CN104135525A - Resource expanding method and device for cloud platform ELB components - Google Patents
Resource expanding method and device for cloud platform ELB components Download PDFInfo
- Publication number
- CN104135525A CN104135525A CN201410375345.5A CN201410375345A CN104135525A CN 104135525 A CN104135525 A CN 104135525A CN 201410375345 A CN201410375345 A CN 201410375345A CN 104135525 A CN104135525 A CN 104135525A
- Authority
- CN
- China
- Prior art keywords
- elb
- cloud platform
- service instance
- service
- assembly
- 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
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012545 processing Methods 0.000 claims abstract description 35
- 238000006116 polymerization reaction Methods 0.000 claims description 34
- 230000007480 spreading Effects 0.000 claims description 22
- 230000004044 response Effects 0.000 description 12
- 230000008569 process Effects 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000012544 monitoring process Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000000205 computational method Methods 0.000 description 3
- 238000001816 cooling Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 101150007711 rps5 gene Proteins 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Landscapes
- Debugging And Monitoring (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a resource expanding method and device for cloud platform ELB components. First, according to an auto scaling parameter and a service cluster parameter of a read cloud platform ELB component, calculate the max service request amount maxRpt processed by a single service example and the average increment per second m of the single service example; according to the service request amount in the obtained historical data, calculate an aggregate value growth rate k; compare the real-time service request amount with the max processing ability of the service example; and when the real-time service request amount is larger than the max processing ability of the service example, calculate the ratio of the aggregate value growth rate K to the average increment per second m of the single service example, and perform resource expanding according to the ratio. The present invention responds to the request of high load situation rapidly with high efficiency, improves the responding ability of the cloud platform, increases the using efficiency of the resource, saves costs, and has a great value in application.
Description
Technical field
The present invention relates to cloud computing technology applied technical field, particularly relate to a kind of resource extended method and device of cloud platform ELB assembly.
Background technology
Natural feature of cloud platform is exactly the access concurrent towards height, and solves the problem of response speed in high Concurrency Access, need to have the way of a load balancing to solve.The core concept of load balancing is exactly to process by request being distributed to different service ends, will promote like this throughput efficiency of whole platform.Similarly ELB (Elastic Load Balancer, elastic load equalizer) can play a part very important at CloudFoundry (a PaaS cloud platform), Google App Engine (Google's web application).In Stratos (a kind of PaaS service platform), ELB is mainly a load balancing to Cartridge access, in the time that too much developer needs same Cartridge, will first carry out a request shunting by ELB, again difference is asked flow point to be dealt into Different Ca rtridge, thereby obtain the mapping of Cartridge.
Because elasticity is one of of paramount importance characteristic of cloud computing platform, be also the major criterion of evaluating cloud computing platform ability, so the resilient expansion decision-making of cloud computing platform is particularly important.Whether the quality of resilient expansion decision-making technique can be expanded at reasonable time according to user's requirement being directly connected to cloud computing platform, so produced a large amount of research at cloud computing platform resilient expansion decision-making level, has also produced ripe product.Auto Scaling (resilient expansion) is the automatic expansion service on Amazon EC2 cloud computing platform, and its function is automatically to create or stop EC2 virtual machine instance according to user-defined trigger.The fundamental mode of Auto Scaling service as shown in Figure 1.Trigger (trigger) in Fig. 1 is user-defined trigger.According to the rule of trigger definition in figure, in the time that average CPU (Central Processing Unit, central processing unit) utilance is greater than 80%, EC2 example is increased to 10%, and in the time that average cpu busy percentage is less than 40%, EC2 example is reduced to 10%.Launch Configurations in figure (starting configuration) is used for the new needed parameter of EC2 example of specify creation.Another service CloudWatch (cloud monitoring) on Amazon EC2 is responsible for monitoring EC2, the expansion defining in the monitor data obtaining according to CloudWatch and trigger and contraction rule, Auto Scaling will expand or shrink EC2 automatically.
Response type expansion decision-making technique is a kind of method of carrying out automatic elastic expansion according to the extension rule of the load of current cloud computing platform and artificial setting.The resilient expansion method of Auto Scaling can range response type expansion decision-making technique.Due to monitored results and the artificial extension rule of setting that this class expansion decision-making technique places one's entire reliance upon to cloud computing platform, realize relatively simply, so adopting more, business-like cloud computing platform carries out in this way resilient expansion decision-making.But the defect of these class methods is fairly obvious, response type expansion decision-making technique is not considered the needed time of cloud computing platform resilient expansion, just in the time that monitored results meets extension rule, start simply resilient expansion, just can obtain required resource so cause user need to experience one period of time of delay.
