CN101316280A - Gridding resource intelligent monitoring method based on feedback - Google Patents
Gridding resource intelligent monitoring method based on feedback Download PDFInfo
- Publication number
- CN101316280A CN101316280A CNA2008101241353A CN200810124135A CN101316280A CN 101316280 A CN101316280 A CN 101316280A CN A2008101241353 A CNA2008101241353 A CN A2008101241353A CN 200810124135 A CN200810124135 A CN 200810124135A CN 101316280 A CN101316280 A CN 101316280A
- Authority
- CN
- China
- Prior art keywords
- monitoring
- value
- poll
- resource
- cycle
- 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
Images
Abstract
The invention discloses an intelligent monitoring method based on network resources fed back, which utilizes a vague mathematics method to analyze data fed back, sets a new polling period according to processing results and leads resource real-time and load to be well balanced on the basis of adopting fixed period polling monitoring. Therefore, the problems of the decrease of the performance of a system or poor monitoring real-time caused by fixed period monitoring in a grid monitoring system are solved, performance data stored in a directory service is more accurate, and resource consumption and the updating operation times of resource monitoring are reduced.
Description
Technical field
The present invention relates generally to a kind of gridding resource method for supervising, is used for solving fixedly polling cycle deficiency of gridding resource monitoring, and the problem that the renewal number of operations of monitoring resource is many belongs to the grid computing technology field.
Background technology
Grid computing has obtained the extensive concern of global range as an important frontier.With the different autonomous territories of leaps numerous in the grid, the computer of the isomery of Fen Buing and resource organizations get up geographically, are the emphasis of studying both at home and abroad.The grid that arises at the historic moment is used the notion of electric power system, utilizes grid, and people can connect switch as electricity consumption, just can use resource easily.Grid is the seamless integrated and collaborative computing environment in the wide scope, and all kinds of resources are integrated application in grid.Resource in the grid comprises computational resource, storage resources, the communication resource, software resource, information resources, knowledge resource etc.And in grid, method for supervising is an important component part.Method for supervising can help the Resource Manager analyzing system performance, in time pinpoints the problems the reasonable disposition resource; For other service in the grid provides the information that needs, help the user to select only resource.
Because the dynamic of gridding resource, promptly resource can dynamically add or withdraw from, and will consider dynamic so gridding resource monitored also, and monitoring must be real-time, because the information of node is dynamic change; Yet monitoring can increase system burden continually, and system effectiveness also can reduce.For example:, will increase unnecessary monitoring burden if we monitor a resource status that does not have bigger variation in a short time.Some are organized and have developed the monitoring grid system at present, such as NWS, and MDS etc.Information collection tool sensor (transducer) has been developed in they or oneself, perhaps utilizes existing monitoring resource instrument Ganglia (sweet lattice Leah) or Hawkeye (good gram is inferior).These supervisory control systems generally all are divided into grid some virtual region VO (Virtual Organization) that concentrate in the geographical position, a global dictionary server is set among each virtual region VO, the static informations such as physical address of main memory node, local directory server of configuration in each node, in each main frame of node, transducer is set for performance data, the transducer of these performance datas is according to fixed cycle poll collecting performance data, in each node, also be provided with the sensor management device, management of sensor and the data that collect are deposited in the local directory service.They have played important function, but some shortcomings are all arranged, and these monitoring resource systems adopt the performance data of the poll strategy collection resource current performance of fixed cycle.The problem of being narrated before will existing like this.Therefore it is significant to study a kind of novel gridding resource supervisory control system.
Summary of the invention
Technical problem: the purpose of this invention is to provide a kind of gridding resource intelligent monitoring method based on feedback, fixedly the polling cycle control real-time is not strong in the solution prior art, the problem that the renewal number of operations of monitoring resource is many.The method that adopts the present invention to propose can solve fixedly, and the caused systematic function of polling cycle descends or the not strong problem of control real-time.
Technical scheme: method of the present invention is a kind of method for supervising of the property improved, propose based on the intelligent control method that feeds back by introducing, its principle is to utilize fuzzy mathematics method that the data that feedback obtains are analyzed, determine poll time of following one-period with analysis result, solved in the method for monitoring grid because problems such as the caused overheads of fixed cycle property monitoring.
One, architecture
Fig. 1 has provided the design architecture of the supervisory control system of a this method of utilization, and its functional part mainly comprises informant (Information Provider), adapter (Adapter), intelligent controller, index server.
Below we provide specifying of several sections:
The informant: the information gathering source of on monitor node, moving, collect various types of resource state informations.Comprise Globus distribution GRAM (grid resource allocation manager) and can with integrated external information supplier such as Ganglia (sweet lattice Leah) or the Hawkeye (good gram Asia) of MDS.The CPU and the memory information of the relevant gridding resource of GRAM (grid resource allocation manager) issue, and with job queue and the relevant schedule information of operation submitted to, the information that is obtained is less, so when the monitoring network case system, usually be used in combination with external information supplier such as Ganglia, it can obtain more host information such as host name, processor, internal memory, operating system and file system.
Adapter: exist difference between the different information acquisition devices.The effect of adapter is exactly to eliminate these difference, obtains data from dissimilar information acquisition devices, gives the intelligent controller parts then and handles.Adapter makes good flexibility when selecting the informant.
Intelligent controller: on the basis of polling cycle, utilize fuzzy mathematics to compare the information data that the information that feeds back and previous moment inquire, make the setting that different reflections is a different cycles according to different results again.
Index server: mainly be to be used to deposit collected system information, the informant reports to GRAM with the system information of obtaining, and GRAM is aggregated into the information that obtains in the index service then, uses for client-requested.
Two, method flow
1, gridding resource method for supervising flow process
The information of monitoring generally includes static information, for example CPU quantity, clock speed, physical memory total amount, virtual memory and free disk space and multidate information comprise the medium pending number of jobs of number, free memory, formation of available CPU, the utilance of current resource etc.The information that present existing method for supervising generally all adopts fixed cycle poll strategy to come necessary for monitoring is representative with Ganglia, and its monitoring flow process as shown in Figure 2.
Ganglia is a distributed surveillance, it has two Daemon (background program), is respectively: client Ganglia Monitoring Daemon (gmond) (background monitoring process) and service end GangliaMeta Daemon (gmetad) (background monitoring process).Background monitoring process (gmond) uses multicast protocol to subscribe to the state of collecting each node..Member node receives that this node of information representation of a certain node is available, represents that then this node is unavailable if all confiscate echo message in several cycles.Set earlier fixedly polling cycle, when the update cycle arrived, the resource of its this locality of monitoring nodes also sent monitor data by multicast protocol.Background monitoring process (gmetad) uses the tree type of point-to-point to connect the state that compiles all clusters between cluster.Because each node in the cluster comprises the whole monitor data of this cluster,, for the consideration of fault-tolerant aspect, can specify a plurality of actual nodes certainly to each leaf node so each leaf node in the tree is represented a different cluster in logic; Non-leaf node is the Rendezvous Point of information, represents the set of some clusters, and they are periodically collected is the information of its child node of poll.
The groundwork flow process of gridding resource method for supervising:
Step1: configuring external informant such as Ganglia, the integrated use of watch-dog that they and grid are carried;
Step2: fixedly polling cycle is set, is made as T;
Step3: data acquisition person is according to fixed cycle poll T collecting performance data, as information such as host name, processor, internal memory, file system;
Step4: the information of utilizing subscribing mechanism or the required monitoring of other machine-processed booking reader;
Step5: the performance data that will utilize the informant to collect regularly is aggregated in the directory service, uses for the user.
2, based on the gridding resource intelligent monitoring method flow process of feeding back
In grid environment, supervisory control system is constantly sent request to resource state information, and when new information produces, the state in the supervisory control system will upgrade.If upgrade too slow, can cause the performance data in the LIST SERVER expired, and regular poll causes the too fast or slow excessively problem of upgrading easily, need to formulate effective poll strategy, when the performance data amplitude of variation is violent, shorten polling cycle, when the performance data amplitude of variation is mild, increase polling cycle.So needing a kind of new scheme of design comes in real time to change polling cycle according to the variation of grid load.In Distributed Calculation, the change of resource state information depends on the change of CPU to a great extent, and fraction depends on the variation of internal memory etc., and same also is like this in grid environment.Therefore, if we establish a weights according to CPU and internal memory to the influence of grid environment, the utilance of observing them just can be predicted the change of other resource substantially.For example: the utilance of supposition CPU and internal memory does not almost change, and other resource status does not have big change yet; If their utilance has very big variation, other resource status also has a lot of variations.In other words, the change of monitoring grid incident depends on the change of CPU and internal memory.Therefore, when their utilance had bigger change, other resource should be monitored immediately, and in this case, we should adjust the interval of monitoring, allowed the monitoring resource service obtain resource state information after the renewal.Gridding resource intelligent monitoring method based on feedback mainly is a method of having used fuzzy mathematics.
This method is mainly used in the grid environment of being made up of informant, adapter, intelligent controller and index server, and the key step of this method is:
Step 3. data acquisition person is according to polling cycle T collecting performance data;
The performance data that step 4. collects step 3 is given adapter and is handled, and the data of different-format are done standard conversion;
Step 5. converges framework and utilizes subscribing mechanism to subscribe to the information of required monitoring;
Step 6. regularly is aggregated into the performance data that collects in the index server, uses for the user, and the data that this obtains constantly are kept in the intelligent controller, is designated as x[1];
2 array x[n of definition in intelligent controller] and t[n], be used for storing a last n observed value and poll time respectively, x[i] expression poll time t[i] observed value of correspondence, x[i] and t[i] determine a poll point;
Step 7. is determined object, and establishing the nearest poll time is object t[1], the poll time before the one-period is object t[2], the rest may be inferred up to object t[n]; Object t[i] corresponding attribute is observed value x[i]={ x
1i, x
2i..., x
Mi; X wherein
1i, x
2i..., x
MiRepresent the information that to monitor respectively; In intelligent controller, the observed value in n-1 the cycle of preserving before taking out, x[2], x[3] ... x[n];
Step 8. is according to n observed value x[1], x[2] ... x[n] set up fuzzy resembling relation; The amplitude of variation of n poll point can be used number r recently
Ij∈ [0,1] describes, according to the minimum method of arithmetic average
Determine r
IjValue, set up fuzzy similarity matrix R=(r
Ij)
N*n, r wherein
Ij=r
Ji, r
Ii=1, i wherein, j=1,2 ..., n; N is the number of observed value, and m is the pairing monitor message number of each observed value, k ∈ [1, m];
Step 9. is in fuzzy similarity matrix R, because this matrix is symmetrical, the lower left quarter of leading diagonal is identical with upper right quarter, and we only need see that lower left quarter gets final product;
If observed r
12Value is at [a
n, 1], promptly this moment remains unchanged substantially with the information that previous moment is monitored, and that is to say that systematic comparison is stable, and the value of the r before at this moment observing again in several cycles if value is very big always, approaches 1, promptly at [a
n, 1] scope in, just mean that also systematic function is all very stable in a very long time, just can enlarge polling cycle, new certainly polling cycle must be less than Tmax, otherwise gets Tmax, turns to step 3 to carry out again, always circulation; If the value of the r in preceding several cycles changes greatly, prove that just this cycle data variation is more steady, the just tentative cycle need not upgrade, and operates to reduce to upgrade, and can keep original polling cycle, turns to step 3 to carry out again, always circulation;
If r
12Value in other scope as [0, a
1] [a
1, a
2], [a
2, a
3] ... [a
N-1, a
n], then get different new polling cycles; As suppose r
12Value belong to [0, a
1], the similitude that then means these two objects is less, promptly has very big-difference, also just think recently during this period of time in performance change very big, need the immediate updating data to preserve the real-time of monitoring, at this moment, just dwindle polling cycle, if polling cycle＜Tmin, then get Tmin, monitoring immediately turns to step 3 to carry out, always circulation again.
Beneficial effect: the present invention is a kind of novel gridding resource supervisory control system, be mainly used in the gridding resource efficiency for monitoring problem that solves, the system that the application of the invention proposes can avoid the deficiency of employing fixed cycle poll monitoring originally, can make the data of monitoring more accurate, and reduce resource consumption.Improved the flexibility of monitoring resource.Provide specific description below:
The monitoring of real-time, efficent use of resources.The supervisory control system in past is owing to be that the fixed cycle poll exists the cycle to fix, lack the deficiency of flexibility.If poll frequency is too high, the most of the time of system and resource consumption have increased the burden of system greatly among inquiry, and systematic function will descend; If poll efficient is too low, systematically real time status just can not in time be reflected.And, gridding resource amplitude of variation instability, as vary within wide limits by day, the fixing cycle has been reduced the accuracy of performance data.On the contrary, night amplitude of variation mild, the fixing cycle has increased the node burden.In order to address this problem, we set one earlier greater than poll time of monitoring period at ordinary times, when n polling cycle arrives, current performance data and last several cycle are preserved the data of getting off does relatively, how calculate this data variation amplitude several times by fuzzy mathematics, also promptly interior during this period of time resource amplitude of variation how.If alter a great deal, then adjust (shortening) polling cycle immediately, the size in cycle is different according to the amplitude difference that changes; If variation is very little or approaching constant, then can consider to increase polling cycle again or remain unchanged.This algorithm has increased the adaptivity of monitoring resource, and monitoring period is changed along with the variation of grid environment, has reached the real-time monitoring that grid environment is changed, and has effectively utilized resource.
Description of drawings
Fig. 1 is this system architecture figure.Comprise among the figure: informant, adapter, intelligent controller, index service.
Fig. 2 is the Ganglia system assumption diagram.
Fig. 3 is a method flow diagram of using the intelligent controller of method of the present invention.
Embodiment
For convenience of description, our supposition has following example:
Step 3. data acquisition person is according to polling cycle T collecting performance data;
The performance data that step 4. collects step 3 is given adapter and is handled, and the data of different-format are done standard conversion;
Step 5.Aggregator (converging) framework utilizes Subscription/Notification (subscribing) mechanism to subscribe to the information of required monitoring;
Step 6. regularly is aggregated into the performance data that collects in the index server, uses for the user, and the data that this obtains constantly are kept in the intelligent controller, is designated as x[1];
2 array x[n of definition in intelligent controller] and t[n], be used for storing a last n observed value and poll time respectively, x[i] expression poll time t[i] observed value of correspondence, x[i] and t[i] determine a poll point.
Step 7. is determined object.If the nearest poll time is object t[1], the poll time before the one-period is object t[2], the rest may be inferred up to object t[n].Object t[i] corresponding attribute is observed value x[i]={ x
1i, x
2i..., x
Mi.X wherein
1i, x
2i..., x
MiRepresent the information that will monitor respectively, as CPU, internal memory etc., corresponding can establish corresponding weights, and for example, the variation of CPU has the greatest impact to the monitoring of whole grid, its weights can be established greatly accordingly.In intelligent controller, the observed value in n-1 the cycle of preserving before taking out, x[2], x[3] ... x[n];
Step 8. is according to n observed value x[1], x[2] ... x[n] set up fuzzy resembling relation.The amplitude of variation of n poll point can be used number r recently
Ij∈ [0,1] describes, according to the minimum method of arithmetic average
Determine r
IjValue, set up fuzzy similarity matrix R=(r
Ij)
N*n, r wherein
Ij=r
Ji, r
Ii=1, i wherein, j=1,2..., n; N is the number of observed value, and m is the pairing monitor message number of each observed value.
Step 9. is in order more effectively accurately to utilize resource, and we are with r
IjDivide the corresponding different cyclomorphosis in different territories, set up r according to user's request
IjWith the corresponding form of polling cycle, here, in order to narrate clearlyer, we suppose that form is as follows, and the user can create different forms according to the needs of oneself certainly:
| r ijScope | r ij∈[0，a 1] | r ij∈[a 1，a 2] | ...... | r ij∈[a n-1，a n] | r ij∈[a n，1] |
| The value of period T | T/ |
2T/n | ...... | (n-1)T/n | T |
Step 10. is in fuzzy similarity matrix R, because this matrix is symmetrical, the lower left quarter of leading diagonal is identical with upper right quarter, and we only need see that lower left quarter gets final product when research.
If observed r
12Value is at [a
n, 1], promptly this moment remains unchanged substantially with the information that previous moment is monitored, and that is to say that systematic comparison is stable.At this moment the value of the r before observing again in several cycles if value is very big always, approaches 1, promptly at [a
n, 1] scope in, just mean that also systematic function is all very stable in a very long time, we just can enlarge polling cycle, new certainly polling cycle must be less than Tmax, otherwise gets Tmax, turns to step 3 to carry out again, always circulation; If the value of the r in preceding several cycles changes greatly, prove that just this cycle data variation is more steady, the just tentative cycle need not upgrade, and operates to reduce to upgrade, and can keep original polling cycle, turns to step 3 to carry out again, always circulation.
If r
12Value in other scope as [0, a
1] [a
1, a
2], [a
2, a
3] ... [a
N-1, a
n], then contrast form and get different new polling cycles respectively.As suppose r
12Value belong to [0, a
1], mean that then the similitude of these two objects is less, promptly there is very big-difference, also just think recently during this period of time in performance change very big, need the immediate updating data to preserve the real-time of monitoring, at this moment, just dwindle polling cycle, if polling cycle＜Tmin, then get Tmin, monitoring immediately, otherwise determine the value of ensuing polling cycle according to the form of setting up in the step 9, turn to step 3 to carry out again, always circulation.
Then its embodiment is:
(1) configuring external informant such as Ganglia or Hawkeye in grid environment, and the watch-dog that they and grid are carried is integrated, and the environment configurations of necessary for monitoring is finished;
(2) initial poll cycle of monitoring resource is set, is made as T, establish maximum polling cycle Tmax and minimum-poll period T min again;
(3) informant is according to polling cycle T collecting performance data, as information such as host name, processor, internal memory, file system;
(4) give adapter with the data of collecting from dissimilar informants in (3) step and handle, the data of different-format are done standard conversion;
(5) utilize notice/subscription or other mechanism to subscribe to the information of required monitoring;
(6) performance data that collects regularly is aggregated in the index server, uses for the user.And the data that this obtains constantly are kept in the intelligent monitor, be designated as x[1];
(7) in the intelligent monitor parts, the observed value in n-1 the cycle of preserving before taking out, x[2], x[3] ... x[n].
(8) according to n observed value x[1], x[2] ... x[n] set up fuzzy resembling relation.The amplitude of variation of n poll point can be used number r recently
Ij∈ [0,1] describes, according to the minimum method of arithmetic average
Determine r
IjValue, set up fuzzy similarity matrix R=(r
Ij)
N*n, r wherein
Ij=r
Ji, r
Ii=1, i wherein, j=1,2..., n; N is the number of observed value, and m is the pairing monitor message number of each observed value.
(9) set up r according to user's request
IjThe corresponding form of value and polling cycle;
(10) if r
12Value smaller, we just need dwindle polling cycle (if polling cycle＜Tmin, then get Tmin), determine the value of ensuing polling cycle to turn to (3) to carry out, always circulation again according to the form of setting up in (9) step.
(11) if r
12Value very big, the value of the r before then observing again in several cycles is if value is very big always, at [a
n, 1] and in the scope, just mean that also systematic function is all very stable in a very long time, we just can enlarge polling cycle (if polling cycle＞Tmax then gets Tmax), turn to (3) to carry out again, always circulation; Not quite then keep the original cycle if value afterwards changes, turn to (3) to carry out again, always circulation.
Claims (1)
1, a kind of gridding resource intelligent monitoring method based on feedback is characterized in that steps of the method are in the grid environment that this method is applied to be made up of informant, adapter, intelligent controller and index server:
Step 1. informant has internal information supplier and external information supplier, and the intelligent controller that external information supplier and grid are carried is integrated, and the environment configurations of necessary for monitoring is finished;
Step 2. is provided with the initial poll period T of monitoring resource, establishes maximum polling cycle Tmax and minimum-poll period T min again;
Step 3. data acquisition person is according to polling cycle T collecting performance data;
The performance data that step 4. collects step 3 is given adapter and is handled, and the data of different-format are done standard conversion;
Step 5. converges framework and utilizes subscribing mechanism to subscribe to the information of required monitoring;
Step 6. regularly is aggregated into the performance data that collects in the index server, uses for the user, and the data that this obtains constantly are kept in the intelligent controller, is designated as x[1];
2 array x[n of definition in intelligent controller] and t[n], be used for storing a last n observed value and poll time respectively, x[i] expression poll time t[i] observed value of correspondence, x[i] and t[i] determine a poll point;
Step 7. is determined object, and establishing the nearest poll time is object t[1], the poll time before the one-period is object t[2], the rest may be inferred up to object t[n]; Object t[i] corresponding attribute is observed value x[i]={ x
1i, x
2i..., x
Mi; X wherein
1i, x
2i..., x
MiRepresent the information that to monitor respectively; In intelligent controller, the observed value in n-1 the cycle of preserving before taking out, x[2], x[3] ... x[n];
Step 8. is according to n observed value x[1], x[2] ... x[n] set up fuzzy resembling relation; The amplitude of variation of n poll point can be used number r recently
Ij∈ [0,1] describes, according to the minimum method of arithmetic average
Determine r
IjValue, set up fuzzy similarity matrix R=(r
Ij)
N*n, r wherein
Ij=r
Ji, r
Ii=1, i wherein, j=1,2 ..., n; N is the number of observed value, and m is the pairing monitor message number of each observed value, k ∈ [1, m];
Step 9. is in fuzzy similarity matrix R, because this matrix is symmetrical, the lower left quarter of leading diagonal is identical with upper right quarter, and we only need see that lower left quarter gets final product;
If observed r
12Value is at [a
n, 1], promptly this moment remains unchanged substantially with the information that previous moment is monitored, and that is to say that systematic comparison is stable, and the value of the r before at this moment observing again in several cycles if value is very big always, approaches 1, promptly at [a
n, 1] scope in, just mean that also systematic function is all very stable in a very long time, just can enlarge polling cycle, new certainly polling cycle must be less than Tmax, otherwise gets Tmax, turns to step 3 to carry out again, always circulation; If the value of the r in preceding several cycles changes greatly, prove that just this cycle data variation is more steady, the just tentative cycle need not upgrade, and operates to reduce to upgrade, and can keep original polling cycle, turns to step 3 to carry out again, always circulation;
If r
12Value in other scope as [0, a
1] [a
1, a
2], [a
2, a
3] ... [a
N-1, a
n], then get different new polling cycles; As suppose r
12Value belong to [0, a
1], the similitude that then means these two objects is less, promptly has very big-difference, also just think recently during this period of time in performance change very big, need the immediate updating data to preserve the real-time of monitoring, at this moment, just dwindle polling cycle, if polling cycle＜Tmin, then get Tmin, monitoring immediately turns to step 3 to carry out, always circulation again.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008101241353A CN101316280A (en) | 2008-06-13 | 2008-06-13 | Gridding resource intelligent monitoring method based on feedback |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008101241353A CN101316280A (en) | 2008-06-13 | 2008-06-13 | Gridding resource intelligent monitoring method based on feedback |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101316280A true CN101316280A (en) | 2008-12-03 |
Family
ID=40107117
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008101241353A Pending CN101316280A (en) | 2008-06-13 | 2008-06-13 | Gridding resource intelligent monitoring method based on feedback |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101316280A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011088697A1 (en) * | 2010-01-22 | 2011-07-28 | 中兴通讯股份有限公司 | Method and system for acquiring mobility related information of equipment and optimizing configuration |
| CN102541721A (en) * | 2010-09-17 | 2012-07-04 | 微软公司 | Integrating external and cluster heat map data |
| CN102761570A (en) * | 2011-04-28 | 2012-10-31 | 同济大学 | System and method for monitoring grid resources based on agents |
| CN102810073A (en) * | 2012-06-13 | 2012-12-05 | 天津大学 | Background real-time monitoring processing method for data |
| WO2014040260A1 (en) * | 2012-09-13 | 2014-03-20 | Accenture Global Services Limited | Method, system and apparatus for load shaving of a power grid and a tangible computer readable medium |
| CN105472009A (en) * | 2015-12-18 | 2016-04-06 | 国云科技股份有限公司 | Self-adapting frequency monitoring method of cloud platform resource |
| CN105577645A (en) * | 2015-12-11 | 2016-05-11 | 中国科学院声学研究所 | Agent-based HLS client-end device and realization method thereof |
| CN106789445A (en) * | 2017-02-06 | 2017-05-31 | 凌云天博光电科技股份有限公司 | The state polling method and system of the network equipment in a kind of CHINA RFTCOM Co Ltd |
| CN108845865A (en) * | 2018-06-28 | 2018-11-20 | 郑州云海信息技术有限公司 | A kind of monitoring service dispositions method, system and storage medium |
| CN109946420A (en) * | 2017-12-21 | 2019-06-28 | 智能城市网络有限公司 | For sending the device and method of atmospheric environment data |
| CN111193904A (en) * | 2019-12-17 | 2020-05-22 | 视联动力信息技术股份有限公司 | Inspection control method and device and readable storage medium |
-
2008
- 2008-06-13 CN CNA2008101241353A patent/CN101316280A/en active Pending
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011088697A1 (en) * | 2010-01-22 | 2011-07-28 | 中兴通讯股份有限公司 | Method and system for acquiring mobility related information of equipment and optimizing configuration |
| CN102541721B (en) * | 2010-09-17 | 2014-09-17 | 微软公司 | Integrating external and cluster heat map data |
| CN102541721A (en) * | 2010-09-17 | 2012-07-04 | 微软公司 | Integrating external and cluster heat map data |
| CN102761570A (en) * | 2011-04-28 | 2012-10-31 | 同济大学 | System and method for monitoring grid resources based on agents |
| CN102810073B (en) * | 2012-06-13 | 2014-11-12 | 天津大学 | Background real-time monitoring processing method for data |
| CN102810073A (en) * | 2012-06-13 | 2012-12-05 | 天津大学 | Background real-time monitoring processing method for data |
| WO2014040260A1 (en) * | 2012-09-13 | 2014-03-20 | Accenture Global Services Limited | Method, system and apparatus for load shaving of a power grid and a tangible computer readable medium |
| US9424613B2 (en) | 2012-09-13 | 2016-08-23 | Accenture Global Services Limited | Method, system and apparatus for load shaving of a power grid and a tangible computer readable medium |
| CN105577645A (en) * | 2015-12-11 | 2016-05-11 | 中国科学院声学研究所 | Agent-based HLS client-end device and realization method thereof |
| CN105472009A (en) * | 2015-12-18 | 2016-04-06 | 国云科技股份有限公司 | Self-adapting frequency monitoring method of cloud platform resource |
| CN106789445A (en) * | 2017-02-06 | 2017-05-31 | 凌云天博光电科技股份有限公司 | The state polling method and system of the network equipment in a kind of CHINA RFTCOM Co Ltd |
| CN109946420A (en) * | 2017-12-21 | 2019-06-28 | 智能城市网络有限公司 | For sending the device and method of atmospheric environment data |
| CN108845865A (en) * | 2018-06-28 | 2018-11-20 | 郑州云海信息技术有限公司 | A kind of monitoring service dispositions method, system and storage medium |
| CN111193904A (en) * | 2019-12-17 | 2020-05-22 | 视联动力信息技术股份有限公司 | Inspection control method and device and readable storage medium |
| CN111193904B (en) * | 2019-12-17 | 2022-12-13 | 视联动力信息技术股份有限公司 | Inspection control method and device and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101316280A (en) | Gridding resource intelligent monitoring method based on feedback | |
| Rao et al. | Distributed coordination of internet data centers under multiregional electricity markets | |
| CN104915407A (en) | Resource scheduling method under Hadoop-based multi-job environment | |
| Cheng et al. | Heterogeneity-aware workload placement and migration in distributed sustainable datacenters | |
| CN104113585A (en) | Hardware Level Generated Interrupts Indicating Load Balancing Status For A Node In A Virtualized Computing Environment | |
| US20110282982A1 (en) | Dynamic application placement based on cost and availability of energy in datacenters | |
| Jin et al. | Energy-efficient task scheduling for CPU-intensive streaming jobs on Hadoop | |
| KR20150030332A (en) | Distributed and parallel processing system on data and method of operating the same | |
| Xu et al. | Resource pre-allocation algorithms for low-energy task scheduling of cloud computing | |
| Arkian et al. | Model-based stream processing auto-scaling in geo-distributed environments | |
| Cheng et al. | Heterogeneity aware workload management in distributed sustainable datacenters | |
| Bergelt et al. | Energy efficient handling of big data in embedded, wireless sensor networks | |
| Banerjee et al. | Efficient resource utilization using multi-step-ahead workload prediction technique in cloud | |
| Shi et al. | MapReduce short jobs optimization based on resource reuse | |
| Diouani et al. | Green cloud computing: efficient energy-aware and dynamic resources management in data centers | |
| Li et al. | Study on the data processing of the IOT sensor network based on hadoop cloud platform and TWLGA scheduling algorithm | |
| Pham et al. | CIRUS: an elastic cloud-based framework for Ubilytics | |
| Roy et al. | A proposal for optimization of data node by horizontal scaling of name node using big data tools | |
| Zhou et al. | AHPA: adaptive horizontal pod autoscaling systems on alibaba cloud container service for kubernetes | |
| Zhang et al. | Monitoring-based task scheduling in large-scale SaaS cloud | |
| Petri et al. | Ensemble-based network edge processing | |
| US9176560B2 (en) | Use and state based power management | |
| Liang et al. | A parallel job execution time estimation approach based on user submission patterns within computational grids | |
| Li et al. | Energy-aware scheduling for spark job based on deep reinforcement learning in cloud | |
| Caux et al. | Smart datacenter electrical load model for renewable sources management |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20081203 |