physical machine, by node pair The sub-services answered are deployed on virtual machine.
Step 5: the drop of message transmission rate is pressed to the side of connection host nodes all in graph model and key point memory node Sequence is ranked up, and successively the tail node on side is mapped in the physical machine of cloud data center, and is key point in the physical machine Distribute memory space.
Step 6: side remaining in graph model being ranked up by the descending of message transmission rate, successively maps tail node Virtual machine is distributed onto the physical machine of cloud data center, and in the physical machine for redundancy services node.
It is being to obtain reliability enhancing under a kind of cloud environment according to step 1 to 6 after all nodes all distributes virtual resource Service arrangement.The reliability of complex services under cloud environment can be improved using the present invention.
The flow diagram for the service arrangement method that Fig. 1 enhances for reliability under the cloud environment of the embodiment of the present invention one, such as Shown in Fig. 1, this method is mainly comprised the steps that
Step a: complex services are modeled as graph model, the host node of figure represents the sub-services of the complex services, figure While representing message transmission rate between sub-services.
A kind of specific method modeled is set forth below:
A1: an empty graph model G (V, E) is constructed for complex services, wherein node set V and line set E is empty.
A2: for complex services S={ s1,s2...si...smax(s)Each sub-services siIncrease a node of graph The type of node is host node, and the attribute of node includes the sub-services to resources such as cpu resource, memory source and disk resources Demand, and the node is added in the node set V of G.
A3: for there is the sub-services s of data interactioniAnd sj, it is its corresponding nodeWithIncrease a lineAnd the side is increased in the line set E of G, while type be main while, the attribute on side is sub-services siAnd sjIt Between message transmission rate.
Step b: traversal each sub-services of complex services select key point to deposit according to the characteristic of each sub-services of complex services for it Storage or resource redundancy are as its Reliability Strategy.
A kind of method of preferable selection Reliability Strategy are as follows:
All sub-services for traversing complex services, for the sub-services s currently traversedi, taken according to being executed without interruption Between T_exe (si), the sub-services are carried out with T_cp (s the time required to the storage of key pointi) and carry out key point storage can Receive cycle T _ interval (si), ineffective time percentage perc is calculated based on following formula:
Then, perc is compared with acceptable degree threshold value delta, if perc is greater than delta, for the sub-services Select resource redundancy as its Reliability Strategy, it may be assumed that by sub-services siIt is added redundancy strategy list RD []；If perc be less than or Person is equal to perc, then selects key point storage as its Reliability Strategy for the sub-services, it may be assumed that by sub-services siIt is added crucial Point storage strategy list CP [].
Here, redundancy strategy list RD [] and key point storage strategy list CP [] by way of example only, in practical application In other modes can also be taken to record the sub-services for selecting resource redundancy or key point to store as Reliability Strategy respectively.
Step c: the Reliability Strategy selected according to step b is extended graph model obtained in step a, respectively Each host node (that is: each sub-services corresponding node) increases key point memory node or redundant node, obtains new artwork Type.A kind of preferable implementation are as follows:
C1: traversal key point storage strategy list CP [], to the sub-services s currently traversediCorresponding host node Increase corresponding key point memory nodeThe type of node is key point node, and the attribute of node is to store crucial dot file Required memory space
Then, it is calculated based on following formulaWithMessage transmission rate:
And increase a connectionWithSideWhile type be storage while, the attribute on side isWith Message transmission rate, and willIt is added in V, it willIt is added in set E.
C2: traversal redundancy strategy list RD [], to the sub-services s currently traversediCorresponding host nodeIncrease phase The redundant node answeredAlso, for any host node in set VIf its withIt is connected directly, increases a lineThe attribute on side isWithMessage transmission rate, while type be redundancy while, willIt is added to In set E.
Step d: the connection side between the host node in graph model is ranked up by the descending of message transmission rate, successively The node end to end on side is mapped in the physical machine of cloud data center, and distributes virtual machine on demand in physical machine, by node Corresponding sub-services are deployed on virtual machine.A kind of preferable implementation are as follows:
D1: main side all in set E is added in list Ranked_EM [], to list Ranked_EM [] by number It sorts from high to low according to transmission rate.
D2: traversal of lists Ranked_EM [], to the side e currently traversedi, obtain eiHead node head (ei) and tail node tail(ei)。
D3: if head (ei) and tail (ei) corresponding sub-services not yet dispose, then the subnet of data center is searched for, such as There are two physical machines for fruitWithSubnet, meetSurplus resources be greater than head (ei) resource requirement, Surplus resources be greater than tail (ei) resource requirement, then existIt is upper to press head (ei) resource requirement distribute virtual machine, will head(ei) corresponding sub-services are deployed on the virtual machine,It is upper to press tail (ei) resource requirement distribute virtual machine, will tail(ei) corresponding sub-services are deployed on the virtual machine, return step d2.
D4: if not finding the subnet of the condition of satisfaction in step d3, searching for the cluster of data center, if there is having Two physical machinesWithCluster, meetSurplus resources be greater than head (ei) resource requirement,Remaining money Source is greater than tail (ei) resource requirement, then existIt is upper to press head (ei) resource requirement distribute virtual machine, by head (ei) right The sub-services answered are deployed on the virtual machine,It is upper to press tail (ei) resource requirement distribute virtual machine, by tail (ei) right The sub-services answered are deployed on the virtual machine, return step d2.
D5: if not meeting the cluster of condition in step d4, two physical machines are found from entire data centerWithMeetSurplus resources be greater than head (ei) resource requirement,Surplus resources be greater than tail (ei) resource need It asks, by head (ei) corresponding sub-services are deployed totail(ei) corresponding sub-services are deployed toReturn step d2.
D6: if only head (ei) not yet dispose, to dispose tail (ei) physical machine centered on, from closely to remote search Surplus resources are greater than head (ei) resource requirement physical machine, in the physical machine press head (ei) to the demand assignment of resource Virtual machine, by head (ei) corresponding sub-services are deployed on the virtual machine, return step d2.
D7: if only tail (ei) not yet dispose, to dispose head (ei) physical machine centered on, from closely to remote search Surplus resources are greater than tail (ei) resource requirement physical machine, in the physical machine press tail (ei) to the demand assignment of resource Virtual machine, by tail (ei) corresponding sub-services are deployed on the virtual machine, return step d2.
Step e: the descending of message transmission rate is pressed to the side between the host node in graph model and key point memory node It is ranked up, successively the tail node on side is mapped in the physical machine of cloud data center, and be key point point in the physical machine With memory space.A kind of preferable implementation are as follows:
E1: storage side all in set E is added in list Ranked_EC [], presses data to Ranked_EC [] Transmission rate sorts from high to low.
E2: traversal Ranked_EC [], to the side e currently traversedi, obtain head node head (ei) and tail node tail (ei)。
E3: with head (ei) deployment physical machine centered on, from being closely greater than tail (e to remote search surplus resourcesi) resource The physical machine is assigned as head (e by the physical machine of demandi) key point store physical machine.
Step f: side remaining in graph model is ranked up by the descending of message transmission rate, successively maps tail node Onto the physical machine of cloud data center, virtual machine is distributed for redundancy services node in the physical machine.A kind of preferably realization side Formula are as follows:
F1: redundancies all in set E side is added in list Ranked_ER [], is passed to Ranked_ER [] by data Defeated rate sorts from high to low.
F2: traversal Ranked_ER [], to the side e currently traversedi, obtain head node head (ei) and tail node tail (ei)。
F3: with head (ei) deployment physical machine centered on, from being closely greater than tail (e to remote search surplus resourcesi) resource need The physical machine asked presses tail (e in the physical machinei) to the demand assignment virtual machine of resource, by tail (ei) corresponding sub-services It is deployed on the virtual machine.
The foregoing is merely the preferred embodiments of the application, not to limit the application, all essences in the application Within mind and principle, any modification, equivalent substitution, improvement and etc. done be should be included within the scope of the application protection.

Claims (7)

1. a kind of service arrangement method that reliability enhances under cloud environment characterized by comprising
A: complex services are modeled as graph model, the host node of figure represents the sub-services of the complex services, and the side of figure represents son Data transmission between services rate；
B: traversal each sub-services of complex services select key point storage or money according to the characteristic of each sub-services of complex services for it Source redundancy is as its Reliability Strategy；
C: being based on the selected Reliability Strategy, be extended to the graph model, and respectively each host node increases crucial Point memory node or redundant node；
D: the side of connection host nodes all in graph model is ranked up by the descending of message transmission rate, successively end to end by side Node is mapped in the physical machine of cloud data center, and distributes virtual machine on demand in physical machine, by the corresponding sub- clothes of node Business is deployed on virtual machine；
E: the side of connection host nodes all in graph model and key point memory node is arranged by the descending of message transmission rate The tail node on side is successively mapped in the physical machine of cloud data center by sequence, and is key point distribution storage in the physical machine Space；
F: side remaining in graph model is ranked up by the descending of message transmission rate, tail node is successively mapped to cloud data In the physical machine at center, and virtual machine is distributed for redundancy services node in the physical machine.
2. the method according to claim 1, wherein the b includes:
All sub-services for traversing the complex services, for the sub-services s currently traversedi, required time is executed according to without interruption T_exe(si), the sub-services are carried out with T_cp (s the time required to the storage of key pointi) and carry out key point storage and connect By cycle T _ interval (si), ineffective time percentage perc is calculated based on following formula:
Then, perc is compared with acceptable degree threshold value delta, if perc is greater than delta, for sub-services selection Resource redundancy is as its Reliability Strategy；If perc is less than or equal to perc, key point storage is selected to make for the sub-services For its Reliability Strategy.
3. method according to claim 1 or 2, which is characterized in that a includes:
A1: an empty graph model G (V, E) is constructed for complex services, wherein node set V and line set E is empty；
A2: for complex services S={ s1,s2...si...smax(s)Each sub-services siIncrease a node of graphNode Type be host node, the attribute of node includes demand of the sub-services to cpu resource, memory source and disk resource, and will The node is added in the node set V of G；
A3: for there is the sub-services s of data interactioniAnd sj, it is its corresponding nodeWithIncrease a line And the side is increased in the line set E of G, while type be main while, the attribute on side is sub-services siAnd sjBetween data transmission speed Rate.
4. method according to claim 1 or 2, which is characterized in that the c includes:
C1: traversal selects key point to store the sub-services as its Reliability Strategy, to the sub-services s currently traversediCorresponding Host nodeIncrease corresponding key point memory nodeThe type of node is key point node, and the attribute of node is storage Memory space required for crucial dot file
Then, it is calculated based on following formulaWithMessage transmission rate:
And increase a connectionWithSideWhile type be storage while, the attribute on side isWithNumber According to transmission rate, and willIt is added in the node set V of the graph model, it willIt is added to the graph model In line set E；
C2: traversal selects sub-services of the resource redundancy as its Reliability Strategy, to the sub-services s currently traversediCorresponding master NodeIncrease corresponding redundant nodeAlso, for any host node in set VIf its withDirect phase Even, increase a lineThe attribute on side isWithMessage transmission rate, while type be redundancy while, willIt is added in set E.
5. method according to claim 1 or 2, which is characterized in that the d includes:
D1: main side all in the line set E of the graph model is added in list Ranked_EM [], to list Ranked_ EM [] sorts from high to low by message transmission rate；
D2: traversal of lists Ranked_EM [], to the side e currently traversedi, obtain eiHead node head (ei) and tail node tail (ei)；
D3: if head (ei) and tail (ei) corresponding sub-services not yet dispose, then the subnet of data center is searched for, if deposited There are two physical machinesWithSubnet, meetSurplus resources be greater than head (ei) resource requirement,It is surplus Remaining resource is greater than tail (ei) resource requirement, then existIt is upper to press head (ei) resource requirement distribute virtual machine, by head (ei) corresponding sub-services are deployed on the virtual machine,It is upper to press tail (ei) resource requirement distribute virtual machine, by tail (ei) corresponding sub-services are deployed on the virtual machine, return step d2；
D4: if not finding the subnet of the condition of satisfaction in step d3, searching for the cluster of data center, if there is there is two Physical machineWithCluster, meetSurplus resources be greater than head (ei) resource requirement,Surplus resources it is big In tail (ei) resource requirement, then existIt is upper to press head (ei) resource requirement distribute virtual machine, by head (ei) corresponding Sub-services are deployed on the virtual machine,It is upper to press tail (ei) resource requirement distribute virtual machine, by tail (ei) corresponding Sub-services are deployed on the virtual machine, return step d2；
D5: if not meeting the cluster of condition in step d4, two physical machines are found from entire data centerWithMeetSurplus resources be greater than head (ei) resource requirement,Surplus resources be greater than tail (ei) resource Demand, by head (ei) corresponding sub-services are deployed totail(ei) corresponding sub-services are deployed toReturn step d2；
D6: if only head (ei) not yet dispose, to dispose tail (ei) physical machine centered on, from closely to remote search it is remaining Resource is greater than head (ei) resource requirement physical machine, in the physical machine press head (ei) virtual to the demand assignment of resource Machine, by head (ei) corresponding sub-services are deployed on the virtual machine, return step d2；
D7: if only tail (ei) not yet dispose, to dispose head (ei) physical machine centered on, from closely to remote search it is remaining Resource is greater than tail (ei) resource requirement physical machine, in the physical machine press tail (ei) virtual to the demand assignment of resource Machine, by tail (ei) corresponding sub-services are deployed on the virtual machine, return step d2.
6. according to the method described in claim 4, it is characterized in that, the e includes:
E1: storage side all in the line set E of the graph model is added in list Ranked_EC [], to Ranked_EC [] sorts from high to low by message transmission rate；
E2: traversal Ranked_EC [], to the side e currently traversedi, obtain head node head (ei) and tail node tail (ei)；
E3: with head (ei) deployment physical machine centered on, from being closely greater than tail (e to remote search surplus resourcesi) resource requirement Physical machine, which is assigned as head (ei) key point store physical machine.
7. according to the method described in claim 4, it is characterized in that, the f includes:
F1: redundancy sides all in the line set E of the graph model are added in list Ranked_ER [], to Ranked_ER [] sorts from high to low by message transmission rate；
F2: traversal Ranked_ER [], to the side e currently traversedi, obtain head node head (ei) and tail node tail (ei)；
F3: with head (ei) deployment physical machine centered on, from being closely greater than tail (e to remote search surplus resourcesi) resource requirement Physical machine presses tail (e in the physical machinei) to the demand assignment virtual machine of resource, by tail (ei) deployment of corresponding sub-services On the virtual machine.
CN201810824345.7A 2018-07-25 2018-07-25 Service deployment method for enhancing reliability in cloud environment Active CN109040214B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810824345.7A CN109040214B (en) 2018-07-25 2018-07-25 Service deployment method for enhancing reliability in cloud environment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810824345.7A CN109040214B (en) 2018-07-25 2018-07-25 Service deployment method for enhancing reliability in cloud environment

Publications (2)

Publication Number Publication Date
CN109040214A true CN109040214A (en) 2018-12-18
CN109040214B CN109040214B (en) 2020-07-17

Family

ID=64645027

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810824345.7A Active CN109040214B (en) 2018-07-25 2018-07-25 Service deployment method for enhancing reliability in cloud environment

Country Status (1)

Country Link
CN (1) CN109040214B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109526012A (en) * 2019-01-24 2019-03-26 重庆邮电大学 A kind of LoRaWAN network spreading factor distribution method based on reliability

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102768670A (en) * 2012-05-31 2012-11-07 哈尔滨工程大学 Webpage clustering method based on node property label propagation
CN104579850A (en) * 2015-01-27 2015-04-29 北京邮电大学 Quality of service (QoS) prediction method for Web service under mobile Internet environment
CN104809168A (en) * 2015-04-06 2015-07-29 华中科技大学 Partitioning and parallel distribution processing method of super-large scale RDF graph data
US20170134519A1 (en) * 2015-11-08 2017-05-11 Vmware, Inc. Deploying an application in a multiple cloud computing environments
CN106708978A (en) * 2016-12-07 2017-05-24 北京邮电大学 Bipartite graph-based service recommendation method and apparatus

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102768670A (en) * 2012-05-31 2012-11-07 哈尔滨工程大学 Webpage clustering method based on node property label propagation
CN104579850A (en) * 2015-01-27 2015-04-29 北京邮电大学 Quality of service (QoS) prediction method for Web service under mobile Internet environment
CN104809168A (en) * 2015-04-06 2015-07-29 华中科技大学 Partitioning and parallel distribution processing method of super-large scale RDF graph data
US20170134519A1 (en) * 2015-11-08 2017-05-11 Vmware, Inc. Deploying an application in a multiple cloud computing environments
CN106708978A (en) * 2016-12-07 2017-05-24 北京邮电大学 Bipartite graph-based service recommendation method and apparatus

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
张峻铭，李静林，王尚广,刘志晗,袁泉,杨放春: "基于时空图的移动对象聚集模式挖掘方法", 《软件学报》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109526012A (en) * 2019-01-24 2019-03-26 重庆邮电大学 A kind of LoRaWAN network spreading factor distribution method based on reliability

Also Published As

Publication number Publication date
CN109040214B (en) 2020-07-17

Similar Documents

Publication Publication Date Title
US11704144B2 (en) Creating virtual machine groups based on request
CN110324164B (en) Network slice deployment method and device
CN109949111B (en) Electronic bill identification distribution method, electronic bill generation method, device and system
CN103095687B (en) metadata processing method and device
CN106909451A (en) A kind of distributed task dispatching system and method
CN103873505B (en) Method, system and the device uploaded offline to cloud storage service device
CN105357322B (en) A kind of virtual machine distribution method based on Topology partition
CN109617807B (en) Route information update method, device, gateway and storage medium
CN103077082A (en) Method and system for distributing data center load and saving energy during virtual machine migration
CN106407244A (en) Multi-database-based data query method, system and apparatus
US10761869B2 (en) Cloud platform construction method and cloud platform storing image files in storage backend cluster according to image file type
CN102438278A (en) Load allocation method and device for mobile communication network
CN106960011A (en) Metadata of distributed type file system management system and method
CN104901999B (en) A kind of virtual data center cross-domain mapping method that reliability perceives
CN109618003A (en) A kind of servers' layout method, server and storage medium
CN103229480A (en) Data processing method, device and client in distributed storage system
CN106940712A (en) Sequence generating method and equipment
CN109040214A (en) A kind of service arrangement method that reliability enhances under cloud environment
CN109951551A (en) A kind of container mirror image management system and method
CN108153494A (en) A kind of I/O request processing method and processing device
CN103677994B (en) Distributed data processing system, device and method
CN112380040B (en) Message processing method and device, electronic equipment and storage medium
CN108829709A (en) Distributed database management method, apparatus, storage medium and processor
CN113688115B (en) Archive big data distributed storage system based on Hadoop
CN110471968A (en) Dissemination method, device, equipment and the storage medium of ETL task

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