CN106453641A - Dynamic enterprise cloud service platform expansion method, device and system - Google Patents
Dynamic enterprise cloud service platform expansion method, device and system Download PDFInfo
- Publication number
- CN106453641A CN106453641A CN201611062102.1A CN201611062102A CN106453641A CN 106453641 A CN106453641 A CN 106453641A CN 201611062102 A CN201611062102 A CN 201611062102A CN 106453641 A CN106453641 A CN 106453641A
- Authority
- CN
- China
- Prior art keywords
- server
- enterprise
- operation requests
- enterprise customer
- frequency
- 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
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
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1031—Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
Abstract
The invention discloses a dynamic enterprise cloud service platform expansion method which comprises the following steps: acquiring the operation volume of each operation request of each enterprise user in a first server and/or the frequency of the operation request of each enterprise user in the duration period; judging whether the operation volume of each operation request of each enterprise user in the first server and/or the frequency of the operation request of each enterprise user in the duration period exceed (s)a preset threshold value; distributing a subsequent operation request of the enterprise user into a second server if the frequency of the operation request of the enterprise user in the first server and/or the operation volume each time are (is) judged to exceed the preset threshold value. According to the dynamic enterprise cloud service platform expansion method, device and system, normal operations can be maintained by server clusters in reasonable quantity, the enterprises can be dynamically distributed into standby server clusters, and normal use of the rest enterprise users is not influenced when high performance response speed of the complicated request is guaranteed.
Description
Technical field
The present invention relates to a kind of enterprise cloud service platform dynamic capacity-expanding method, apparatus and system.
Background technology
Enterprises service platform based on cloud computing, when providing enterprises service for enterprise customer, typically can use Cloud Server
Cluster, focuses on the request of all enterprises.The data operation request of each enterprise, understands random be sent in cluster one
Server is processed.Due to each enterprise of cloud service platform service, to there is scale, use time and mode variant,
When some enterprises are when having the computation requests of complexity some indefinite or specific time period, then can consume corresponding server
Calculation resources, thus impact simultaneously using other enterprises of this corresponding server arithmetic speed.In prior art, typically
To be solved the above problems using solution below:First, buy substantial amounts of server and form cluster, this mode, enterprise can be given
Service platform brings substantial amounts of cost, when request amount is very low, can produce a large amount of idle machine, cause very big waste.2nd, every
Individual enterprise disposes the privately owned Cloud Server of oneself, this mode, and the computing capability of the server that each enterprise can enjoy reduces,
Although itself complex requests does not interfere with other people, be the calculating performance of privately owned Cloud Server compared with Cloud Server cluster, property
Can substantially reduce, the processing speed amount of complex requests is greatly reduced.
Content of the invention
For above-mentioned the deficiencies in the prior art, the technical problem to be solved is:Offer one kind can be with reasonable
The server of quantity maintains enterprise's cloud service platform dynamic capacity-expanding method of normal operation, apparatus and system.
For solving above-mentioned technical problem, the technical problem that the present invention adopts is:A kind of enterprise cloud service platform is provided
Dynamic capacity-expanding method, comprises the following steps:
S101, obtain first server in each enterprise customer the operational ton of operation requests each time and/or in the persistent period
The frequency of the operation requests of each enterprise customer in section;
S103, judge in first server the operational ton of operation requests each time of each enterprise customer and/or in the persistent period
Whether the frequency of the operation requests of each enterprise customer in section exceedes predetermined threshold;
If S105 judge to obtain in first server the frequency of the operation requests of enterprise customer and/or each time operational ton exceed
Predetermined threshold, then by the subsequent operation request distribution of this enterprise customer to second server.
Further, described first server and second server are some, some first server compositions first
Cluster of servers, some second servers form second server cluster.
Further, described first server cluster is preferred server cluster, and described second server cluster is standby
Server cluster.
Further:
In S101 step, obtained by each operation requests of each enterprise customer in each first server of storage
The operational ton of operation requests each time of each enterprise customer and/or every in persistent period section in each first server
The frequency of the operation requests of one enterprise customer;
In S103 step, judge in each server the operational ton of operation requests each time of each enterprise customer and/or
The frequency of the operation requests of each enterprise customer in persistent period section;
In S105 step, if judging to obtain the behaviour of the one or more enterprise customers in wherein one or more first server
Ask and/or operational ton exceedes predetermined threshold each time, then by the subsequent operation of this corresponding one or more enterprise customer
Request distribution is to the second server not less than predetermined threshold and/or first server.
Further, in S105 step, specifically include:
If judging to obtain operation requests of one or more enterprise customers in wherein one or more first server and/or every
Once-through operation amount exceedes predetermined threshold, then the first server that this exceedes predetermined threshold will exceed threshold information and feed back to reverse generation
Reason server, wherein, described exceedes the enterprise customer ID that threshold information includes exceeding predetermined threshold;Reverse Proxy will be right
The subsequent operation request distribution of the one or more enterprise customers answering is to the second server and/or first not less than predetermined threshold
In server.