Thereby prediction type expansion decision-making technique is to predict the method for carrying out resilient expansion decision-making by the load capacity to cloud computing platform future.These class methods are regarded the history monitoring value of cloud computing platform as a predictable sequence, and this sequence is carried out to mathematical modeling, can use the method such as recurrence, time series similarity analysis, the monitoring value of so next time point just can be used Mathematical Modeling to predict in the value of next time point.After the following load capacity of cloud computing platform is predicted, just can utilize the judgement that predicts the outcome when should carry out resilient expansion, make resilient expansion decision-making.Expansion decision-making technique based on pattern matching, predicts the following load capacity of cloud computing platform by improving KMP (linear session string matching algorithm) algorithm, thereby expands decision-making.Prediction type expansion decision-making technique can be tackled cloud computing platform and expand required time of delay, but simultaneously because needs are according to historical load capacity predict future load capacity, and this method needs in detail accurately historical monitor data as support.The existing research about quality of service monitor has adopted different frameworks, agreement or algorithm to reach the object of carrying out quality of service monitor in the situation that producing a small amount of traffic load, but its prerequisite is in the scene of quality of service monitor, not need value monitored accurately, or do not require and obtain at short notice monitored results, so existing quality of service monitor method can not obtain monitor data in detail accurately in time, so can not support well expansion decision-making.
In the practical application of cloud platform ELB assembly, general way is directly to adopt response type expanding policy, response type expansion decision-making technique is not considered the needed time of cloud computing platform resilient expansion, just in the time that monitored results meets extension rule, start simply resilient expansion, just can obtain required resource so cause user need to experience one period of time of delay.Prediction type expansion decision-making technique can be tackled cloud computing platform and expand required time of delay, but simultaneously because needs are according to historical load capacity predict future load capacity, this method needs in detail accurately historical monitor data as support, in real work, workload is often larger, can not meet the demand of real work.
Summary of the invention
Based on above-mentioned situation, the present invention proposes a kind of resource extended method of cloud platform ELB assembly, efficiently respond fast the request response of high load condition, improve the responding ability of cloud platform, improve resource utilization, cost-saving, there is very high using value.
To achieve these goals, the embodiment of technical solution of the present invention is:
A resource extended method for cloud platform ELB assembly, comprises the following steps:
According to automatic elastic spreading parameter and the service cluster parameter of the cloud platform ELB assembly reading, calculate maximum service request quantity maxRpt and the average increment m per second of single Service Instance that single Service Instance is accepted;
Calculate polymerization value growth rate k according to service request quantity in the historical data of obtaining;
Real time service request quantity and Service Instance maximum processing capability are compared to the maximum service request quantity maxRpt that described Service Instance maximum processing capability is accepted by described single Service Instance and real time execution Service Instance number n
in real timeproduct determine;
In the time that described real time service request quantity is greater than described Service Instance maximum processing capability, calculate the ratio of described polymerization value growth rate k and the average increment m per second of described single Service Instance, carry out resource expansion according to described ratio.
For prior art problem, the invention allows for a kind of resource expanding unit of cloud platform ELB assembly, effectively solve existing response type expansion and cause the problem that user need to experience one period of time of delay, prediction type expansion workload can not meet greatly real work demand, be applicable to application.
The embodiment of technical solution of the present invention is:
A resource expanding unit for cloud platform ELB assembly, comprising:
Processing module, automatic elastic spreading parameter and the service cluster parameter of the cloud platform ELB assembly reading for basis, calculate maximum service request quantity maxRpt and the average increment m per second of single Service Instance that single Service Instance is accepted, and calculate polymerization value growth rate k according to service request quantity in the historical data of obtaining;
Comparison module, for real time service request quantity and Service Instance maximum processing capability are compared, the maximum service request quantity maxRpt that described Service Instance maximum processing capability is accepted by described single Service Instance and real time execution Service Instance number n
in real timeproduct determine;
Resource expansion module, in the time that described real time service request quantity is greater than described Service Instance maximum processing capability, calculates the ratio of described polymerization value growth rate k and the average increment m per second of described single Service Instance, carries out resource expansion according to described ratio.
Compared with prior art, beneficial effect of the present invention is: resource extended method and the device of cloud platform ELB assembly of the present invention, the maximum service request quantity of accepting by introducing single Service Instance, the parameters such as the average increment per second of single Service Instance and polymerization value growth rate, calculate the ratio of polymerization value growth rate and the average increment per second of single Service Instance, according to the number of this Service Instance that need to start than value prediction, pre-cooling Service Instance, before peak value arrives, reach the stock number expansion of peak performance demand, effectively overcome response type expansion decision-making technique and do not considered the needed time of resilient expansion, cause user need to experience the problem that just can obtain required resource one period of time of delay, reduce prediction type expansion decision-making simultaneously and needed a large amount of historical load capacity predict future load capacity of digital simulation analysis, the efficient request response that responds fast high load condition, improve the responding ability of cloud platform, improve resource utilization, cost-saving, there is very high using value.
Brief description of the drawings
Fig. 1 is existing Auto Scaling service fundamental mode schematic diagram;
Fig. 2 is the resource extended method flow chart of an embodiment medium cloud platform ELB assembly;
Fig. 3 is the resource extended method flow chart based on the concrete example medium cloud platform ELB assembly of method one shown in Fig. 2;
Fig. 4 is the resource expanding unit structural representation of an embodiment medium cloud platform ELB assembly.
Embodiment
For making object of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is described in further detail.Should be appreciated that embodiment described herein, only in order to explain the present invention, does not limit protection scope of the present invention.
The resource extended method of an embodiment medium cloud platform ELB assembly, as shown in Figure 2, comprises the following steps:
Step S201: according to automatic elastic spreading parameter and the service cluster parameter of the cloud platform ELB assembly reading, calculate maximum service request quantity maxRpt and the average increment m per second of single Service Instance that single Service Instance is accepted;
Step S202: calculate polymerization value growth rate k according to service request quantity in the historical data of obtaining;
Step S203: real time service request quantity and Service Instance maximum processing capability are compared to the maximum service request quantity maxRpt that described Service Instance maximum processing capability is accepted by described single Service Instance and real time execution Service Instance number n
in real timeproduct determine;
Step S204: in the time that described real time service request quantity is greater than described Service Instance maximum processing capability, calculate the ratio of described polymerization value growth rate k and the average increment m per second of described single Service Instance, carry out resource expansion according to described ratio.
Known from the above description, the resource extended method of cloud platform ELB assembly of the present invention, efficiently responds the request response of high load condition fast, improves the responding ability of cloud platform, improves resource utilization, is applicable to practical application.
As an embodiment, the maximum service request quantity maxRpt that described single Service Instance is accepted passes through formula: maxRpt=Rps × Δ t × AUR is definite, and the average increment m per second of described single Service Instance passes through formula:
determine, wherein Rps represents single Service Instance maximum service request quantity of accepting per second, and Δ t represents the time interval of twice tasks carrying, and AUR represents high alarm setting, and t represents that Service Instance starts time delay;
The maximum service request quantity of accepting by the single Service Instance of calculation of parameter configuring in cloud platform ELB assembly, the parameter value of the average increment per second of single Service Instance, predict future peaking capacity, pre-cooling Service Instance, has very high using value.
As an embodiment, described polymerization value growth rate k passes through formula:
determine, wherein vi (t) represents the service request quantity of t moment i node in historical data, and n represents node sum, and Δ t represents the time interval of twice tasks carrying, and guarantee subsequent treatment is normally carried out, and is applicable to practical application.
As an embodiment, the automatic elastic spreading parameter of the cloud platform ELB assembly reading in described basis and service cluster parameter also comprise step before calculating:
Automatic elastic spreading parameter and the service cluster parameter of configuration cloud platform ELB assembly, according to actual conditions configuration parameter, fully practical requirement, is applicable to application.
As an embodiment, after the automatic elastic spreading parameter and service cluster parameter of described configuration cloud platform ELB assembly, the automatic elastic spreading parameter of the cloud platform ELB assembly that described basis reads and service cluster parameter also comprise step before calculating:
The log information of configuration cloud platform ELB assembly, realizes journal function, follows the tracks of detailed algorithm design flow process and computational methods, is applicable to practical application.
In order to understand better this method, below elaborate a present invention and be applied to the application example of the resource extended method of cloud platform ELB assembly, this application example is realized in WSO2-StratOS PaaS cloud platform ELB assembly.
WSO2Stratos is a PaaS platform of increasing income, it is mainly enterprise-level application development the solution based on cloud is provided, WSO2Stratos externally provides ApplicationServer (application server) at present, 16 kinds of serviced components such as Elastic Load Balancer (elastic load equalizer), every kind of serviced component all can be responsible for providing specific function, and all serviced components intercom mutually by WebService (application program of a web), finally form WSO2Stratos platform by all these serviced components, WSO2Stratos overall architecture, the solution that WSO2Stratos externally provides as a PaaS platform, all to realize by combining different serviced components.
As shown in Figure 3, this application example can comprise the following steps:
Step S301: automatic elastic spreading parameter, the service cluster parameter of configuration WSO2-StratOS PaaS cloud platform ELB assembly:
Automatic elastic spreading parameter configuration: (1) starts resilient expansion, automatic extension under default situations in ELB is closed, need to arrive in <ELB_HOME>/repository/conf/lo adbalancer.conf file, to ' loadbalancer ' amendment attribute, ' enable_autoscaler '=' true '; (2) autoscaler_task_delay represents two time intervals between tasks carrying; (3) server_startup_delay represents the maximum duration cycle that automatic expansion task was waited for before starting, and refers to the time delay that adds ELB due to network problem Service Instance; (4) instances represents minimum service example number in ELB; (5) enable_autoscaler represents whether to start automatic elastic expansion; (6) autoscaler_task represents that automatic elastic expansion determines generation task, is configured to org.wso2.carbon.mediator.autoscale.lbautoscale.task.Serv iceRequestsIn FlightAutoscaler; (7) use_embedded_autoscaler represents whether use built-in automatic elastic expander; (8) size_of_cache represents cache size; (9) autoscaler_service_epr represents that automatic elastic expander service terminal point quotes, and is configured to https: //host_address:https_port/services/Autoscaler
service; (10) autoscaler_task_interval represents the time interval (Δ t, millisecond) of twice tasks carrying; (11) server_startup_interval represents that Service Instance starts time delay (t); (12) session_timeout represents session timeout; (13) whether fail_over standby starts;
Service cluster parameter configuration: (1) min_app_instances represents minimum service example number, the downward expansion of automatic expansion device can be lower than this value; (2) max_app_instances represents maximum Service Instance number, and automatic expansion device is upwards expanded and can not be exceeded this value; (3) the single Service Instance of max_requests_per_second (Rps) maximum service request quantity of accepting per second; (4) iterations of rounds_to_average (r) automatic expansion; (5) alarming_upper_rate (AUR) represents high alarm setting, and span is 0～1, and acquiescence is 0.7; (6) message_expiry_time represents message expired time;
Step S302: the log information of configuration WSO2-StratOS PaaS cloud platform ELB assembly, in <ELB_HOME>/repository/conf/lo g4j.properties file, add log4j.logger.org.wso2.carbon.mediator.autoscale.lbautosc ale.task.ServiceRequestsInFlightAutoscaler=DEBUG and restart server and realize journal function, follow the tracks of detailed algorithm design flow process and computational methods;
Step S303: the automatic elastic spreading parameter, the service cluster parameter that read WSO2-StratOS PaaS cloud platform ELB assembly, the average increment m per second of the maximum service request quantity maxRpt accepting according to the single Service Instance of the calculation of parameter reading and single Service Instance, the maximum service request quantity maxRpt that wherein single Service Instance is accepted passes through formula: maxRpt=Rps × Δ t × AUR is definite, and the average increment m per second of single Service Instance passes through formula:
determine, wherein Rps represents single Service Instance maximum service request quantity of accepting per second, and Δ t represents the time interval of twice tasks carrying, and AUR represents high alarm setting, and t represents that Service Instance starts time delay;
Step S304: calculate polymerization value Vtotal, average nodal polymerization value Vavg and polymerization value growth rate k according to service request quantity in the historical data of obtaining, wherein polymerization value Vtotal passes through formula:
determine, average nodal polymerization value Vavg passes through formula:
determine, polymerization value growth rate k passes through formula:
determine, wherein vi (t) represents the service request quantity of t moment i node in historical data, and n represents node sum, and Δ t represents the time interval of twice tasks carrying, node sum n=1 in this application example;
Step S305: real time service request quantity and Service Instance maximum processing capability are compared to the maximum service request quantity maxRpt that this Service Instance maximum processing capability is accepted by above-mentioned single Service Instance and real time execution Service Instance number n
in real timeproduct determine;
Step S306: in the time that real time service request quantity is greater than above-mentioned Service Instance maximum processing capability, calculate the ratio of above-mentioned polymerization value growth rate k and the average increment m per second of single Service Instance, carry out resource expansion according to this ratio;
Configuration parameter explanation: the time interval Δ t60000ms of twice tasks carrying, single Service Instance maximum service request quantity Rps5, the high alarm setting AUR0.7 accepting per second, minimum service example number 1, Service Instance start time delay t180000ms, and (Δ is 60000ms, max_requests_per_second (Rps) 5, alarming_upper_rate (AUR) 0.7, min_app_instances1, server_startup_interval (t) 180000ms t) for autoscaler_task_interval;
Calculate the maximum service request quantity that single Service Instance is accepted:
The average increment per second of single Service Instance:
The process one of carrying out resource expansion in WSO2-StratOS PaaS cloud platform ELB assembly can be as shown in table 1 below;
Table 1 process one
The 3rd iteration real time service request quantity as shown in table 1 is greater than Service Instance maximum processing capability, is greater than maximum service request quantity maxRpt and real time execution Service Instance number n that described single Service Instance is accepted
in real timeproduct, be 250>210, calculate the ratio of polymerization value growth rate k and the average increment m per second of single Service Instance, carry out resource expansion according to this ratio, the value of the 3rd iteration polymerization value growth rate k is 4.15, the ratio of polymerization value growth rate k and the average increment m per second of single Service Instance
round, obtaining need to be in the 2nd iteration, should start 4 Service Instances and carry out resource expansion, like this in the 3rd iteration, in the situation that considering startup of server time delay, can meet request performance requirement, effectively prevent Service Instance operating overload, cause the problem of request delay or interruption, the process two of carrying out resource expansion in WSO2-StratOS PaaS cloud platform ELB assembly can be as shown in table 2 below;
Table 2 process two
Server_startup_interval in this application example (t) 180000ms, be that Service Instance starts time delay t=180000ms, the time interval of each iteration is 1min, so 4 Service Instances that start in the 2nd iteration, progressively start, when the 3rd iteration, start 1, real time execution Service Instance number n
in real time=2, when the 4th iteration, start 2, real time execution Service Instance number n
in real time=3, when the 5th iteration, start 3, real time execution Service Instance number n
in real timewhen the=4,6th iteration, 4 are started, real time execution Service Instance number n
in real time=5.
In ELB assembly in WSO2-StratOS, introduce polymerization value, average nodal polymerization value, and the parameter value such as polymerization value growth rate, by calculating the ratio of polymerization value growth rate and the average increment per second of single Service Instance, prediction needs the number of the Service Instance starting, therefore, can be before asking peak value arrival next time, the quantity of Service Instance just can meet performance requirement, overcome the service pressure overload that startup of server time delay brings, can not meet the pressure requirement of peak value request, also overcome the intrinsic algorithm errors of ELB assembly in paas framework, reduce the extensive work that history matching brings, improve resource utilization, cost-saving.
The resource expanding unit of an embodiment medium cloud platform ELB assembly, as shown in Figure 4, comprising:
Processing module, automatic elastic spreading parameter and the service cluster parameter of the cloud platform ELB assembly reading for basis, calculate maximum service request quantity maxRpt and the average increment m per second of single Service Instance that single Service Instance is accepted, and calculate polymerization value growth rate k according to service request quantity in the historical data of obtaining;
Comparison module, for real time service request quantity and Service Instance maximum processing capability are compared, the maximum service request quantity maxRpt that described Service Instance maximum processing capability is accepted by described single Service Instance and real time execution Service Instance number n
in real timeproduct determine;
Resource expansion module, in the time that described real time service request quantity is greater than described Service Instance maximum processing capability, calculates the ratio of described polymerization value growth rate k and the average increment m per second of described single Service Instance, carries out resource expansion according to described ratio.
As shown in Figure 4, this preferred embodiment of installing each module annexation is: processing module, comparison module and resource expansion module are linked in sequence successively.
First processing module is according to automatic elastic spreading parameter and the service cluster parameter of the cloud platform ELB assembly reading, calculate maximum service request quantity maxRpt and the average increment m per second of single Service Instance that single Service Instance is accepted, calculate polymerization value growth rate k according to service request quantity in the historical data of obtaining; Then comparison module compares the request of real time service quantity and Service Instance maximum processing capability; In the time that the request of real time service quantity is greater than Service Instance maximum processing capability, resource expansion module calculates the ratio of polymerization value growth rate k and the average increment m per second of single Service Instance, carries out resource expansion according to this ratio.
Known from the above description, the resource expanding unit of cloud platform ELB assembly of the present invention, effectively solves existing response type expansion and causes the problem that user need to experience one period of time of delay, prediction type expansion workload can not meet greatly real work demand, is applicable to application.
As an embodiment, the maximum service request quantity maxRpt that described single Service Instance is accepted passes through formula: maxRpt=Rps × Δ t × AUR is definite, and the average increment m per second of described single Service Instance passes through formula:
determine, wherein Rps represents single Service Instance maximum service request quantity of accepting per second, and Δ t represents the time interval of twice tasks carrying, and AUR represents high alarm setting, and t represents that Service Instance starts time delay;
The maximum service request quantity of accepting by the single Service Instance of calculation of parameter configuring in cloud platform ELB assembly, the parameter value of the average increment per second of single Service Instance, predict future peaking capacity, pre-cooling Service Instance, has very high using value.
As an embodiment, described polymerization value growth rate k passes through formula:
determine, wherein vi (t) represents the service request quantity of t moment i node in historical data, and n represents node sum, and Δ t represents the time interval of twice tasks carrying, and guarantee subsequent treatment is normally carried out, and is applicable to practical application.
As an embodiment, also comprise configuration module, for configuring automatic elastic spreading parameter and the service cluster parameter of cloud platform ELB assembly;
Described processing module is according to automatic elastic spreading parameter and the service cluster parameter of the cloud platform ELB assembly reading from described configuration module, calculate maximum service request quantity maxRpt and the average increment m per second of single Service Instance that single Service Instance is accepted, according to actual conditions configuration parameter, fully practical requirement, is applicable to application.
As an embodiment, described configuration module also, for configuring the log information of cloud platform ELB assembly, is realized journal function, follows the tracks of detailed algorithm design flow process and computational methods, is applicable to practical application.
The above embodiment has only expressed several execution mode of the present invention, and it describes comparatively concrete and detailed, but can not therefore be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection range of patent of the present invention should be as the criterion with claims.
Claims (10)
1. a resource extended method for cloud platform ELB assembly, is characterized in that, comprises the following steps:
According to automatic elastic spreading parameter and the service cluster parameter of the cloud platform ELB assembly reading, calculate maximum service request quantity maxRpt and the average increment m per second of single Service Instance that single Service Instance is accepted;
Calculate polymerization value growth rate k according to service request quantity in the historical data of obtaining;
Real time service request quantity and Service Instance maximum processing capability are compared to the maximum service request quantity maxRpt that described Service Instance maximum processing capability is accepted by described single Service Instance and real time execution Service Instance number n
in real timeproduct determine;
In the time that described real time service request quantity is greater than described Service Instance maximum processing capability, calculate the ratio of described polymerization value growth rate k and the average increment m per second of described single Service Instance, carry out resource expansion according to described ratio.
2. the resource extended method of cloud platform ELB assembly according to claim 1, it is characterized in that, the maximum service request quantity maxRpt that described single Service Instance is accepted passes through formula: maxRpt=Rps × Δ t × AUR is definite, and the average increment m per second of described single Service Instance passes through formula:
t is definite, and wherein Rps represents single Service Instance maximum service request quantity of accepting per second, and Δ t represents the time interval of twice tasks carrying, and AUR represents high alarm setting, and t represents that Service Instance starts time delay.
3. the resource extended method of cloud platform ELB assembly according to claim 1 and 2, is characterized in that, described polymerization value growth rate k passes through formula:
determine, wherein vi (t) represents the service request quantity of t moment i node in historical data, and n represents node sum, and Δ t represents the time interval of twice tasks carrying.
4. the resource extended method of cloud platform ELB assembly according to claim 1, is characterized in that, the automatic elastic spreading parameter of the cloud platform ELB assembly reading in described basis and service cluster parameter also comprise step before calculating:
Automatic elastic spreading parameter and the service cluster parameter of configuration cloud platform ELB assembly.
5. the resource extended method of cloud platform ELB assembly according to claim 4, it is characterized in that, after the automatic elastic spreading parameter and service cluster parameter of described configuration cloud platform ELB assembly, the automatic elastic spreading parameter of the cloud platform ELB assembly that described basis reads and service cluster parameter also comprise step before calculating:
The log information of configuration cloud platform ELB assembly.
6. a resource expanding unit for cloud platform ELB assembly, is characterized in that, comprising:
Processing module, automatic elastic spreading parameter and the service cluster parameter of the cloud platform ELB assembly reading for basis, calculate maximum service request quantity maxRpt and the average increment m per second of single Service Instance that single Service Instance is accepted, and calculate polymerization value growth rate k according to service request quantity in the historical data of obtaining;
Comparison module, for real time service request quantity and Service Instance maximum processing capability are compared, the maximum service request quantity maxRpt that described Service Instance maximum processing capability is accepted by described single Service Instance and real time execution Service Instance number n
in real timeproduct determine;
Resource expansion module, in the time that described real time service request quantity is greater than described Service Instance maximum processing capability, calculates the ratio of described polymerization value growth rate k and the average increment m per second of described single Service Instance, carries out resource expansion according to described ratio.
7. the resource expanding unit of cloud platform ELB assembly according to claim 6, it is characterized in that, the maximum service request quantity maxRpt that described single Service Instance is accepted passes through formula: maxRpt=Rps × Δ t × AUR is definite, and the average increment m per second of described single Service Instance passes through formula:
determine, wherein Rps represents single Service Instance maximum service request quantity of accepting per second, and Δ t represents the time interval of twice tasks carrying, and AUR represents high alarm setting, and t represents that Service Instance starts time delay.
8. according to the resource expanding unit of the cloud platform ELB assembly described in claim 6 or 7, it is characterized in that, described polymerization value growth rate k passes through formula:
determine, wherein vi (t) represents the service request quantity of t moment i node in historical data, and n represents node sum, and Δ t represents the time interval of twice tasks carrying.
9. the resource expanding unit of cloud platform ELB assembly according to claim 6, is characterized in that, also comprises configuration module, for configuring automatic elastic spreading parameter and the service cluster parameter of cloud platform ELB assembly;
Described processing module, according to automatic elastic spreading parameter and the service cluster parameter of the cloud platform ELB assembly reading from described configuration module, is calculated maximum service request quantity maxRpt and the average increment m per second of single Service Instance that single Service Instance is accepted.
10. the resource expanding unit of cloud platform ELB assembly according to claim 9, is characterized in that, described configuration module is also for configuring the log information of cloud platform ELB assembly.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410375345.5A CN104135525B (en) | 2014-07-31 | 2014-07-31 | The resource expansion method and apparatus of cloud platform ELB components |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410375345.5A CN104135525B (en) | 2014-07-31 | 2014-07-31 | The resource expansion method and apparatus of cloud platform ELB components |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104135525A true CN104135525A (en) | 2014-11-05 |
| CN104135525B CN104135525B (en) | 2017-12-01 |
Family
ID=51808047
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410375345.5A Expired - Fee Related CN104135525B (en) | 2014-07-31 | 2014-07-31 | The resource expansion method and apparatus of cloud platform ELB components |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104135525B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105068919A (en) * | 2015-07-15 | 2015-11-18 | 华南理工大学 | WSO2 Stratos-based WS-BPEL flow load testing tool |
| CN107643975A (en) * | 2017-09-25 | 2018-01-30 | 郑州云海信息技术有限公司 | A kind of method, apparatus and computer-readable recording medium for counting pressure information |
| CN113254224A (en) * | 2021-07-15 | 2021-08-13 | 中电金信软件有限公司 | Computing resource capacity expansion method and device, electronic equipment and readable storage medium |
| CN113783799A (en) * | 2020-06-09 | 2021-12-10 | 马上消费金融股份有限公司 | Flow control method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103220337A (en) * | 2013-03-22 | 2013-07-24 | 合肥工业大学 | Cloud computing resource optimizing collocation method based on self-adaptation elastic control |
| CN103425535A (en) * | 2013-06-05 | 2013-12-04 | 浙江大学 | Agile elastic telescoping method in cloud environment |
| CN103778474A (en) * | 2012-10-18 | 2014-05-07 | 华为技术有限公司 | Resource load capacity prediction method, analysis prediction system and service operation monitoring system |
-
2014
- 2014-07-31 CN CN201410375345.5A patent/CN104135525B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103778474A (en) * | 2012-10-18 | 2014-05-07 | 华为技术有限公司 | Resource load capacity prediction method, analysis prediction system and service operation monitoring system |
| CN103220337A (en) * | 2013-03-22 | 2013-07-24 | 合肥工业大学 | Cloud computing resource optimizing collocation method based on self-adaptation elastic control |
| CN103425535A (en) * | 2013-06-05 | 2013-12-04 | 浙江大学 | Agile elastic telescoping method in cloud environment |
Non-Patent Citations (1)
| Title |
|---|
| 吴和生等: ""TeraPELB:云计算中基于预测的弹性负载均衡算法"", 《系统仿真学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105068919A (en) * | 2015-07-15 | 2015-11-18 | 华南理工大学 | WSO2 Stratos-based WS-BPEL flow load testing tool |
| CN105068919B (en) * | 2015-07-15 | 2018-04-13 | 华南理工大学 | A kind of WS BPEL flow load testing instruments based on WSO2 Stratos |
| CN107643975A (en) * | 2017-09-25 | 2018-01-30 | 郑州云海信息技术有限公司 | A kind of method, apparatus and computer-readable recording medium for counting pressure information |
| CN113783799A (en) * | 2020-06-09 | 2021-12-10 | 马上消费金融股份有限公司 | Flow control method and device |
| CN113254224A (en) * | 2021-07-15 | 2021-08-13 | 中电金信软件有限公司 | Computing resource capacity expansion method and device, electronic equipment and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104135525B (en) | 2017-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109766182B (en) | System resource dynamic expansion and contraction method and device, computer equipment and storage medium | |
| US8782215B2 (en) | Performance testing in a cloud environment | |
| Liu et al. | Aggressive resource provisioning for ensuring QoS in virtualized environments | |
| CN110677305B (en) | Automatic scaling method and system in cloud computing environment | |
| CN107247651B (en) | Cloud computing platform monitoring and early warning method and system | |
| WO2020253111A1 (en) | Automatic expansion method and apparatus for blockchain node, and operation and maintenance terminal and storage medium | |
| CN111966289B (en) | Partition optimization method and system based on Kafka cluster | |
| Fu et al. | Layered virtual machine migration algorithm for network resource balancing in cloud computing | |
| CN112003763B (en) | Network link monitoring method, monitoring device, monitoring equipment and storage medium | |
| WO2016169166A1 (en) | Virtual machine scheduling method and device | |
| KR101630125B1 (en) | Method for resource provisioning in cloud computing resource management system | |
| CN104135525A (en) | Resource expanding method and device for cloud platform ELB components | |
| WO2014194704A1 (en) | A grouping processing method and system | |
| CN107402851B (en) | Data recovery control method and device | |
| CN106464733A (en) | Method and device for adjusting virtual resources in cloud computing | |
| US11669374B2 (en) | Using machine-learning methods to facilitate experimental evaluation of modifications to a computational environment within a distributed system | |
| CN112214303A (en) | Kubernetes cluster automatic scaling system | |
| CN107493333B (en) | SDN-based web system resource management method and system in cloud computing environment | |
| Oral et al. | Supporting performance isolation in software as a service systems with rich clients | |
| CN115048186A (en) | Method and device for processing expansion and contraction of service container, storage medium and electronic equipment | |
| CN114880079A (en) | Kubernetes cluster scale adjustment method, system and equipment based on reinforcement learning | |
| Debbi | Modeling and performance analysis of resource provisioning in cloud computing using probabilistic model checking | |
| CN113747506A (en) | Resource scheduling method, device and network system | |
| US20140359104A1 (en) | Grouping processing method and system | |
| Skulysh et al. | Management of multiple stage queuing systems |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171201 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |