CN103490956A - Self-adaptive energy-saving control method, device and system based on traffic predication - Google Patents
Self-adaptive energy-saving control method, device and system based on traffic predication Download PDFInfo
- Publication number
- CN103490956A CN103490956A CN201310433553.1A CN201310433553A CN103490956A CN 103490956 A CN103490956 A CN 103490956A CN 201310433553 A CN201310433553 A CN 201310433553A CN 103490956 A CN103490956 A CN 103490956A
- Authority
- CN
- China
- Prior art keywords
- carrying capacity
- traffic carrying
- server cluster
- server
- traffic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012937 correction Methods 0.000 claims abstract description 17
- 230000003044 adaptive effect Effects 0.000 claims description 30
- 238000013528 artificial neural network Methods 0.000 claims description 11
- 238000013459 approach Methods 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 8
- 230000000052 comparative effect Effects 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 description 11
- 230000000694 effects Effects 0.000 description 7
- 238000004134 energy conservation Methods 0.000 description 7
- 238000005265 energy consumption Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a self-adaptive energy-saving control method, device and system based on traffic predication. The method includes the steps that various indexes relevant to traffic of a server cluster in all historical time periods are collected; historical traffic of the server cluster in all the historical time periods is calculated by utilizing the various indexes relevant to the traffic in all the historical time periods; according to the historical traffic in all the historical time periods, traffic of the server cluster in the next time period is predicated; by utilizing correction factors corresponding to date classification which the next time period belongs to, the traffic in the next time period is corrected; if the largest historical traffic is larger than or equal to traffic of the next time period after being corrected, the number and/or the work states of servers operating in the service cluster are adjusted according to the corrected traffic in the next time period; when the work states of the operating servers are different, the traffics processed by the operating servers are different. Therefore, self-adaptive energy-saving control can be performed on the server cluster.
Description
Technical field
The present invention relates to the server cluster technical field, relate in particular to the self-adapting energy-saving control method of a kind of service based amount prediction and equipment, system.
Background technology
Under cloud computing and large data background, the scale of server cluster sharply increases, and the energy consumption problem of thing followed server cluster also is on the rise.In actual applications, the traffic carrying capacity of server, in the different time, for example there will be operating time, weekend, festivals or holidays larger fluctuation.Wherein, when traffic carrying capacity is at a low ebb, the energy consumption of server is very low, when traffic carrying capacity during in peak the energy consumption of server very high.Because server cluster is deployed in operator usually, and the service performance that operator externally provides in order to ensure server cluster is unaffected, usually can require the server in server cluster to move according to traffic peak, thereby can cause very large energy consumption.
Summary of the invention
The embodiment of the invention discloses the self-adapting energy-saving control method of a kind of service based amount prediction and equipment, system, can carry out the efficient adaptive Energy Saving Control to server cluster.
Embodiment of the present invention first aspect discloses a kind of self-adapting energy-saving control method of service based amount prediction, comprising:
Collect the indices relevant to traffic carrying capacity of each historical time section of server cluster;
Utilize the indices relevant to traffic carrying capacity of each historical time section of described server cluster to calculate respectively the historical traffic of each historical time section of described server cluster;
Predict the traffic carrying capacity of described next time period of server cluster according to the historical traffic of each historical time section of described server cluster;
Utilize date under described next time period corresponding correction factor of classifying to be revised the traffic carrying capacity of described next time period, obtain the traffic carrying capacity of revised next time period;
Whether the historical traffic peak relatively calculated is more than or equal to the traffic carrying capacity of described revised next time period, if not, adjust the operating state of the server of the number of servers moved in described server cluster and/or operation according to the traffic carrying capacity of described revised next time period; The traffic carrying capacity difference of the server of described operation when wherein, the operating state of the server of described operation is different.
In the first of embodiment of the present invention first aspect in possible implementation, after the described traffic carrying capacity according to described revised next time period is adjusted the operating state of server of the number of servers moved in described server cluster and/or operation, described method also comprises:
Monitor the service quality of every business in described server cluster;
Whether the service quality that judges every business in described server cluster exceeds default tolerance range, if do not exceed, whether the traffic carrying capacity that judges next time period is greater than described historical traffic peak, if be greater than, described historical traffic peak is updated to the traffic carrying capacity of described next time period.
At the second of embodiment of the present invention first aspect in possible implementation, the historical traffic that the described indices relevant to traffic carrying capacity that utilizes each historical time section of described server cluster calculates respectively each historical time section of described server cluster comprises:
Utilize the indices relevant to traffic carrying capacity of each historical time section of described server cluster, and the binding hierarchy analytic approach is calculated respectively the historical traffic of each historical time section of described server cluster.
The first or the possible implementation of the second in conjunction with embodiment of the present invention first aspect or embodiment of the present invention first aspect, in the third possible implementation of embodiment of the present invention first aspect, the described historical traffic according to each historical time section of described server cluster predicts that the traffic carrying capacity of described next time period of server cluster comprises:
Using the historical traffic of each historical time section of described server cluster as output variable, and in conjunction with neural network prediction algorithm or the traffic carrying capacity of described next time period of server cluster of grey system forecasting algorithm predicts.
The third possible implementation in conjunction with embodiment of the present invention first aspect, in the 4th kind of possible implementation of embodiment of the present invention first aspect, the operating state that the described traffic carrying capacity according to described revised next time period is adjusted the server of the number of servers moved in described server cluster and/or operation comprises:
Calculate the traffic carrying capacity of each server in each the class server in described server cluster under current time, wherein, the maximum that the traffic carrying capacity of described each server equals described server allows the product of the traffic carrying capacity coefficient of regime corresponding with the operating state of described server under current time, coefficient of regime corresponding to the operating state of described server is greater than 0, and is less than or equal to 1; Wherein, when the operating state of described server is different, the coefficient of regime difference that the operating state of described server is corresponding;
According to the traffic carrying capacity of each server in each the class server in described server cluster under current time, calculate the traffic carrying capacity of each the class server in described server cluster under current time;
According to the traffic carrying capacity of each the class server in described server cluster under described current time, calculate the traffic carrying capacity of described server cluster under described current time;
Calculate the traffic carrying capacity of described server cluster under described current time and the difference of the traffic carrying capacity of described revised next time period;
Adjust the operating state of the server of the number of servers moved in described server cluster and/or operation according to described difference.
In conjunction with the 4th kind of possible implementation of embodiment of the present invention first aspect, in the 5th kind of possible implementation of embodiment of the present invention first aspect, the described indices relevant to traffic carrying capacity comprises number of concurrent, number of users and page browsing amount.
Embodiment of the present invention second aspect discloses a kind of adaptive power conservation control appliance of service based amount prediction, comprising:
Collector unit, for collecting the indices relevant to traffic carrying capacity of each historical time section of server cluster;
Statistic unit, calculate respectively the historical traffic of each historical time section of described server cluster for the indices relevant to traffic carrying capacity that utilizes each historical time section of described server cluster that described collector unit collects;
Predicting unit, predict the traffic carrying capacity of described next time period of server cluster for the historical traffic of each historical time section of described server cluster of calculating according to described statistic unit;
Amending unit, revised the traffic carrying capacity of described next time period of described predicting unit prediction for utilizing date under described next time period corresponding correction factor of classifying, and obtains the traffic carrying capacity of revised next time period;
Comparing unit, whether the historical traffic peak calculated for more described statistic unit is more than or equal to the traffic carrying capacity of described revised next time period of described amending unit acquisition;
Adjustment unit, for at the comparative result of described comparing unit while being no, the traffic carrying capacity of described revised next time period obtained according to described amending unit is adjusted the operating state of the server of the number of servers moved in described server cluster and/or operation; The traffic carrying capacity difference of the server of described operation when wherein, the operating state of the server of described operation is different.
In the first of embodiment of the present invention second aspect, in possible implementation, described equipment also comprises:
Monitoring unit, after the operating state for the server of adjusting number of servers that described server cluster moves and/or operation at described adjustment unit, monitor the service quality of every business in described server cluster;
Whether judging unit, exceed default tolerance range for the service quality of every business judging described server cluster, if do not exceed, judges whether the traffic carrying capacity of next time period is greater than described historical traffic peak;
Updating block, while for the traffic carrying capacity that goes out described next time period in described judgment unit judges, being greater than described historical traffic peak, be updated to described historical traffic peak the traffic carrying capacity of described next time period.
At the second of embodiment of the present invention second aspect in possible implementation, described statistic unit is specifically for utilizing the indices relevant to traffic carrying capacity of each historical time section of described server cluster, and the binding hierarchy analytic approach is calculated respectively the historical traffic of each historical time section of described server cluster.
The first or the possible implementation of the second in conjunction with embodiment of the present invention second aspect or embodiment of the present invention second aspect, in the third possible implementation of embodiment of the present invention second aspect, described predicting unit, specifically for the historical traffic using each historical time section of described server cluster as output variable, and in conjunction with neural network prediction algorithm or the traffic carrying capacity of described next time period of server cluster of grey system forecasting algorithm predicts.
In conjunction with the third possible implementation of embodiment of the present invention second aspect, in the 4th kind of possible implementation of embodiment of the present invention second aspect, described adjustment unit comprises:
The first computation subunit, for at the comparative result of described comparing unit while being no, calculate the traffic carrying capacity of each server in each the class server in described server cluster under current time, wherein, the maximum that the traffic carrying capacity of described each server equals described server allows the product of the traffic carrying capacity coefficient of regime corresponding with the operating state of described server under current time, coefficient of regime corresponding to the operating state of described server is greater than 0, and is less than or equal to 1; Wherein, when the operating state of described server is different, the coefficient of regime difference that the operating state of described server is corresponding;
The second computation subunit, for the traffic carrying capacity of each server in each the class server according to described server cluster under current time, calculate the traffic carrying capacity of each the class server in described server cluster under current time;
The 3rd computation subunit, for the traffic carrying capacity of each the class server according to described server cluster under described current time, calculate the traffic carrying capacity of described server cluster under described current time;
The 4th computation subunit, for the traffic carrying capacity of calculating described server cluster under described current time and the difference of the traffic carrying capacity of described revised next time period;
Adjust subelement, for the operating state of the server of adjusting number of servers that described server cluster moves and/or operation according to described difference.
In conjunction with the 4th kind of possible implementation of embodiment of the present invention second aspect, in the 5th kind of possible implementation of embodiment of the present invention second aspect, the described indices relevant to traffic carrying capacity comprises number of concurrent, number of users and page browsing amount.
The embodiment of the present invention third aspect discloses a kind of server cluster system, described server cluster system comprises the adaptive power conservation control appliance of server cluster and the disclosed described service based amount prediction of embodiment of the present invention second aspect, wherein, the adaptive power conservation control appliance of described service based amount prediction is connected with each server communication in described server cluster.
In the embodiment of the present invention, after the historical traffic of each historical time section of calculation server cluster, can be according to the traffic carrying capacity of historical traffic predictive server next time period of cluster of each historical time section of server cluster, and utilize date under next time period corresponding correction factor of classifying to be revised the traffic carrying capacity of next time period, obtain the traffic carrying capacity of revised next time period, when the historical traffic peak calculated is less than the traffic carrying capacity of revised next time period, can adjust according to the traffic carrying capacity of revised next time period the operating state of the server of the number of servers moved in server cluster and/or operation, the traffic carrying capacity difference of the server of operation when wherein, the operating state of the server of operation is different.Visible, the embodiment of the present invention can be predicted the traffic carrying capacity of next time period according to historical traffic, and by the traffic carrying capacity of prediction, the operating state of the server of the number of servers moved in server cluster and/or operation is adjusted, thereby can carry out the efficient adaptive Energy Saving Control to server cluster, reach energy-conservation effect.
The accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, below will the accompanying drawing of required use in embodiment be briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is the schematic flow sheet of the self-adapting energy-saving control method of the disclosed a kind of service based amount prediction of the embodiment of the present invention;
Fig. 2 is the schematic flow sheet of the self-adapting energy-saving control method of the disclosed another kind of service based amount prediction of the embodiment of the present invention;
Fig. 3 is the structural representation of the adaptive power conservation control appliance of the disclosed a kind of service based amount prediction of the embodiment of the present invention;
Fig. 4 is the structural representation of the adaptive power conservation control appliance of the disclosed another kind of service based amount prediction of the embodiment of the present invention;
Fig. 5 is the structural representation of the adaptive power conservation control appliance of the disclosed another kind of service based amount prediction of the embodiment of the present invention;
Fig. 6 is the structural representation of the disclosed a kind of server cluster system of the embodiment of the present invention.
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 clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, rather than whole embodiment.Embodiment based in the present invention, those of ordinary skills, not making under the creative work prerequisite the every other embodiment obtained, belong to the scope of protection of the invention.
The embodiment of the invention discloses the self-adapting energy-saving control method of a kind of service based amount prediction and equipment, system, can carry out the efficient adaptive Energy Saving Control to server cluster.Below be elaborated respectively.
Refer to Fig. 1, Fig. 1 is the schematic flow sheet of the self-adapting energy-saving control method of the disclosed a kind of service based amount prediction of the embodiment of the present invention.As shown in Figure 1, the self-adapting energy-saving control method of this service based amount prediction can comprise the following steps.
101, collect the indices relevant to traffic carrying capacity of each historical time section of server cluster.
In the embodiment of the present invention, can be collected by control appliance the indices relevant to traffic carrying capacity of each historical time section of server cluster, wherein, control appliance can be equipment or the terminals such as controller, server in physics realization, and the embodiment of the present invention is not construed as limiting.
In the embodiment of the present invention, the indices relevant to traffic carrying capacity can comprise number of concurrent, number of users and page browsing amount (PageView, PV).
102, utilize the historical traffic of indices difference calculation server cluster each the historical time section relevant to traffic carrying capacity of each historical time section of server cluster.
In the embodiment of the present invention, can be utilized by control appliance the indices relevant to traffic carrying capacity of each historical time section of server cluster, and binding hierarchy analytic approach (being not limited to the method) is carried out the historical traffic of each historical time section of calculation server cluster respectively.
Wherein, analytic hierarchy process (AHP) can comprise the following steps:
1, set up hierarchical structure model;
2, construct all judgment matrixs in each level;
3, Mode of Level Simple Sequence and consistency check;
4, level always sorts and consistency check.
In the embodiment of the present invention, control appliance utilizes the indices relevant to traffic carrying capacity of each historical time section of server cluster, and binding hierarchy analytic approach (being not limited to the method) to carry out the detailed process of the historical traffic of each historical time section of calculation server cluster respectively be general knowledge known in those skilled in the art, the embodiment of the present invention is not described in detail.
103, according to the traffic carrying capacity of historical traffic predictive server next time period of cluster of each historical time section of server cluster.
In the embodiment of the present invention, control appliance can be using the historical traffic of each historical time section of server cluster as output variable, and carrys out the traffic carrying capacity of next time period of predictive server cluster in conjunction with neural network prediction algorithm or grey system forecasting algorithm (being not limited to this two kinds of algorithms).
Wherein, the traffic carrying capacity prediction mainly comprises the following steps:
1) calculation stages: using historical traffic as input, calculated according to definite Forecasting Methodology (as neural network prediction algorithm or grey system forecasting algorithm etc. does not enumerate), wherein, different Forecasting Methodologies has larger difference aspect calculating.
2) the data detection stage: the prediction traffic carrying capacity usually once calculated is not necessarily accurate, at this moment need according to the verification rule of setting, the prediction traffic carrying capacity to be tested, for example, in neural net, data detection is exactly a repetition learning, the process of repeatedly adjusting, need according to check situation adjustment model repeatedly.When the prediction traffic carrying capacity meets test condition, show that this model comparatively mates True Data, can use the traffic carrying capacity of the prediction traffic carrying capacity of this model as next time period.
3) upgrade the historical traffic stage: after the traffic carrying capacity that obtains next time period, historical traffic upgrades in time.Proceed to calculate, data detection, adjustment model until meet certain accuracy, is predicted the traffic carrying capacity of next time period repeatedly, upgrades historical traffic, so moves in circles.
In actual applications, while adopting the traffic carrying capacity of neural network prediction next time period of server cluster, can by historical traffic (D1, D2 ... Dk) as the input layer of input vector input neural network, after the network layer neuron of the output variable input neural network of input layer, can dope next time period traffic carrying capacity (D1 ', D2 ',, Dk ').
104, utilize date under next time period corresponding correction factor of classifying to be revised the traffic carrying capacity of next time period, obtain the traffic carrying capacity of revised next time period.
In the embodiment of the present invention, after control appliance dopes the traffic carrying capacity of next time period of server cluster, can utilize date under next time period corresponding correction factor of classifying to be revised the traffic carrying capacity of next time period, obtain the traffic carrying capacity of revised next time period.For instance, if the date under next time period is categorized as working day, so corresponding correction factor can be 1.05; If the date under next time period is categorized as weekend, so corresponding correction factor can be 1.1; If the date under next time period is categorized as festivals or holidays, so corresponding correction factor can be 1.2.
Whether the historical traffic peak 105, relatively calculated is more than or equal to the traffic carrying capacity of revised next time period, the operating state of the number of servers of moving in the traffic carrying capacity adjustment server cluster according to revised next time period if not, and/or the server of operation; The traffic carrying capacity difference of the server process of this operation when wherein, the operating state of the server of operation is different.
In the embodiment of the present invention, control appliance can be adjusted the operating state of the server of the number of servers moved in server cluster and/or operation according to the traffic carrying capacity of revised next time period, specifically comprise the following steps:
11), calculate the traffic carrying capacity of each server in each the class server in server cluster under current time, wherein, the maximum that the traffic carrying capacity of each server equals server allows the product of the traffic carrying capacity coefficient of regime corresponding with the operating state of this server under current time, coefficient of regime corresponding to the operating state of this server is greater than 0, and is less than or equal to 1; Wherein, when the operating state of this server is different, the coefficient of regime difference that the operating state of this server is corresponding.
In the embodiment of the present invention, can server be divided into to the operating states such as S1, S2, S3 according to running statuses such as the CPU of server, internal memory, IO.
12), according to the traffic carrying capacity of each server in each the class server in server cluster under current time, calculate the traffic carrying capacity of each the class server in server cluster under current time.
13), according to the traffic carrying capacity of each the class server in server cluster under current time, calculate the traffic carrying capacity of server cluster under current time.
The implementation of in the embodiment of the present invention, above-mentioned steps 11)～step 13) can adopt following formula (1) to mean, under current time, the traffic carrying capacity Dreal of server cluster is:
Wherein, N1 means the quantity of the 1st class server, and N2 means the quantity of the 2nd class server ..., Nn means the quantity of n class server; Data1 means that the maximum of the 1st class server allows traffic carrying capacity, and Data2 means that the maximum of the 2nd class server allows traffic carrying capacity ..., Datan means that the maximum of n class server allows traffic carrying capacity; I means i server in each class server, and Si means the coefficient of regime corresponding to operating state of i server in each class server, wherein, and 0<Si≤1.
14), calculate the traffic carrying capacity of server cluster under current time and the difference of the traffic carrying capacity of revised next time period.
In the embodiment of the present invention, above-mentioned steps 14) the middle traffic carrying capacity of server cluster under current time and the difference DELTA D of the traffic carrying capacity of revised next time period of calculating, that is:
Wherein, D_pre(1+J_*) mean the traffic carrying capacity of revised next time period, the number of servers of all kinds of operations in the server cluster that △ N1～△ N2 means to need to adjust, the operating state of the server of all kinds of operations in the server cluster that Δ Si means to need to adjust.
As a kind of optional execution mode, in the embodiment of the present invention, after the operating state of the number of servers of moving in the traffic carrying capacity adjustment server cluster of control appliance according to revised next time period and/or the server of operation, can also carry out following steps:
The service quality of every business in the monitoring server cluster;
Whether the service quality that judges every business in server cluster exceeds default tolerance range, if do not exceed, whether the traffic carrying capacity that judges next time period is greater than historical traffic peak, if be greater than, historical traffic peak is updated to the traffic carrying capacity of next time period.
In the described method of Fig. 1, control appliance can be predicted the traffic carrying capacity of next time period according to historical traffic, and by the traffic carrying capacity of prediction, the operating state of the server of the number of servers moved in server cluster and/or operation is adjusted, thereby can carry out the efficient adaptive Energy Saving Control to server cluster, reach energy-conservation effect.
Refer to Fig. 2, Fig. 2 is the schematic flow sheet of the self-adapting energy-saving control method of the disclosed another kind of service based amount prediction of the embodiment of the present invention.As shown in Figure 2, the self-adapting energy-saving control method of this service based amount prediction can comprise the following steps.
201, control appliance is collected the indices relevant to traffic carrying capacity of each historical time section of server cluster.
Wherein, the indices relevant to traffic carrying capacity can comprise number of concurrent, number of users and page browsing amount (PV).
202, control appliance utilizes the historical traffic of indices difference calculation server cluster each the historical time section relevant to traffic carrying capacity of each historical time section of server cluster.
203, control appliance is preserved the historical traffic of each historical time section of server cluster, and determines historical traffic peak.
204, control appliance is according to the traffic carrying capacity of historical traffic predictive server next time period of cluster of each historical time section of server cluster.
205, control appliance utilizes date under next time period corresponding correction factor of classifying to be revised the traffic carrying capacity of next time period, obtains the traffic carrying capacity of revised next time period.
Whether the historical traffic peak that 206, control appliance relatively calculates is more than or equal to the traffic carrying capacity of revised next time period, if not, performs step 207; If so, perform step 213.
The operating state of the number of servers of 207, moving in the traffic carrying capacity adjustment server cluster of control appliance according to revised next time period and/or the server of operation, and perform step 208.
The traffic carrying capacity difference of the server of this operation when wherein, the operating state of the server of operation is different.
After the operating state of the number of servers of 208, moving in the traffic carrying capacity adjustment server cluster of control appliance according to revised next time period and/or the server of operation, carry out business migration, and perform step 209.
209, the service quality of every business in control appliance monitoring server cluster, and perform step 210.
210, control appliance judges that whether the service quality of every business in server cluster exceeds default tolerance range, if exceed, performs step 211; If do not exceed, perform step 217.
211, control appliance is further adjusted the operating state of the server of the number of servers moved in server cluster and/or operation, and performs step 212.
212, control appliance judges whether full speed running of server, if not, returns to step 209; If so, perform step 214.
In the embodiment of the present invention, the server full speed running refers to the operation of the maximum permission of server traffic carrying capacity.
213, control appliance Control Server full speed running, and perform step 214.
214, the service quality of every business in control appliance monitoring server cluster, and perform step 215.
215, control appliance judges that whether the service quality of every business in server cluster exceeds default tolerance range, if exceed, performs step 216; If do not exceed, perform step 217.
216, the control appliance alarm need to increase the number of servers of operation.
217, control appliance judges whether the traffic carrying capacity of next time period is greater than historical traffic peak, if be less than or equal to, keeps historical traffic peak constant; If be greater than, historical traffic peak is updated to the traffic carrying capacity of next time period, to continue the traffic carrying capacity of another next time period of prediction.
In the described method of Fig. 2, control appliance can be predicted the traffic carrying capacity of next time period according to historical traffic, and by the traffic carrying capacity of prediction, the operating state of the server of the number of servers moved in server cluster and/or operation is adjusted, thereby can carry out the efficient adaptive Energy Saving Control to server cluster, reach energy-conservation effect.
In order to understand better the present invention, below further by specific embodiment, be described.
Example one: conventional embodiment
Wherein, preset following condition:
1. establishing the indices relevant to traffic carrying capacity is number of concurrent, number of users and PV.
2. establish in server cluster and have 2 class servers, the quantity of all kinds of servers is respectively 1 and 1.
3. the maximum of establishing the 1st, 2 class servers allows traffic carrying capacity to be respectively (100,150).
4. establish the 1st, 2 class servers and work in S1, S2 state, corresponding coefficient of regime is respectively 1 and 0.9, and coefficient of regime corresponding to S3 is 0.96.
5. establishing the date is categorized as working day, weekend and festivals or holidays, and wherein, working day, weekend and festivals or holidays, corresponding correction factor was respectively 1.05,1.1,1.2.
6. the indices relevant to traffic carrying capacity of establishing each historical time section of server cluster is respectively:
| ? | Number of users | Number of concurrent | PV |
| Historical time section 1 | 100 | 102 | 100 |
| Historical time section 2 | 200 | 202 | 210 |
| Historical time section 3 | 104 | 102 | 105 |
In the embodiment of the present invention, the indices relevant to traffic carrying capacity that control appliance can utilize each historical time section of the above-mentioned server cluster historical traffic of each historical time section of calculation server cluster respectively is:
| ? | Historical traffic |
| Historical time section 1 | 200 |
| Historical time section 2 | 230 |
| Historical time section 3 | 205 |
From the historical traffic of each historical time section of above-mentioned server cluster, can find out, historical traffic peak is 230, supposes that the date under next time period is categorized as weekend, can show that correction factor is 1.1.
Predict that according to the traffic carrying capacity prediction algorithm next traffic carrying capacity constantly is 210, utilize so date under next time period corresponding correction factor 1.1 of classifying to be revised the traffic carrying capacity 210 of next time period, the traffic carrying capacity that obtains revised next time period is 231.
Wherein, the traffic carrying capacity 231 of amended next time period is less than historical traffic peak, needs the operating state of the server of the number of servers of adjust operation and/or operation.
The traffic carrying capacity Dreal that particularly, can calculate server cluster under current time according to formula (3) is:
Wherein, N1 means the quantity of the 1st class server, and N2 means the quantity of the 2nd class server, and Data1 means that the maximum of the 1st class server allows traffic carrying capacity 100,, Data2 means that the maximum of the 2nd class server allows traffic carrying capacity 150; I means i server in each class server, and Si means the coefficient of regime corresponding to operating state of i server in each class server, wherein, and 0<Si≤1.
Particularly, can calculate the traffic carrying capacity Dreal of server cluster under current time and the difference DELTA D of the traffic carrying capacity of revised next time period according to formula (4), that is:
Can draw according to difference DELTA D, maximum can be allowed the operating state of the 1st class server that traffic carrying capacity is 100 to be adjusted into operating state S3 by S1, thereby can carry out the efficient adaptive Energy Saving Control to server cluster, reach energy-conservation effect.
Further, if the service quality of every business in server cluster does not exceed default tolerance range service quality (such as service delay), the operating state of the 1st class server is successfully modified.
Further, new historical traffic peak more, to continue the traffic carrying capacity of next time period of prediction.
Example two: traffic carrying capacity increases embodiment
Suppose in above-mentioned example one, predict that according to the traffic carrying capacity prediction algorithm traffic carrying capacity of next time period is 220, the traffic carrying capacity of so revised next time period is 242.Correspondingly, follow-up being treated to:
1. the traffic carrying capacity of next time period is greater than historical traffic peak, need to carry out the operating state of the server of the number of servers of adjust operation and/or operation.
2. the traffic carrying capacity Dreal that calculates server cluster under current time according to formula (3) is:
3. need increasing 7(is 242-235) traffic carrying capacity, if maximum, allow the operating state of the 2nd class server that traffic carrying capacity is 150 to change to S3, the traffic carrying capacity of next time period of server cluster is 100*1+150*0.96=244 so, can meet business demand.
4. new historical traffic peak more.
Refer to Fig. 3, Fig. 3 is the structural representation of the adaptive power conservation control appliance of the disclosed a kind of service based amount prediction of the embodiment of the present invention.As shown in Figure 3, the adaptive power conservation control appliance of this service based amount prediction can comprise:
Predicting unit 303, for the traffic carrying capacity of historical traffic predictive server next time period of cluster of each historical time section of server cluster of calculating according to statistic unit 302;
Amending unit 304, revised the traffic carrying capacity of next time period of predicting unit 303 predictions for utilizing date under next time period corresponding correction factor of classifying, and obtains the traffic carrying capacity of revised next time period;
Comparing unit 305, whether the historical traffic peak calculated for comparative statistics unit 302 is more than or equal to the traffic carrying capacity of revised next time period of amending unit 304 acquisitions;
As a kind of optional execution mode, the control appliance shown in Fig. 3 can also comprise:
Whether judging unit 308, exceed default tolerance range for the service quality of every business judging server cluster, if do not exceed, judges whether the traffic carrying capacity of next time period is greater than historical traffic peak;
Updating block 309, while for the traffic carrying capacity of judging next time period at judging unit 308, being greater than historical traffic peak, the historical traffic peak that statistic unit 302 is calculated is updated to the traffic carrying capacity of next time period.
In one embodiment, statistic unit 303 is specifically for utilizing the indices relevant to traffic carrying capacity of each historical time section of server cluster, and the historical traffic of each historical time section of binding hierarchy analytic approach difference calculation server cluster.
In one embodiment, predicting unit 303 specifically for the historical traffic using each historical time section of server cluster as output variable, and in conjunction with neural network prediction algorithm or the traffic carrying capacity of grey system forecasting algorithm predicts next time period of server cluster.
Refer to Fig. 4, Fig. 4 is the structural representation of the adaptive power conservation control appliance of the disclosed another kind of service based amount prediction of the embodiment of the present invention.Wherein, the adaptive power conservation control appliance of the service based amount shown in Fig. 4 prediction is that the adaptive power conservation control appliance of service based amount prediction as shown in Figure 3 is optimized and obtains.In the adaptive power conservation control appliance of the prediction of the service based amount shown in Fig. 4, adjustment unit 306 comprises:
The first computation subunit 3061, for at the comparative result of comparing unit 305 while being no, calculate the traffic carrying capacity of each server in each the class server in server cluster under current time, wherein, the maximum that the traffic carrying capacity of each server equals server allows the product of the traffic carrying capacity coefficient of regime corresponding with the operating state of server under current time, coefficient of regime corresponding to the operating state of server is greater than 0, and is less than or equal to 1; Wherein, when the operating state of server is different, the coefficient of regime difference that the operating state of server is corresponding;
The second computation subunit 3062, for the traffic carrying capacity of each server in each the class server according to server cluster under current time, calculate the traffic carrying capacity of each the class server in server cluster under current time;
The 3rd computation subunit 3063, for the traffic carrying capacity of each the class server according to server cluster under current time, calculate the traffic carrying capacity of server cluster under current time;
The 4th computation subunit 3064, for the traffic carrying capacity of calculating server cluster under current time and the difference of the traffic carrying capacity of revised next time period;
Adjust subelement 3065, for the operating state of the server of adjusting number of servers that server cluster moves and/or operation according to this difference.
In the embodiment of the present invention, the indices relevant to traffic carrying capacity comprises number of concurrent, number of users and page browsing amount.
In the embodiment of the present invention, control appliance shown in Fig. 3, Fig. 4 can be predicted the traffic carrying capacity of next time period according to historical traffic, and by the traffic carrying capacity of prediction, the operating state of the server of the number of servers moved in server cluster and/or operation is adjusted, thereby can carry out the efficient adaptive Energy Saving Control to server cluster, reach energy-conservation effect.
Refer to Fig. 5, Fig. 5 is the structural representation of the adaptive power conservation control appliance of the disclosed another kind of service based amount prediction of the embodiment of the present invention.As shown in Figure 5, the adaptive power conservation control appliance of this service based amount prediction can comprise:
Collect the indices relevant to traffic carrying capacity of each historical time section of server cluster;
Utilize the historical traffic of indices difference calculation server cluster each the historical time section relevant to traffic carrying capacity of each historical time section of server cluster;
Traffic carrying capacity according to historical traffic predictive server next time period of cluster of each historical time section of server cluster;
Utilize date under next time period corresponding correction factor of classifying to be revised the traffic carrying capacity of next time period, obtain the traffic carrying capacity of revised next time period;
Whether the historical traffic peak relatively calculated is more than or equal to the traffic carrying capacity of revised next time period, the operating state of the number of servers of moving in the traffic carrying capacity adjustment server cluster according to revised next time period if not, and/or the server of operation; The traffic carrying capacity difference of the server process of this operation when wherein, the operating state of the server of operation is different.
As a kind of optional execution mode, after the operating state of the number of servers of moving in the traffic carrying capacity adjustment server cluster of processor 501 according to revised next time period and/or the server of operation, also carry out following operation:
The service quality of every business in the monitoring server cluster;
Whether the service quality that judges every business in server cluster exceeds default tolerance range, if do not exceed, whether the traffic carrying capacity that judges next time period is greater than historical traffic peak, if be greater than, historical traffic peak is updated to the traffic carrying capacity of next time period.
As a kind of optional execution mode, the historical traffic that processor 501 utilizes the indices relevant to traffic carrying capacity of each historical time section of server cluster to calculate respectively each historical time section of described server cluster comprises:
Utilize the indices relevant to traffic carrying capacity of each historical time section of server cluster, and the historical traffic of each historical time section of binding hierarchy analytic approach difference calculation server cluster.
As a kind of optional execution mode, processor 501 comprises according to the traffic carrying capacity of historical traffic predictive server next time period of cluster of each historical time section of server cluster:
Using the historical traffic of each historical time section of server cluster as output variable, and in conjunction with neural network prediction algorithm or the traffic carrying capacity of grey system forecasting algorithm predicts next time period of server cluster.
As a kind of optional execution mode, the operating state of the number of servers of moving in the traffic carrying capacity adjustment server cluster of processor 501 according to revised next time period and/or the server of operation comprises:
Calculate the traffic carrying capacity of each server in each the class server in server cluster under current time, wherein, the maximum that the traffic carrying capacity of each server equals this server allows the product of the traffic carrying capacity coefficient of regime corresponding with the operating state of this server under current time, coefficient of regime corresponding to the operating state of this server is greater than 0, and is less than or equal to 1; Wherein, when the operating state of this server is different, the coefficient of regime difference that the operating state of this server is corresponding;
According to the traffic carrying capacity of each server in each the class server in server cluster under current time, calculate the traffic carrying capacity of each the class server in server cluster under current time;
According to the traffic carrying capacity of each the class server in server cluster under current time, calculate the traffic carrying capacity of server cluster under current time;
Calculate the traffic carrying capacity of server cluster under current time and the difference of the traffic carrying capacity of revised next time period;
Adjust the operating state of the server of the number of servers moved in server cluster and/or operation according to difference.
In the embodiment of the present invention, the indices relevant to traffic carrying capacity comprises number of concurrent, number of users and page browsing amount.
In the embodiment of the present invention, control appliance shown in Fig. 5 can be predicted the traffic carrying capacity of next time period according to historical traffic, and by the traffic carrying capacity of prediction, the operating state of the server of the number of servers moved in server cluster and/or operation is adjusted, thereby can carry out the efficient adaptive Energy Saving Control to server cluster, reach energy-conservation effect.
Refer to Fig. 6, Fig. 6 is the structural representation of the disclosed a kind of server cluster system of the embodiment of the present invention.As shown in Figure 6, this server cluster system comprises the adaptive power conservation control appliance 602 of server cluster 601 and the prediction of service based amount, wherein, the adaptive power conservation control appliance 602 of service based amount prediction is connected with each server communication in server cluster 601.In the embodiment of the present invention, the structure of the adaptive power conservation control appliance 602 of service based amount prediction, function are identical with structure, the function of the adaptive power conservation control appliance that the described service based amount of Fig. 3, Fig. 4 is predicted, the embodiment of the present invention is not repeated.In the embodiment of the present invention, the type of server in server cluster 601 can be for multiple, and the number of servers of each type is not limit.In the embodiment of the present invention, this server cluster system can be predicted the traffic carrying capacity of next time period according to historical traffic, and by the traffic carrying capacity of prediction, the operating state of the server of the number of servers moved in server cluster and/or operation is adjusted, thereby can carry out the efficient adaptive Energy Saving Control to server cluster, reach energy-conservation effect.
The self-adapting energy-saving control method of the disclosed service based amount prediction of the embodiment of the present invention and equipment, system can dynamically be adjusted the server in server cluster system by the mode of prediction traffic carrying capacity, thereby save the energy.
The self-adapting energy-saving control method of the disclosed service based amount prediction of the embodiment of the present invention and equipment, the applicable search of system, mailbox, door, network game, Internet video, memcache, web cache etc. have front end applications scene (these front ends are not stored critical data) or distributed system, and the larger value of server scale is larger.
One of ordinary skill in the art will appreciate that all or part of step in the whole bag of tricks of above-described embodiment is to come the hardware that instruction is relevant to complete by program, this program can be stored in a computer-readable recording medium, storage medium can comprise: flash disk, read-only memory (Read-Only Memory, ROM), random access device (Random Access Memory, RAM), disk or CD etc.
The above self-adapting energy-saving control method to the disclosed a kind of service based amount prediction of the embodiment of the present invention and equipment, system are described in detail, applied specific case herein principle of the present invention and execution mode are set forth, the explanation of above embodiment is just for helping to understand method of the present invention and core concept thereof; , for one of ordinary skill in the art, according to thought of the present invention, all will change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention simultaneously.
Claims (13)
1. the self-adapting energy-saving control method of a service based amount prediction, is characterized in that, comprising:
Collect the indices relevant to traffic carrying capacity of each historical time section of server cluster;
Utilize the indices relevant to traffic carrying capacity of each historical time section of described server cluster to calculate respectively the historical traffic of each historical time section of described server cluster;
Predict the traffic carrying capacity of described next time period of server cluster according to the historical traffic of each historical time section of described server cluster;
Utilize date under described next time period corresponding correction factor of classifying to be revised the traffic carrying capacity of described next time period, obtain the traffic carrying capacity of revised next time period;
Whether the historical traffic peak relatively calculated is more than or equal to the traffic carrying capacity of described revised next time period, if not, adjust the operating state of the server of the number of servers moved in described server cluster and/or operation according to the traffic carrying capacity of described revised next time period; The traffic carrying capacity difference of the server process of described operation when wherein, the operating state of the server of described operation is different.
2. method according to claim 1, is characterized in that, after the described traffic carrying capacity according to described revised next time period is adjusted the operating state of server of the number of servers moved in described server cluster and/or operation, described method also comprises:
Monitor the service quality of every business in described server cluster;
Whether the service quality that judges every business in described server cluster exceeds default tolerance range, if do not exceed, whether the traffic carrying capacity that judges next time period is greater than described historical traffic peak, if be greater than, described historical traffic peak is updated to the traffic carrying capacity of described next time period.
3. method according to claim 1, is characterized in that, the historical traffic that the described indices relevant to traffic carrying capacity that utilizes each historical time section of described server cluster calculates respectively each historical time section of described server cluster comprises:
Utilize the indices relevant to traffic carrying capacity of each historical time section of described server cluster, and the binding hierarchy analytic approach is calculated respectively the historical traffic of each historical time section of described server cluster.
4. according to the described method of claim 1～3 any one, it is characterized in that, the described historical traffic according to each historical time section of described server cluster predicts that the traffic carrying capacity of described next time period of server cluster comprises:
Using the historical traffic of each historical time section of described server cluster as output variable, and in conjunction with neural network prediction algorithm or the traffic carrying capacity of described next time period of server cluster of grey system forecasting algorithm predicts.
5. method according to claim 4, is characterized in that, the operating state that the described traffic carrying capacity according to described revised next time period is adjusted the server of the number of servers moved in described server cluster and/or operation comprises:
Calculate the traffic carrying capacity of each server in each the class server in described server cluster under current time, wherein, the maximum that the traffic carrying capacity of described each server equals described server allows the product of the traffic carrying capacity coefficient of regime corresponding with the operating state of described server under current time, coefficient of regime corresponding to the operating state of described server is greater than 0, and is less than or equal to 1; Wherein, when the operating state of described server is different, the coefficient of regime difference that the operating state of described server is corresponding;
According to the traffic carrying capacity of each server in each the class server in described server cluster under current time, calculate the traffic carrying capacity of each the class server in described server cluster under current time;
According to the traffic carrying capacity of each the class server in described server cluster under described current time, calculate the traffic carrying capacity of described server cluster under described current time;
Calculate the traffic carrying capacity of described server cluster under described current time and the difference of the traffic carrying capacity of described revised next time period;
Adjust the operating state of the server of the number of servers moved in described server cluster and/or operation according to described difference.
6. method according to claim 5, is characterized in that, the described indices relevant to traffic carrying capacity comprises number of concurrent, number of users and page browsing amount.
7. the adaptive power conservation control appliance of a service based amount prediction, is characterized in that, comprising:
Collector unit, for collecting the indices relevant to traffic carrying capacity of each historical time section of server cluster;
Statistic unit, calculate respectively the historical traffic of each historical time section of described server cluster for the indices relevant to traffic carrying capacity that utilizes each historical time section of described server cluster that described collector unit collects;
Predicting unit, predict the traffic carrying capacity of described next time period of server cluster for the historical traffic of each historical time section of described server cluster of calculating according to described statistic unit;
Amending unit, revised the traffic carrying capacity of described next time period of described predicting unit prediction for utilizing date under described next time period corresponding correction factor of classifying, and obtains the traffic carrying capacity of revised next time period;
Comparing unit, whether the historical traffic peak calculated for more described statistic unit is more than or equal to the traffic carrying capacity of described revised next time period of described amending unit acquisition;
Adjustment unit, for at the comparative result of described comparing unit while being no, the traffic carrying capacity of described revised next time period obtained according to described amending unit is adjusted the operating state of the server of the number of servers moved in described server cluster and/or operation; The traffic carrying capacity difference of the server process of described operation when wherein, the operating state of the server of described operation is different.
8. equipment according to claim 7, is characterized in that, described equipment also comprises:
Monitoring unit, after the operating state for the server of adjusting number of servers that described server cluster moves and/or operation at described adjustment unit, monitor the service quality of every business in described server cluster;
Whether judging unit, exceed default tolerance range for the service quality of every business judging described server cluster, if do not exceed, judges whether the traffic carrying capacity of next time period is greater than described historical traffic peak;
Updating block, while for the traffic carrying capacity that goes out described next time period in described judgment unit judges, being greater than described historical traffic peak, be updated to described historical traffic peak the traffic carrying capacity of described next time period.
9. equipment according to claim 7, is characterized in that,
Described statistic unit is specifically for utilizing the indices relevant to traffic carrying capacity of each historical time section of described server cluster, and the binding hierarchy analytic approach is calculated respectively the historical traffic of each historical time section of described server cluster.
10. according to the described equipment of claim 7～9 any one, it is characterized in that,
Described predicting unit, specifically for the historical traffic using each historical time section of described server cluster as output variable, and in conjunction with neural network prediction algorithm or the traffic carrying capacity of described next time period of server cluster of grey system forecasting algorithm predicts.
11. equipment according to claim 10, is characterized in that, described adjustment unit comprises:
The first computation subunit, for at the comparative result of described comparing unit while being no, calculate the traffic carrying capacity of each server in each the class server in described server cluster under current time, wherein, the maximum that the traffic carrying capacity of described each server equals described server allows the product of the traffic carrying capacity coefficient of regime corresponding with the operating state of described server under current time, coefficient of regime corresponding to the operating state of described server is greater than 0, and is less than or equal to 1; Wherein, when the operating state of described server is different, the coefficient of regime difference that the operating state of described server is corresponding;
The second computation subunit, for the traffic carrying capacity of each server in each the class server according to described server cluster under current time, calculate the traffic carrying capacity of each the class server in described server cluster under current time;
The 3rd computation subunit, for the traffic carrying capacity of each the class server according to described server cluster under described current time, calculate the traffic carrying capacity of described server cluster under described current time;
The 4th computation subunit, for the traffic carrying capacity of calculating described server cluster under described current time and the difference of the traffic carrying capacity of described revised next time period;
Adjust subelement, for the operating state of the server of adjusting number of servers that described server cluster moves and/or operation according to described difference.
12. equipment according to claim 11, is characterized in that, the described indices relevant to traffic carrying capacity comprises number of concurrent, number of users and page browsing amount.
A 13. server cluster system, it is characterized in that, described server cluster system comprises the adaptive power conservation control appliance of server cluster and the described service based amount prediction of claim 7～12 any one, wherein, the adaptive power conservation control appliance of described service based amount prediction is connected with each server communication in described server cluster.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310433553.1A CN103490956A (en) | 2013-09-22 | 2013-09-22 | Self-adaptive energy-saving control method, device and system based on traffic predication |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310433553.1A CN103490956A (en) | 2013-09-22 | 2013-09-22 | Self-adaptive energy-saving control method, device and system based on traffic predication |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103490956A true CN103490956A (en) | 2014-01-01 |
Family
ID=49830929
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310433553.1A Pending CN103490956A (en) | 2013-09-22 | 2013-09-22 | Self-adaptive energy-saving control method, device and system based on traffic predication |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103490956A (en) |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104158911A (en) * | 2014-09-01 | 2014-11-19 | 青岛海信电器股份有限公司 | Server cluster-oriented control method and system |
| CN104244436A (en) * | 2014-09-29 | 2014-12-24 | 中国联合网络通信集团有限公司 | Method and device for adjusting network equipment |
| CN105138691A (en) * | 2015-09-18 | 2015-12-09 | 北京百度网讯科技有限公司 | Method and system for analyzing user traffic |
| CN105227676A (en) * | 2015-10-20 | 2016-01-06 | 珠海许继芝电网自动化有限公司 | A kind of method of distribution terminal data acquisition |
| CN105426978A (en) * | 2014-09-04 | 2016-03-23 | 中兴通讯股份有限公司 | Service concurrency prediction method and service concurrency prediction system |
| CN105808347A (en) * | 2014-12-31 | 2016-07-27 | 北京神州泰岳软件股份有限公司 | Cluster server deployment calculation method and device |
| CN106126339A (en) * | 2016-06-21 | 2016-11-16 | 青岛海信传媒网络技术有限公司 | resource adjusting method and device |
| CN106549772A (en) * | 2015-09-16 | 2017-03-29 | 华为技术有限公司 | Resource prediction method, system and capacity management device |
| WO2017084385A1 (en) * | 2015-11-16 | 2017-05-26 | 广州广电运通金融电子股份有限公司 | Self-service equipment energy saving control method and device |
| CN106803815A (en) * | 2015-11-26 | 2017-06-06 | 阿里巴巴集团控股有限公司 | A kind of flow control methods and device |
| CN107046505A (en) * | 2016-02-05 | 2017-08-15 | 华为技术有限公司 | A service control method and a service control device |
| CN107093096A (en) * | 2016-12-15 | 2017-08-25 | 口碑控股有限公司 | A kind of Traffic prediction method and device |
| CN107092973A (en) * | 2016-11-25 | 2017-08-25 | 口碑控股有限公司 | The Forecasting Methodology and device of a kind of portfolio |
| CN107222540A (en) * | 2017-06-06 | 2017-09-29 | 国家计算机网络与信息安全管理中心 | A Group Scheduling Method for Server Clusters Based on Negative Feedback |
| WO2017167041A1 (en) * | 2016-04-01 | 2017-10-05 | 阿里巴巴集团控股有限公司 | Method and apparatus for correcting service data prediction |
| CN107301570A (en) * | 2017-07-13 | 2017-10-27 | 北京小度信息科技有限公司 | Traffic prediction method, abnormal traffic quantity measuring method, device and electronic equipment |
| CN107608788A (en) * | 2017-08-29 | 2018-01-19 | 北京三快在线科技有限公司 | A kind of control method, device and equipment |
| CN107743138A (en) * | 2017-07-24 | 2018-02-27 | 上海壹账通金融科技有限公司 | Push server distribution method, device, computer equipment and storage medium |
| WO2018119568A1 (en) * | 2016-12-26 | 2018-07-05 | Morgan Stanley Services Group Inc. | Predictive asset optimization for computer resources |
| WO2018191844A1 (en) * | 2017-04-17 | 2018-10-25 | 深圳市沃信科技有限公司 | Trade playback system and method |
| CN109413121A (en) * | 2017-08-16 | 2019-03-01 | 中兴通讯股份有限公司 | A kind of control method and device of server cluster |
| CN110109800A (en) * | 2019-04-10 | 2019-08-09 | 网宿科技股份有限公司 | A kind of management method and device of server cluster system |
| WO2019174410A1 (en) * | 2018-03-16 | 2019-09-19 | 阿里巴巴集团控股有限公司 | Service occurrence amount prediction method, apparatus and device |
| CN110928748A (en) * | 2019-12-04 | 2020-03-27 | 中国银行股份有限公司 | Business system operation monitoring method and device |
| CN111352639A (en) * | 2019-12-16 | 2020-06-30 | 深圳市智微智能软件开发有限公司 | After-sale system upgrading method and system |
| CN112350952A (en) * | 2020-10-28 | 2021-02-09 | 武汉绿色网络信息服务有限责任公司 | Controller distribution method and network service system |
| CN112512057A (en) * | 2019-09-16 | 2021-03-16 | 中国移动通信集团河北有限公司 | Network slice abnormity identification method, device, equipment and computer storage medium |
| CN112866129A (en) * | 2021-03-31 | 2021-05-28 | 中国工商银行股份有限公司 | Method, device and system for packing by automatically adjusting intervals according to aging marks |
| CN112887219A (en) * | 2021-03-31 | 2021-06-01 | 中国工商银行股份有限公司 | Message packet interval adjusting method and device |
| CN113703974A (en) * | 2021-08-27 | 2021-11-26 | 深圳前海微众银行股份有限公司 | Method and device for predicting server capacity |
| WO2025044750A1 (en) * | 2023-08-31 | 2025-03-06 | 三维通信股份有限公司 | Energy consumption control method, indoor distribution system, computer device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101695050A (en) * | 2009-10-19 | 2010-04-14 | 浪潮电子信息产业股份有限公司 | Dynamic load balancing method based on self-adapting prediction of network flow |
| US20100218005A1 (en) * | 2009-02-23 | 2010-08-26 | Microsoft Corporation | Energy-aware server management |
| CN101820384A (en) * | 2010-02-05 | 2010-09-01 | 浪潮(北京)电子信息产业有限公司 | Method and device for dynamically distributing cluster services |
-
2013
- 2013-09-22 CN CN201310433553.1A patent/CN103490956A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100218005A1 (en) * | 2009-02-23 | 2010-08-26 | Microsoft Corporation | Energy-aware server management |
| CN101695050A (en) * | 2009-10-19 | 2010-04-14 | 浪潮电子信息产业股份有限公司 | Dynamic load balancing method based on self-adapting prediction of network flow |
| CN101820384A (en) * | 2010-02-05 | 2010-09-01 | 浪潮(北京)电子信息产业有限公司 | Method and device for dynamically distributing cluster services |
Cited By (53)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104158911A (en) * | 2014-09-01 | 2014-11-19 | 青岛海信电器股份有限公司 | Server cluster-oriented control method and system |
| CN105426978A (en) * | 2014-09-04 | 2016-03-23 | 中兴通讯股份有限公司 | Service concurrency prediction method and service concurrency prediction system |
| CN104244436A (en) * | 2014-09-29 | 2014-12-24 | 中国联合网络通信集团有限公司 | Method and device for adjusting network equipment |
| CN104244436B (en) * | 2014-09-29 | 2018-09-18 | 中国联合网络通信集团有限公司 | A kind of method of adjustment and device of the network equipment |
| CN105808347B (en) * | 2014-12-31 | 2018-11-09 | 北京神州泰岳软件股份有限公司 | Cluster server disposes computational methods and device |
| CN105808347A (en) * | 2014-12-31 | 2016-07-27 | 北京神州泰岳软件股份有限公司 | Cluster server deployment calculation method and device |
| CN106549772B (en) * | 2015-09-16 | 2019-11-19 | 华为技术有限公司 | Resource forecasting method, system and capacity management device |
| CN106549772A (en) * | 2015-09-16 | 2017-03-29 | 华为技术有限公司 | Resource prediction method, system and capacity management device |
| CN105138691A (en) * | 2015-09-18 | 2015-12-09 | 北京百度网讯科技有限公司 | Method and system for analyzing user traffic |
| CN105138691B (en) * | 2015-09-18 | 2018-10-02 | 北京百度网讯科技有限公司 | Analyze the method and system of subscriber traffic |
| CN105227676A (en) * | 2015-10-20 | 2016-01-06 | 珠海许继芝电网自动化有限公司 | A kind of method of distribution terminal data acquisition |
| WO2017084385A1 (en) * | 2015-11-16 | 2017-05-26 | 广州广电运通金融电子股份有限公司 | Self-service equipment energy saving control method and device |
| US10394302B2 (en) | 2015-11-16 | 2019-08-27 | Grg Banking Equipment Co., Ltd. | Self-service equipment energy saving control method and device |
| CN106803815A (en) * | 2015-11-26 | 2017-06-06 | 阿里巴巴集团控股有限公司 | A kind of flow control methods and device |
| CN106803815B (en) * | 2015-11-26 | 2020-03-24 | 阿里巴巴集团控股有限公司 | Flow control method and device |
| CN107046505A (en) * | 2016-02-05 | 2017-08-15 | 华为技术有限公司 | A service control method and a service control device |
| CN107046505B (en) * | 2016-02-05 | 2020-01-03 | 华为技术有限公司 | Service control method and service control device |
| US10776176B2 (en) | 2016-02-05 | 2020-09-15 | Huawei Technologies Co., Ltd. | Service control method and service control apparatus |
| WO2017167041A1 (en) * | 2016-04-01 | 2017-10-05 | 阿里巴巴集团控股有限公司 | Method and apparatus for correcting service data prediction |
| CN106126339B (en) * | 2016-06-21 | 2019-07-30 | 青岛海信传媒网络技术有限公司 | Resource adjusting method and device |
| CN106126339A (en) * | 2016-06-21 | 2016-11-16 | 青岛海信传媒网络技术有限公司 | resource adjusting method and device |
| CN107092973A (en) * | 2016-11-25 | 2017-08-25 | 口碑控股有限公司 | The Forecasting Methodology and device of a kind of portfolio |
| US11443251B2 (en) | 2016-11-25 | 2022-09-13 | Koubei Holding Limited | Method and device for predicting traffic |
| CN108921333A (en) * | 2016-11-25 | 2018-11-30 | 口碑(上海)信息技术有限公司 | A kind of prediction technique and device of portfolio |
| CN108921333B (en) * | 2016-11-25 | 2021-04-27 | 口碑(上海)信息技术有限公司 | Traffic prediction method and device |
| CN107093096A (en) * | 2016-12-15 | 2017-08-25 | 口碑控股有限公司 | A kind of Traffic prediction method and device |
| CN107093096B (en) * | 2016-12-15 | 2022-03-25 | 口碑(上海)信息技术有限公司 | Traffic prediction method and device |
| US11113612B2 (en) | 2016-12-26 | 2021-09-07 | Morgan Stanley Services Group Inc. | Predictive asset optimization for computer resources |
| JP2020504382A (en) * | 2016-12-26 | 2020-02-06 | モルガン スタンレー サービシーズ グループ，インコーポレイテッドＭｏｒｇａｎ Ｓｔａｎｌｅｙ Ｓｅｒｖｉｃｅｓ Ｇｒｏｕｐ，Ｉｎｃ． | Predictive asset optimization for computer resources |
| US11481653B2 (en) | 2016-12-26 | 2022-10-25 | Morgan Stanley Services Group Inc. | Predictive asset optimization for computer resources |
| WO2018119568A1 (en) * | 2016-12-26 | 2018-07-05 | Morgan Stanley Services Group Inc. | Predictive asset optimization for computer resources |
| WO2018191844A1 (en) * | 2017-04-17 | 2018-10-25 | 深圳市沃信科技有限公司 | Trade playback system and method |
| CN107222540A (en) * | 2017-06-06 | 2017-09-29 | 国家计算机网络与信息安全管理中心 | A Group Scheduling Method for Server Clusters Based on Negative Feedback |
| CN107222540B (en) * | 2017-06-06 | 2020-11-20 | 国家计算机网络与信息安全管理中心 | A Negative Feedback-Based Group Scheduling Method for Server Clusters |
| CN107301570A (en) * | 2017-07-13 | 2017-10-27 | 北京小度信息科技有限公司 | Traffic prediction method, abnormal traffic quantity measuring method, device and electronic equipment |
| CN107301570B (en) * | 2017-07-13 | 2021-01-26 | 北京星选科技有限公司 | Traffic prediction method, abnormal traffic detection device and electronic equipment |
| CN107743138A (en) * | 2017-07-24 | 2018-02-27 | 上海壹账通金融科技有限公司 | Push server distribution method, device, computer equipment and storage medium |
| CN109413121A (en) * | 2017-08-16 | 2019-03-01 | 中兴通讯股份有限公司 | A kind of control method and device of server cluster |
| CN107608788A (en) * | 2017-08-29 | 2018-01-19 | 北京三快在线科技有限公司 | A kind of control method, device and equipment |
| CN107608788B (en) * | 2017-08-29 | 2020-08-07 | 北京三快在线科技有限公司 | Control method, device and equipment |
| WO2019174410A1 (en) * | 2018-03-16 | 2019-09-19 | 阿里巴巴集团控股有限公司 | Service occurrence amount prediction method, apparatus and device |
| CN110109800A (en) * | 2019-04-10 | 2019-08-09 | 网宿科技股份有限公司 | A kind of management method and device of server cluster system |
| CN112512057A (en) * | 2019-09-16 | 2021-03-16 | 中国移动通信集团河北有限公司 | Network slice abnormity identification method, device, equipment and computer storage medium |
| CN112512057B (en) * | 2019-09-16 | 2023-02-10 | 中国移动通信集团河北有限公司 | Network slicing abnormality identification method, device, equipment and computer storage medium |
| CN110928748B (en) * | 2019-12-04 | 2024-04-26 | 中国银行股份有限公司 | Service system operation monitoring method and device |
| CN110928748A (en) * | 2019-12-04 | 2020-03-27 | 中国银行股份有限公司 | Business system operation monitoring method and device |
| CN111352639A (en) * | 2019-12-16 | 2020-06-30 | 深圳市智微智能软件开发有限公司 | After-sale system upgrading method and system |
| CN112350952A (en) * | 2020-10-28 | 2021-02-09 | 武汉绿色网络信息服务有限责任公司 | Controller distribution method and network service system |
| CN112866129A (en) * | 2021-03-31 | 2021-05-28 | 中国工商银行股份有限公司 | Method, device and system for packing by automatically adjusting intervals according to aging marks |
| CN112887219A (en) * | 2021-03-31 | 2021-06-01 | 中国工商银行股份有限公司 | Message packet interval adjusting method and device |
| CN113703974A (en) * | 2021-08-27 | 2021-11-26 | 深圳前海微众银行股份有限公司 | Method and device for predicting server capacity |
| WO2023024679A1 (en) * | 2021-08-27 | 2023-03-02 | 深圳前海微众银行股份有限公司 | Method and apparatus for predicting server capacity |
| WO2025044750A1 (en) * | 2023-08-31 | 2025-03-06 | 三维通信股份有限公司 | Energy consumption control method, indoor distribution system, computer device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103490956A (en) | Self-adaptive energy-saving control method, device and system based on traffic predication | |
| US11409347B2 (en) | Method, system and storage medium for predicting power load probability density based on deep learning | |
| CN109300310B (en) | Traffic flow prediction method and device | |
| JP6756048B2 (en) | Predictive asset optimization for computer resources | |
| CN114066071B (en) | Power parameter optimization method based on energy consumption, terminal equipment and storage medium | |
| CN105989420B (en) | Method for determining electricity utilization behavior characteristics of user, and method and device for predicting electricity utilization load of user | |
| CN110289994B (en) | Cluster capacity adjusting method and device | |
| CN104715292A (en) | City short-term water consumption prediction method based on least square support vector machine model | |
| CN105160444A (en) | Electrical equipment failure rate determining method and system | |
| CN113039506A (en) | Data center infrastructure optimization method based on causal learning | |
| CN107046505B (en) | Service control method and service control device | |
| CN114500578B (en) | Distributed storage system load balancing scheduling method, device and storage medium | |
| CN104679590A (en) | Map optimization method and device in distributive calculating system | |
| CN109242150A (en) | A kind of electric network reliability prediction technique | |
| CN110647381A (en) | Virtual machine resource balancing and deployment optimizing method | |
| Phung et al. | A prediction based autoscaling in serverless computing | |
| CN115238599B (en) | Energy-saving method and model reinforcement learning training method and device for refrigerating system | |
| CN115876004A (en) | Water-equalizing control method based on cooling tower, cooling system, equipment and storage medium | |
| CN108471362B (en) | Resource allocation prediction technique and device | |
| CN117477581B (en) | Power system load balancing control method and power system | |
| CN111047465B (en) | Power grid friendly load response method based on power big data | |
| CN111191815B (en) | Ultra-short-term output prediction method and system for wind power cluster | |
| CN110309873A (en) | Integrated Transient Stability Evaluation Method and System Based on Bayesian Classification of Power Flow Vector | |
| CN103873388A (en) | Network content control method and network equipment | |
| KR102507185B1 (en) | Power system, method and apparatus for clustering of power system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140101 |