For solving above-mentioned technical problem, another technical problem that the present invention adopts is:A kind of cloud service of enterprise is provided to put down
Platform dynamic capacity-expanding method, comprises the following steps:
First server obtains and logs on to the operation requests of enterprise customer thereon and store;
First server according to storage record obtain its interior each enterprise customer operation requests each time operational ton and/or
The frequency of the operation requests of each enterprise customer in persistent period section;
First server judges in it operational ton of operation requests each time of each enterprise customer and/or in persistent period section
The frequency of the operation requests of each enterprise customer whether exceed predetermined threshold;
If first server judges to obtain operation requests in described persistent period section for one or more enterprise customers in it
Frequency and/or each time operational ton exceed predetermined threshold, then will exceed threshold information and feed back to Reverse Proxy, for anti-
To proxy server, the subsequent operation request of this one or more enterprise customer is distributed to second server.
For solving above-mentioned technical problem, another technical problem that the present invention adopts is:A kind of cloud service of enterprise is provided to put down
Platform dynamic capacity-expanding method, comprises the following steps:
Reverse Proxy receives the operation requests of each enterprise customer;
The operation requests receiving are randomly assigned to first server Reverse Proxy;
What Reverse Proxy reception first server was fed back exceedes threshold information;
Reverse Proxy exceedes threshold information according to feedback, will exceed after the corresponding enterprise customer of threshold information with this
Continuous operation requests are randomly assigned to second server.
For solving above-mentioned technical problem, another technical problem that the present invention adopts is:A kind of cloud service of enterprise is provided to put down
Platform dynamic capacity-expanding device, including:
Acquisition module, for obtain in first server the operational ton of the operation requests each time of each enterprise customer and/or
The frequency of the operation requests of each enterprise customer in persistent period section;
Judge module, for judge in first server the operational ton of the operation requests each time of each enterprise customer and/or
Whether the frequency of the operation requests of each enterprise customer in persistent period section exceedes predetermined threshold;
Operation requests distribute module, for obtaining the operation requests of enterprise customer in first server when the judgement of described judge module
And/or each time when operational ton exceedes predetermined threshold, by the subsequent operation request distribution of this enterprise customer to second server.
For solving above-mentioned technical problem, the technical problem that the present invention adopts is:A kind of enterprise cloud service platform is provided
Dynamic capacity-expanding system, including:
First server;
Second server;
Reverse Proxy, for receiving the operation requests of each enterprise customer, and operation requests is randomly assigned to described
First server;
Described first server, for receiving the operation requests of Reverse Proxy distribution and being processed, storage simultaneously should
Operation requests;It is additionally operable to judge the operational ton of operation requests each time of each enterprise customer and/or in persistent period section
Whether the frequency of the operation requests of each enterprise customer exceedes predetermined threshold, if exceeding predetermined threshold, this is exceeded threshold value letter
Breath feeds back to Reverse Proxy;
Described Reverse Proxy is additionally operable to exceed threshold information according to feed back, after the corresponding enterprise customer of this this information
Continuous operation requests are randomly assigned to second server.
Further, described first server and second server are some, and some first server compositions are conventional
Server cluster, some second servers form backup server cluster.
Enterprise's cloud service platform dynamic capacity-expanding method of the present invention, apparatus and system, compared with existing two schemes, have
Remarkable advantage.Can be maintained with the server cluster of fair amount and run at ordinary times.When complex requests fluctuation occurs, dynamic
This Some Enterprises is assigned in backup server cluster, while ensureing to have high-performance response speed to complex requests, no
Affect the normal use of remaining enterprise customer.
Brief description
In order to be illustrated more clearly that the technical scheme of the embodiment of the present invention, below will be in embodiment or description of the prior art
The accompanying drawing of required use be briefly described it should be apparent that, drawings in the following description be only the present invention some are real
Apply example, for those skilled in the art, on the premise of not paying creative work, can also be obtained according to these accompanying drawings
Other accompanying drawings.
The flow chart of Tu1Shi enterprise of the present invention cloud service platform dynamic capacity-expanding method first embodiment.
The flow chart of Tu2Shi enterprise of the present invention cloud service platform dynamic capacity-expanding method first embodiment.
The flow chart of Tu3Shi enterprise of the present invention cloud service platform dynamic capacity-expanding method first embodiment.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation description is it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of not making creative work
Embodiment, broadly falls into the scope of protection of the invention.
Refer to Fig. 1, the flow chart of Tu1Shi enterprise of the present invention cloud service platform dynamic capacity-expanding method first embodiment.This
The first server being related in embodiment is some, and second server is some, the quantity of second server and first
The quantity of server can differ.Some first servers form first server cluster, and this first server cluster is
Preferred server cluster, then, when Reverse Proxy batch operation is asked, preferentially distribute to this first server cluster, if
Dry platform second server composition second server cluster is backup server cluster.First server cluster and second server collection
Servers-all under group is respectively provided with consistent function.It is appreciated that in other examples, described first server and
The quantity of second server can have different changes, such as first server and second server all can be one.
Enterprise's cloud service platform dynamic capacity-expanding method of the present embodiment, comprises the following steps:
S101, obtain first server in each enterprise customer the operational ton of operation requests each time and/or in the persistent period
The frequency of the operation requests of each enterprise customer in section;
The operation requests of each enterprise customer in this persistent period section, may refer to a certain User logs in cycle(Such as 24
Hour)The operation requests that interior first time gets this enterprise customer are rising of persistent period section, with cutting of predetermined persistent period
Only the time is the persistent period.For example, the initial time receiving the operation requests of this enterprise customer in first time is 8:00, make a reservation for
Persistent period be 30 minutes, then, persistent period Duan Zewei 8:00 to 8:30.
In the present embodiment, when the operational ton of the operation requests each time of each enterprise customer of acquisition does not then limit lasting
Between, when first server often receives the operation requests of an enterprise customer, then obtain once-through operation amount.
Operation requests mainly include User logs in and business operation request, and the operation requests obtaining in this step are then
Business operation is asked(Hereinafter referred to as operation requests).
Before this step, each enterprise customer also needs to first carry out user name login step, enterprise customer after login
Then business operation request can be carried out, so that after this first server cluster and second server cluster all can process login
The operation requests that enterprise customer sends.When logging in, first send logging request, Reverse Proxy to Reverse Proxy
Logging request is randomly assigned to any one first server, this first server can issue an encryption string, Yong Huhou
Each continuous request can bring this encryption string, and any server can parse to judge whether login is effective.
In corporate user registration, identity can be carried out for enterprise customer, such as write enterprise in cookie
The ID of user.In instantiation, login effective time can be set, for example, set effective time as 24 hours, then at this
In 24 hours, enterprise customer can only log in once, you can carries out operation requests below.The body that enterprise customer logs in each time
Part mark can be inconsistent, can distribute identity temporarily, as long as its of this identity of this enterprise customer and login
The identity of his enterprise customer is inconsistent.
During each enterprise customer's operation requests, all carry corresponding identity, when specifically carrying out operation requests,
The operation requests of all enterprise customers are first sent to Reverse Proxy(Such as nginx Reverse Proxy)On, then
Carried out point to agency to actual service server from described Reverse Proxy(I.e. above-mentioned first server).For example there is enterprise
Industry user 1, enterprise customer 2 ..., enterprise customer n, then they log in after, the operation requests of this n enterprise customer are all sent out
Deliver to Reverse Proxy, then the operation requests of n enterprise customer are randomly assigned to Reverse Proxy
One server(Can be allocated according to the process state of first server).
During this step obtains suddenly, each operation especially by each enterprise customer in each first server of storage please
Ask to obtain the frequency of the operation requests of each enterprise customer in persistent period section for each first server and/or each
Secondary operational ton.The cumulative of the operation requests that all first servers receive can be stored using data base, memory module etc.
Number of times and each time operational ton(Resource shared by operation requests each time)Such that it is able to obtain each first server
The frequency of operation requests within the predetermined persistent period for each enterprise customer that processes and/or each time operational ton.
This obtaining step, can be that first server obtains corresponding each enterprise from memory module, database
The operational ton of industry user and operation requests frequency or separately set acquisition module and obtained, then inform this first clothes again
Business device, acquisition modes have diversification however it is not limited to above-mentioned acquisition modes.
In the present embodiment, need to obtain frequency and the operational ton each time of operation requests, the frequency according to operation requests with
Operational ton dynamically to judge the load mechanism of each server each time.It should be understood that in various embodiments, permissible
To judge the load mechanism of each server according only to operation frequency or operational ton.
S103, judge in each server the operational ton of operation requests each time of each enterprise customer and/or holding
Whether the frequency of the operation requests of each enterprise customer in the continuous time period exceedes predetermined threshold;
In the present embodiment, this judgment mechanism includes:Operational ton judges, operation requests frequency judges, operational ton and frequency synthesis are sentenced
Disconnected.
Set the operation requests allocation rule of a high load capacity rule in advance:Divide for operation requests frequency and each time operational ton
Other given threshold, also can process the time of the multiple request of this frequency according to the current first server of this threshold value prediction(I.e. this
One server needs high load capacity to process the high load capacity time of this multi-pass operation request), predict current first server process this behaviour
The time of the once-through operation request that work is measured(I.e. this first server needs high load capacity to process the high load capacity time of this operational ton),
And the normal recovery time of this first server is calculated according to this duration of load application(I.e. from high load capacity recover positive normal duty when
Between), then feedback exceed threshold information to Reverse Proxy, wherein, this exceedes to carry in threshold information and has more than setting
The enterprise customer ID of threshold value, the first server normal recovery time of this high load capacity.Exceed when Reverse Proxy receives this
After threshold information, then the subsequent operation of this enterprise customer can be asked extension set to distribute to the first server of the non-high load capacity of others
And standby second server;After reaching normal recovery time, Reverse Proxy then can be by after this enterprise customer
Continuous operation requests recover distribution to conventional first server(Including before high load capacity first server), and second server
Standby again.
Specifically combine example:
1. operational ton judges:
The element that specified event package contains exceedes setting high load capacity threshold value n(I.e. operational ton exceedes given threshold)It is assumed that the event of specifying
For group mail, the high load capacity threshold value of first server A single group mail is 5000 envelopes, and prediction exceedes this high load capacity threshold value
When, first server needs high load capacity to process the high load capacity time of this operational ton, specific as follows:
The once-through operation of corporate client 1 asks, for sending group mail 5000 envelope, this operation requests to be distributed to first server
A, then when this first server A mass-sends 5000 envelope mail of corporate client 1, it is little that the high load capacity time increases by 1 in current time
When.First server A exceedes threshold information to Reverse Proxy feedback, and this exceedes threshold information and then includes first service
Device A processes the ID of the high load capacity time, normal recovery time and corporate client 1 of this 5000 envelope mail.Reverse Proxy
Upon receipt, if downstream industry client 1 sends operation requests again, then Reverse Proxy is then each by follow-up transmission
Secondary operation requests are randomly assigned to other first servers thought except this first server A and/or standby second server.
When first server A recovers normal process, Reverse Proxy then can distribute corporate client according to being randomly assigned rule
Operation requests.
2nd, operation requests frequency judges:
Specify the request amount of request more than m in unit interval n(I.e. in the persistent period same corporate client operation requests frequency
Exceed predetermined threshold).
Assume first server A in 30 minutes can newly-built customer data 100, the company that illustrate in participation exhibition or
In typing client, may be exactly subsequently to send out mail, the peak of operation client, the high load capacity time setting first server A is little as 1
When, specific as follows:
Corporate client 1 newly-built customer data 100 in 30 minutes, sends 100 operation requests to first server A, and first
Server A then exceedes threshold information to Reverse Proxy feedback, includes first server A and processes this 100 client's phases
The ID of the high load capacity time, normal recovery time and corporate client 1 of pass data.The process of Reverse Proxy ibid, this
Place no longer repeats one by one.
3. comprehensive descision
Operational ton is more than x, and in unit interval y, request amount is more than z.
More than 100,000, corporate client 1 screened client to first server A in 5 minutes to such as corporate client 1 client's sum
Operation be more than 10, may explanation corporate client 1 company carrying out client's screening distribution, these operating resources consume very big,
Half an hour high capacity can be set.
High load capacity rule:Client's sum is more than 100,000, and screening guest operation is more than 10, during high capacity in 5 minutes
Between 30 minutes, first server A then sends more than threshold information to Reverse Proxy, during by corporate client 1ID, high load capacity
Between and feed back to Reverse Proxy normal recovery time, Reverse Proxy carries out above-mentioned similar operations, herein not
An another detailed description.
It should be understood that above-mentioned 1,2,3 to explain the thinking of the present invention as just a kind of embodiment, different
In embodiment however it is not limited to above-mentioned 1,2,3 mode of operation.For example in certain embodiments, as long as current first server is sentenced
When any one exceedes preset value in its interior operation requests frequency of breaking and operational ton, then send more than threshold information to reverse proxy
Server, the described threshold information that exceedes can also only include making the enterprise customer ID of this current first server high load capacity.And work as
The high load capacity time of front server or normal recovery time, might not can also pass through existing by the way of above-mentioned prediction
Any monitoring, monitoring, acquisition modes whether recover normal duty come the first server to judge high load capacity, so that reversely
Proxy server is in follow-up operation requests, it is determined whether need to send to the first server of high load capacity before this.
If S105 judges that the operation obtaining the one or more enterprise customers in wherein one or more first server please
Ask and/or operational ton exceedes predetermined threshold each time, then the subsequent operation of this corresponding one or more enterprise customer is asked
Distribute to the first server not less than predetermined threshold and/or second server.
In this step, when judging to obtain the one or more enterprise customers' in wherein one or more first server
Operation requests and/or each time operational ton exceed predetermined threshold, then the first server that this exceedes predetermined threshold will exceed threshold value
Feedback of the information to Reverse Proxy, wherein, described exceedes the enterprise customer ID that threshold information includes exceeding predetermined threshold;Instead
To proxy server by the subsequent operation of corresponding one or more enterprise customers request distribution to not less than the of predetermined threshold
In one server and/or second server.
This judgement step can be operation frequency and/or the behaviour of each enterprise customer to itself for each first server
Judged or arranged judge module, flow process etc. to judge each enterprise customer's in each server as amount
Operation frequency and/or operational ton.
Embodiment of the present invention, arranges preferred server cluster and backup server cluster to realize enterprise's cloud service platform
Dynamically distributes rule, each first server obtain its operation in persistent period section for the interior each enterprise customer in real time
The frequency of request and the operational ton obtaining its interior each enterprise customer, set high load capacity rule, that is, as operation frequency or behaviour
When work amount reaches respective settings threshold value, feed back to Reverse Proxy, feedback exceed during threshold information simultaneously by prediction from
Bear lotus and feed back to Reverse Proxy recovery time, Reverse Proxy obtains after this exceedes threshold information, will exceed
Being randomly assigned to preferred server cluster and/or backup server cluster of the subsequent operation request dynamic of the enterprise customer of threshold value
Middle others server is processed.So, it is possible to maintain with the server cluster of fair amount and run at ordinary times.Complicated when occurring
During request fluctuation, this Some Enterprises being transferred in other servers in server cluster of dynamic, ensureing to complexity
While request has high-performance response speed, do not affect the normal use of remaining enterprise customer.
Refer to Fig. 2, the flow chart of Tu2Shi enterprise of the present invention cloud service platform dynamic capacity-expanding method first embodiment.This
Enterprise's cloud service platform of embodiment equally includes some first service same or analogous with first embodiment structure or function
Device and some second servers, some first servers form preferred server cluster, and some second servers form standby clothes
Business device cluster.Enterprise's cloud service platform dynamic capacity-expanding method of the present embodiment, comprises the following steps:
S201, first server obtain and log on to the operation requests of enterprise customer thereon and store;
Described enterprise cloud service platform serves some enterprises, the such as business such as process of corporate mail, corporate client management.By
Scale in each enterprise customer differs, and therefore each enterprise customer corresponding customers quantity, scale are variant.One
A little more enterprise customers of customer quantity are processing mail mass-sending, client resources management(For example newly-built customer profile, Customer mail
Mass-sending)When, when single mail is mass-sended, its operational ton is larger may to make first server produce to the enterprise customer of more customer quantity
Raw high load capacity, then if this server is further continued for the process processing this enterprise customer, makes other enterprises on this server
The process of industry user is slack-off, makes troubles for other enterprise customers and bad experience sense and satisfaction.
When each enterprise customer needs using this cloud service platform processing business, need first to log on to preferred server
On, corresponding business operation request can be carried out after login, this process is all allocated by reverse server, when being assigned to
When on one server, first server stores number of operations and the operational ton of this enterprise customer.
Preferably, each server is all by the operation requests number of times of its each interior enterprise customer, operate each time
Amount stores to shared memory module or data base, is so capable of the getting in each first server of more transparence
Information about operation requests.
S203, first server obtain the behaviour of the operation requests each time of its interior each enterprise customer according to storage record
Measure and/or operation requests of each enterprise customer in persistent period section frequency;
When obtaining, each operation requests especially by each enterprise customer in each first server of storage to obtain
The frequency of the operation requests of each enterprise customer in persistent period section of each first server and/or operate each time
Amount.
S205, first server judge in it operational ton of operation requests each time of each enterprise customer and/or are holding
Whether the frequency of the operation requests of each enterprise customer in the continuous time period exceedes predetermined threshold;
If S207 first server judges that obtaining operation in described persistent period section for one or more enterprise customers in it asks
The frequency asked and/or each time operational ton exceed predetermined threshold, then will exceed threshold information and feed back to Reverse Proxy, with
For Reverse Proxy, the subsequent operation request of this one or more enterprise customer is distributed to second server.
In this step, when judging to obtain operational ton and/or the behaviour of its interior a certain enterprise customer or multiple enterprise customer
The frequency making to ask exceedes predetermined threshold, and this first server then predicts that it processes this operational ton and/or this operates several times
Request needs high load capacity time of high load capacity, then by with high load capacity process time, recovery time, exceed the enterprise of predetermined threshold
Industry ID sends to Reverse Proxy, and Reverse Proxy receives after exceeding threshold information, in this high load capacity
The interior subsequent operation request no longer sending this enterprise customer to the first server of above-mentioned high load capacity, then to preferred server
Other first servers in cluster and/or backup server cluster are randomly assigned the subsequent operation request of this enterprise customer.
Above-mentioned dynamically distributes mechanism is same or similar with first embodiment, just no longer repeats one by one herein.The present invention is implemented
Mode, is stored the operation requests number of times of each enterprise customer in all first servers, is operated each time using memory module
Request time, each time operational ton, each first server only needs to directly invoke the behaviour that can get all users in it
Making the frequency asked and operational ton each time, thus obtaining itself whether high load capacity, setting up high load capacity mechanism, by superthreshold value information
Send to Reverse Proxy, so that the subsequent operation request of this one or more enterprise customer is divided by Reverse Proxy
It is assigned in other first servers and/or second server.
Refer to Fig. 3, Fig. 3 is the flow chart of the 3rd embodiment of cloud service platform dynamic capacity-expanding method of the present invention.This reality
Apply example and dynamically distributes expansion method is elaborated for visual angle with Reverse Proxy.The method comprises the following steps:
S301, Reverse Proxy receive the operation requests of each enterprise customer;
The operation requests receiving are randomly assigned to first server for S303, Reverse Proxy;
What S305, Reverse Proxy reception first server were fed back exceedes threshold information;
S307, Reverse Proxy exceed threshold information according to feedback, will exceed the corresponding enterprise of threshold information with this and use
The subsequent operation request at family is randomly assigned to second server.
Embodiment of the present invention, Reverse Proxy can be learnt in preferred server cluster according to exceeding threshold information
High load capacity first server, thus the operation requests of follow-up enterprise customer are distributed to the first server of normal duty
Or distribute to standby server, using this dynamically distributes mechanism it is not necessary to as conventional art purchases enough clothes over there
Business device, the server using fair amount can achieve dynamic capacity-expanding, and the business providing high-quality for each corporate client can simultaneously
Enough reduces costs.
The invention also discloses a kind of enterprise cloud service platform dynamic capacity-expanding device, including:
Acquisition module, for obtaining the frequency of the operation requests of each enterprise customer in persistent period section for the first server
And/or operational ton each time;
Judge module, for judge in first server the operational ton of the operation requests each time of each enterprise customer and/or
Whether the frequency of the operation requests of each enterprise customer in persistent period section exceedes predetermined threshold;
Operation requests distribute module, for obtaining the operation requests of enterprise customer in first server when the judgement of described judge module
And/or each time when operational ton exceedes predetermined threshold, by the subsequent operation request distribution of this enterprise customer to second server.
The concrete grammar of the acquisition module of the present embodiment, judge module and operation requests distribute module is shown in that said method is real
Apply example, no longer repeat one by one herein
The invention also discloses a kind of enterprise cloud service platform dynamic capacity-expanding system, including:
First server;
Second server;
Reverse Proxy, for receiving the operation requests of each enterprise customer, and operation requests is randomly assigned to described
First server;
Described first server, for receiving the operation requests of Reverse Proxy distribution and being processed, storage simultaneously should
Operation requests;It is additionally operable to judge the frequency of the operation requests of same enterprise customer in a persistent period section and/or grasp each time
Measure and whether exceed predetermined threshold, if exceeding predetermined threshold, this being exceeded threshold information and feeds back to Reverse Proxy;
Described Reverse Proxy is additionally operable to exceed threshold information according to feed back, after the corresponding enterprise customer of this this information
Continuous operation requests are randomly assigned to second server.
Described first server and second server are some, and some first servers form preferred server collection
Group, some second servers form backup server cluster.
Embodiment of the present invention, compared with traditional two schemes, has remarkable advantage.Can be with the service of fair amount
Device cluster maintains and runs at ordinary times.When occur complex requests fluctuation when, dynamic this Some Enterprises is assigned to standby server
In cluster, while ensureing to have high-performance response speed to complex requests, do not affect the normal use of remaining enterprise customer.Carry
The satisfaction of high enterprise customer, Experience Degree, have saved server purchase cost simultaneously.
These are only embodiments of the present invention, not thereby limit the scope of the claims of the present invention, every utilization present invention
Equivalent structure or equivalent flow conversion that description and accompanying drawing content are made, or directly or indirectly it is used in other related technology
Field, is included within the scope of the present invention.
Claims (10)
1. a kind of enterprise cloud service platform dynamic capacity-expanding method, comprises the following steps:Each in S101, acquisition first server
The operational ton of operation requests each time of enterprise customer and/or the operation requests of each enterprise customer in persistent period section
Frequency;S103, judge in first server the operational ton of operation requests each time of each enterprise customer and/or when lasting
Between the frequency of operation requests of each enterprise customer in section whether exceed predetermined threshold;If S105 judges to obtain first service
In device the frequency of the operation requests of enterprise customer and/or each time operational ton exceed predetermined threshold, then by after this enterprise customer
Continuous operation requests are distributed to second server.
2. enterprise as claimed in claim 1 cloud service platform dynamic capacity-expanding method it is characterised in that:Described first server and
Second server is some, and some first servers form first server cluster, some second server compositions second
Cluster of servers.
3. enterprise as claimed in claim 2 cloud service platform dynamic capacity-expanding method it is characterised in that:Described first server collection
Group is preferred server cluster, and described second server cluster is backup server cluster.
4. enterprise as claimed in claim 3 cloud service platform dynamic capacity-expanding method it is characterised in that:In S101 step, lead to
The each operation requests crossing each enterprise customer in each first server of storage are obtaining in each first server
The operation of the operational ton of operation requests each time of each enterprise customer and/or each enterprise customer in persistent period section please
The frequency asked;In S103 step, judge the operational ton of the operation requests each time of each enterprise customer in each server
And/or the frequency of the operation requests of each enterprise customer in persistent period section;In S105 step, if judging to obtain it
In operation requests of one or more enterprise customers in one or more first server and/or each time operational ton exceed pre-
Determine threshold value, then by the subsequent operation request distribution of this corresponding one or more enterprise customer to not less than the second of predetermined threshold
In server and/or first server.
5. enterprise as claimed in claim 4 cloud service platform dynamic capacity-expanding method is it is characterised in that in S105 step, have
Body includes:If judge to obtain operation requests of one or more enterprise customers in wherein one or more first server and/
Or operational ton exceedes predetermined threshold each time, then the first server that this exceedes predetermined threshold will exceed threshold information and feed back to instead
To proxy server, wherein, described exceed the enterprise customer ID that threshold information includes exceeding predetermined threshold;Reverse Proxy
By the subsequent operation of corresponding one or more enterprise customers request distribution to the second server not less than predetermined threshold and/or
In first server.
6. a kind of enterprise cloud service platform dynamic capacity-expanding method, comprises the following steps:First server obtains and logs on to thereon
The operation requests of enterprise customer simultaneously store;First server obtains each of its interior each enterprise customer according to storage record
The operational ton of secondary operation requests and/or the frequency of the operation requests of each enterprise customer in persistent period section;First service
Device judges that in it, operational ton of operation requests each time of each enterprise customer and/or each enterprise in persistent period section use
Whether the frequency of the operation requests at family exceedes predetermined threshold;If first server judges that obtaining one or more enterprises in it uses
The frequency of operation requests in described persistent period section for the family and/or each time operational ton exceed predetermined threshold, then will exceed threshold
Value information feeds back to Reverse Proxy, so that Reverse Proxy is by the subsequent operation of this one or more enterprise customer
Request distribution is to second server.
7. a kind of enterprise cloud service platform dynamic capacity-expanding method, comprises the following steps:Reverse Proxy receives each enterprise
The operation requests of user;The operation requests receiving are randomly assigned to first server Reverse Proxy;Reversely
What proxy server reception first server was fed back exceedes threshold information;Reverse Proxy exceedes threshold value according to feedback
Information, the subsequent operation exceeding the corresponding enterprise customer of threshold information with this is asked to be randomly assigned to second server.
8. a kind of enterprise cloud service platform dynamic capacity-expanding device is it is characterised in that include:Acquisition module, for obtaining the first clothes
The operational ton of operation requests each time of each enterprise customer and/or each enterprise customer in persistent period section in business device
The frequency of operation requests;Judge module, for judging the behaviour of the operation requests each time of each enterprise customer in first server
Measure and/or whether the frequency of operation requests of each enterprise customer in persistent period section exceedes predetermined threshold;Operation please
Seek distribute module, for judging to obtain in first server operation requests of enterprise customer and/or each when described judge module
When secondary operational ton exceedes predetermined threshold, by the subsequent operation request distribution of this enterprise customer to second server.
9. a kind of enterprise cloud service platform dynamic capacity-expanding system, including:First server;Second server;Reverse proxy takes
Business device, for receiving the operation requests of each enterprise customer, and operation requests is randomly assigned to described first server;Institute
State first server, for receiving the operation requests of Reverse Proxy distribution and being processed, store this operation please simultaneously
Ask;It is additionally operable to judge the operational ton of operation requests each time of each enterprise customer and/or each enterprise in persistent period section
Whether the frequency of the operation requests of industry user exceedes predetermined threshold, if exceeding predetermined threshold, this is exceeded threshold information feedback
To Reverse Proxy;Described Reverse Proxy is additionally operable to exceed threshold information according to feed back, by this this information pair
The enterprise customer's subsequent operation request answered is randomly assigned to second server.
10. enterprise as claimed in claim 9 cloud service platform dynamic capacity-expanding system it is characterised in that:Described first server
And second server is some, some first servers form preferred server cluster, and some second server compositions are standby
Use server cluster.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611062102.1A CN106453641B (en) | 2016-11-24 | 2016-11-24 | Enterprise's cloud service platform dynamic capacity-expanding method, apparatus and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611062102.1A CN106453641B (en) | 2016-11-24 | 2016-11-24 | Enterprise's cloud service platform dynamic capacity-expanding method, apparatus and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106453641A true CN106453641A (en) | 2017-02-22 |
CN106453641B CN106453641B (en) | 2018-05-22 |
Family
ID=58219063
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611062102.1A Active CN106453641B (en) | 2016-11-24 | 2016-11-24 | Enterprise's cloud service platform dynamic capacity-expanding method, apparatus and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106453641B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107682391A (en) * | 2017-08-04 | 2018-02-09 | 平安科技(深圳)有限公司 | Electronic installation, server-assignment control method and computer-readable recording medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101695049A (en) * | 2009-11-10 | 2010-04-14 | 杭州华三通信技术有限公司 | Method and device for processing businesses in monitoring system |
US20110078303A1 (en) * | 2009-09-30 | 2011-03-31 | Alcatel-Lucent Usa Inc. | Dynamic load balancing and scaling of allocated cloud resources in an enterprise network |
CN202906933U (en) * | 2012-05-28 | 2013-04-24 | 深圳市谷米科技有限公司 | An automatic fault-tolerant system based on javascript |
CN103257683A (en) * | 2013-05-07 | 2013-08-21 | 华为技术有限公司 | Method and device of cloud calculation service expansion and contraction |
CN103309946A (en) * | 2013-05-16 | 2013-09-18 | 华为技术有限公司 | Method, device and system for processing multimedia file |
CN103905220A (en) * | 2012-12-25 | 2014-07-02 | 腾讯科技(北京)有限公司 | Data synchronization processing method and system |
CN105024842A (en) * | 2014-04-25 | 2015-11-04 | 深圳市腾讯计算机系统有限公司 | Method and device for capacity expansion of server |
CN105553721A (en) * | 2015-12-15 | 2016-05-04 | 浪潮电子信息产业股份有限公司 | Cloud application stretching method, application management side and system |
-
2016
- 2016-11-24 CN CN201611062102.1A patent/CN106453641B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110078303A1 (en) * | 2009-09-30 | 2011-03-31 | Alcatel-Lucent Usa Inc. | Dynamic load balancing and scaling of allocated cloud resources in an enterprise network |
CN101695049A (en) * | 2009-11-10 | 2010-04-14 | 杭州华三通信技术有限公司 | Method and device for processing businesses in monitoring system |
CN202906933U (en) * | 2012-05-28 | 2013-04-24 | 深圳市谷米科技有限公司 | An automatic fault-tolerant system based on javascript |
CN103905220A (en) * | 2012-12-25 | 2014-07-02 | 腾讯科技(北京)有限公司 | Data synchronization processing method and system |
CN103257683A (en) * | 2013-05-07 | 2013-08-21 | 华为技术有限公司 | Method and device of cloud calculation service expansion and contraction |
CN103309946A (en) * | 2013-05-16 | 2013-09-18 | 华为技术有限公司 | Method, device and system for processing multimedia file |
CN105024842A (en) * | 2014-04-25 | 2015-11-04 | 深圳市腾讯计算机系统有限公司 | Method and device for capacity expansion of server |
CN105553721A (en) * | 2015-12-15 | 2016-05-04 | 浪潮电子信息产业股份有限公司 | Cloud application stretching method, application management side and system |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107682391A (en) * | 2017-08-04 | 2018-02-09 | 平安科技(深圳)有限公司 | Electronic installation, server-assignment control method and computer-readable recording medium |
CN107682391B (en) * | 2017-08-04 | 2020-06-30 | 平安科技(深圳)有限公司 | Electronic device, server allocation control method, and computer-readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN106453641B (en) | 2018-05-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112162865B (en) | Scheduling method and device of server and server | |
US8949410B2 (en) | Server load balancer scaling for virtual servers | |
US11206193B2 (en) | Method and system for provisioning resources in cloud computing | |
Rabbani et al. | On achieving high survivability in virtualized data centers | |
CN103428008A (en) | Big data distribution strategy oriented to multiple user groups | |
EP2335162A1 (en) | Dynamic distribution of virtual machines in a communication network | |
CN104426694B (en) | A kind of method and apparatus of adjustment resources of virtual machine | |
CN108933829A (en) | A kind of load-balancing method and device | |
Ma et al. | Distributed load balancing allocation of virtual machine in cloud data center | |
CN108600300A (en) | Daily record data processing method and processing device | |
CN110677274A (en) | Event-based cloud network service scheduling method and device | |
CN103491024A (en) | Job scheduling method and device for streaming data | |
CN106375328A (en) | Adaptive optimization operation method of large-scale data distribution system | |
US20060031242A1 (en) | Method, system, and program for distributing application transactions among work servers | |
CN103488538B (en) | Application extension device and application extension method in cloud computing system | |
Vashistha et al. | Comparative study of load balancing algorithms | |
CN108282526A (en) | Server dynamic allocation method and system between double clusters | |
CN105357239A (en) | Method and device for providing service, and method and device for acquiring service | |
Marandi et al. | Filo: Consolidated consensus as a cloud service | |
CN113315825A (en) | Distributed request processing method, device, equipment and storage medium | |
CN106453641A (en) | Dynamic enterprise cloud service platform expansion method, device and system | |
Chahal et al. | SLA-aware workload scheduling using hybrid cloud services | |
CN116703071A (en) | Resource sharing method, device and equipment and storage medium | |
CN114598656B (en) | Load balancing method and system applied to delivery service platform | |
CN106453677A (en) | Address allocation method and apparatus |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |