CN103780646A - Cloud resource scheduling method and system - Google Patents
Cloud resource scheduling method and system Download PDFInfo
- Publication number
- CN103780646A CN103780646A CN201210404063.4A CN201210404063A CN103780646A CN 103780646 A CN103780646 A CN 103780646A CN 201210404063 A CN201210404063 A CN 201210404063A CN 103780646 A CN103780646 A CN 103780646A
- Authority
- CN
- China
- Prior art keywords
- server
- factor
- policy group
- value
- constraints
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention is suitable for the technical field of cloud computing and provides a cloud resource scheduling method and system. The method comprises the following steps: receiving a submitted resource request; based on a constraint condition of the resource request, screening out servers according with the constraint condition; based on external data values, internal status values and operation and maintenance data values, calculating efficiency values of the servers according with the constraint condition; and selecting the server, of which the efficiency value is the largest, from the servers according with the constraint condition and allocating the server, of which the efficiency value is the largest, to the resource request. The efficiency values of the serves are calculated based on historical records, performance statistics and external data input, the efficiency values are optimized by continuously optimizing calculating parameters of the efficiency values, and the resources are allocated to the resource request based on the efficiency values, so that dynamic and reasonable resource allocation can be realized, and resource allocation can be automatically optimized continuously.
Description
Technical field
The invention belongs to cloud computing technology field, relate in particular to a kind of dispatching method and system of cloud resource.
Background technology
Cloud computing utilizes the transmittability of high speed internet, the resources such as calculating, storage, software and services are transplanted to the extensive high-performance computer, personal computer, virtual machine of the Internet centralized management from the personal computer that disperses or server, thereby make user use these resources as using electric power, a kind of new computation schema has been explained in cloud computing: application, data and IT resource offer user in the mode of service by network and use.
Cloud scheduling of resource refers in a specific cloud environment, according to resource service regeulations, carries out the process of scheduling of resource between different resources.Under prior art, cloud platform resource scheduling strategy adopts average scheduling mode, random schedule mode or the utilization of resources to maximize the single tactful modes such as scheduling, make resource can not get rational utilization, thereby be difficult to rational Resources allocation, reduced the performance of whole task.
Summary of the invention
The object of the embodiment of the present invention is to provide a kind of method and system of cloud scheduling of resource, cannot realize the problem of resource rational utilization to solve prior art.
Embodiments of the invention are achieved in that a kind of method of cloud scheduling of resource, said method comprising the steps of:
Receive the resource request of submitting to;
According to the constraints of resource request, filter out the server that meets described constraints;
Calculating meets external data value, internal state value and the O&M data value of the server of described constraints;
According to default weights, outside data value, internal state value and O&M data value are weighted to summation, calculate the efficiency value of the server that meets constraints;
In the server that meets described constraints, select the server of efficiency value maximum, and give described resource request the server-assignment of described efficiency value maximum.
Another object of embodiments of the invention is to provide a kind of system of cloud scheduling of resource, and described system comprises:
Receiving element, for receiving the resource request of submission;
Screening unit, for according to the constraints of resource request, filters out the server that meets described constraints;
The first computing unit, for calculating external data value, internal state value and the O&M data value of the server that meets described constraints;
The second computing unit, for according to default weights, is weighted summation to outside data value, internal state value and O&M data value, calculates the efficiency value of the server that meets constraints;
Allocation units, for select the server of efficiency value maximum at the server that meets described constraints, and give described resource request the server-assignment of described efficiency value maximum.
The embodiment of the present invention is passed through server historical data, performance statistics and external data COMPREHENSIVE CALCULATING server efficiency value, realize processing by demand and dynamic assignment data, solve prior art and existed resource distribution mode single, the problem of resource unreasonable distribution.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skills, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is the realization flow figure of the method for cloud scheduling of resource provided by the invention;
Fig. 2 is the modular structure figure of the system of cloud scheduling of resource provided by the invention;
Fig. 3 is the realization flow figure of the method for the cloud scheduling of resource that provides of one embodiment of the invention;
Fig. 4 is the realization flow figure of the method for the cloud scheduling of resource that provides of another embodiment of the present invention;
Fig. 5 is the modular structure figure of the system of the cloud scheduling of resource that provides of another embodiment of the present invention;
Fig. 6 is the modular structure figure of the system of the cloud scheduling of resource that provides of another embodiment of the present invention;
Fig. 7 is the strategy group evolution schematic diagram that the embodiment of the present invention provides.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
The invention provides a kind of method of cloud scheduling of resource, as shown in Figure 1, concrete steps comprise described method:
The resource request that S101, reception are submitted to;
S102, according to the constraints of resource request, filter out the server that meets described constraints;
S103, calculating meet external data value, internal state value and the O&M data value of the server of described constraints;
S104, according to default weights, outside data value, internal state value and O&M data value are weighted to summation, calculate the efficiency value of server that meets constraints;
S105, in the each server that meets described constraints, select the server of efficiency value maximum, and give described resource request the server-assignment of described efficiency value maximum.
Optionally, the method that realizes S103 specifically can comprise:
According to the each assembly combination property of server P
prodcut, server failure rate R
error, brand public credibility S
brandwith product testing score P
evaluate, and external data in the initial policy group input EDI factor
with breakdown loss factor sigma, calculate the external data value of the server that meets constraints,
and
The average occupancy Avg_O of CPU obtaining according to collection
cpu, the average occupancy Avg_O of internal memory
memory, hard disk average read-write speed Avg_S
rw, switch average delay Avg_D
switch, the comprehensive caloric value H of server
serverwith server energy consumption C
server, and cpu load factor ω, internal memory load factor θ and input and output I/O raising efficiency factor mu in initial policy group, calculates the internal state value of the server that meets constraints,
and
According to the server O&M failure rate E in historical record
history, server uses duration T, server status record S
history, server performance parameter P
history, year power consumption C
power, year Operational Visit amount V, maximum concurrent number N and maximum application load N
apps,, and ODI factor set δ, server decay factor ε and concurrent load factor ρ in initial policy group, calculates the O&M data value of the server that meets constraints,
Optionally, said method also comprises:
By the default time interval, obtain optimal policy group, use the factor in optimal policy group to replace the factor in initial policy group.
Optionally, described pressing the default time interval, obtains optimal policy group, and the factor that uses the factor in optimal policy group to replace in initial policy group is specially:
By the default EDI factor
breakdown loss factor sigma, cpu load factor ω, internal memory load factor θ, I/O raising efficiency factor mu, ODI factor delta, server decay factor ε and an initial policy group of concurrent load factor ρ composition;
Use random function to generate M group randomized policy group, described M is greater than 1 integer, the EDI factor in randomized policy group
breakdown loss factor sigma, cpu load factor ω, internal memory load factor θ, I/O raising efficiency factor mu, ODI factor delta, server decay factor ε and concurrent load factor ρ are in threshold range;
The average occupancy Avg_O of CPU obtaining according to collection
cpu, the average occupancy Avg_O of internal memory
memorywith hard disk average read-write speed Avg_S
rwcarry out equivalent weighted sum and obtain the load value of server, in Servers-all, choose the server of load value minimum as simulated environment.
In default simulated environment, test randomized policy group, in the time that the operation of randomized policy group causes the direct fault of simulated environment, directly delete the randomized policy group that causes fault;
According to the default time interval, the performance parameter of the rear system of randomized policy group operation that randomized policy group more of the same clan is maximum, the optimum corresponding randomized policy group of performance parameter of getting the rear system of operation is optimal policy group, and uses the factor in optimal policy group to replace the factor in initial policy group.
The invention provides a kind of system of cloud scheduling of resource, described system as shown in Figure 2, specifically comprises:
Receiving element 21, for receiving the resource request of submission;
The first computing unit 23, for calculating external data value, internal state value and the O&M data value of the server that meets described constraints;
The second computing unit 24, for according to default weights, is weighted summation to outside data value, internal state value and O&M data value, calculates the efficiency value of the server that meets constraints;
Optionally, described the second computing unit 23 specifically for:
According to the each assembly combination property of server P
prodcut, server failure rate R
error, brand public credibility S
brandwith product testing score P
evaluate, and external data in the initial policy group input EDI factor
with breakdown loss factor sigma, calculate the external data value of the server that meets constraints,
and
The average occupancy Avg_O of CPU obtaining according to collection
cpu, the average occupancy Avg_O of internal memory
memory, hard disk average read-write speed Avg_S
rw, switch average delay Avg_D
switch, the comprehensive caloric value H of server
serverwith server energy consumption C
server, and cpu load factor ω, internal memory load factor θ and input and output I/O raising efficiency factor mu in initial policy group, calculates the internal state value of the server that meets constraints,
and
According to the server O&M failure rate E in historical record
history, server uses duration T, server status record S
history, server performance parameter P
history, year power consumption C
power, year Operational Visit amount V, maximum concurrent number N and maximum application load N
apps,, and ODI factor set δ, server decay factor ε and concurrent load factor ρ in initial policy group, calculates the O&M data value of the server that meets constraints,
Optionally, described system further comprises:
Replacement unit, for according to the default time interval, obtains optimal policy group, uses the factor in optimal policy group to replace the factor in initial policy group.
Optionally, described replacement unit is further used for:
By the default EDI factor
breakdown loss factor sigma, cpu load factor ω, internal memory load factor θ, I/O raising efficiency factor mu, ODI factor delta, server decay factor ε and an initial policy group of concurrent load factor ρ composition;
Use random function to generate M group randomized policy group, described M is greater than 1 integer, the EDI factor in randomized policy group
breakdown loss factor sigma, cpu load factor ω, internal memory load factor θ, I/O raising efficiency factor mu, ODI factor delta, server decay factor ε and concurrent load factor ρ are in threshold range;
The average occupancy Avg_O of CPU obtaining according to collection
cpu, the average occupancy Avg_O of internal memory
memorywith hard disk average read-write speed Avg_S
rwcarry out equivalent weighted sum and obtain the load value of server, divide default space size as simulated environment at the server of load value minimum.
In default simulated environment, test randomized policy group, in the time that the operation of randomized policy group causes the direct fault of simulated environment, directly delete the randomized policy group that causes fault;
According to the default time interval, the performance parameter of the rear system of randomized policy group operation that randomized policy group more of the same clan is maximum, the optimum corresponding randomized policy group of performance parameter of getting the rear system of operation is optimal policy group, and uses the factor in optimal policy group to replace the factor in initial policy group.
Fig. 3 shows the realization flow of the method for the cloud scheduling of resource that one embodiment of the invention provides, and details are as follows for the method process:
The resource request that S301, reception are submitted to;
S302, according to the constraints of resource request, in cloud system, filter out the server that meets described constraints;
In the present embodiment, suppose and have 10 station server clusters, a new request, constraints is that server O&M failure rate must not exceed the average occupancy of 0.2%, CPU and is no more than 70%, the application load amount on server is no more than 8.Three parameters such as the application load amount on server O&M failure rate, the average occupancy of CPU and the server of 10 station servers are as shown in table 1:
Table 1
In table 1, filter out this 4 station server of Server_2, Server_5, Server_9 and Server_10 according to constraints and meet constraints.
S303, calculating meet external data value, internal state value and the O&M data value of the server of constraints;
In the present embodiment, needing the server of considering is Server_2, Server_5, Server_9 and the Server_10 filtering out in step S302, calculates external data value, internal state value and the O&M data value of these servers.
External data input (External Data Input, EDI) refers to obtains from outside channel, to equipment in cloud resource pool or apply the relevant index such as reliability, actual performance; Cloud resource pool (Cloud pool, CP) refers to the IT resources such as server that cloud platform controls by Intel Virtualization Technology and distributed computing technology, memory device, network.Obtaining after the formation inventory of cloud resource pool, can connect the external data input that EDI center obtains the equipment in CP, also can direct labor import aforesaid external data input, the major parameter of external data input and the computational methods of external data value are as shown in table 2;
Table 2
In table 2, the EDI factor and the breakdown loss factor are to read from initial policy group.
Status data input (Status Data Input, SDI) is the data that monitor state obtains, and the major parameter of SDI and the computational methods of internal state value are as shown in table 3;
Table 3
In table 3, the cpu load factor, internal memory load factor and the I/O raising efficiency factor are to obtain from initial policy group.
O&M data inputs (Operation Data Input, ODI) refer to the state recording of accumulative total and the data of personnel's operation, and the major parameter of ODI and O&M data value are as shown in table 4;
Table 4
In table 4, the ODI factor, server decay factor and concurrent load factor are to obtain from initial policy group.
S304, calculating meet the efficiency value of the server of constraints;
In the present embodiment, first the efficiency value of each server calculates needs to set EDI weights, SDI weights and ODI weights, and described weights can be for being less than 1, be greater than 1 or equal 1 positive number;
Efficiency value Potency=EDI*EDI weights+SDI*SDI weights+ODI*ODI weights;
Suppose that in step S303, result of calculation is as shown in table 5, and EDI weights, SDI weights and ODI weights are all 1;
| Computer | Server_2 | Server_5 | Server_9 | Server_10 |
| EDI | 1702 | 2201 | 1910 | 1855 |
| SDI | 271 | 255 | 301 | 299 |
| ODI | 3301 | 2901 | 3088 | 3125 |
Table 5
According to data in table 5 and weights, the efficiency value that the efficiency value that the efficiency value that the efficiency value that can calculate Server_2 is 5274, Server_5 is 5357, Server_9 is 5299, Server_10 is 5279.
S305, in the described server that meets constraints, select in the server of efficiency value maximum, and give described resource request the server-assignment of described efficiency value maximum.
In the present embodiment, step S304 calculates the efficiency value maximum of Server_9, and therefore this resource request is assigned to Server_9.
Fig. 4 shows the realization flow of the method for the cloud scheduling of resource that another embodiment of the present invention provides, and details are as follows for the method process:
S401, by the default EDI factor
breakdown loss factor sigma, cpu load factor ω, internal memory load factor θ, I/O raising efficiency factor mu, ODI factor delta, server decay factor ε and an initial policy group of concurrent load factor ρ composition;
In the present embodiment, Figure 7 shows that example, before system brings into operation, preset above-mentioned whole factor assignment, and by an initial policy group of above-mentioned whole factor composition.
S402, use random function generate M(M and are greater than 1 integer) group randomized policy group, the EDI factor of described randomized policy group
breakdown loss factor sigma, cpu load factor ω, internal memory load factor θ, I/O raising efficiency factor mu, ODI factor delta, server decay factor ε and concurrent load factor ρ are in threshold range;
In the present embodiment, first whole factors are set to effective scope, in effective range, give at random different values by all factors of initial policy group; Figure 7 shows that example, start to generate 3 groups of randomized policy groups at first evolution cycle and be respectively strategy group 1, strategy group 2, strategy group 3; Before the second evolution cycle starts, strategy group 2 is because cause simulated environment fault and be eliminated in first cycle, start strategy group 1 random generation strategy group a, strategy group b and strategy group c at the second evolution cycle, strategy group 3 random generation strategy group e, strategy group f and strategy group g.
S403, the average occupancy Avg_O of the CPU obtaining according to collection
cpu, the average occupancy Avg_O of internal memory
memorywith hard disk average read-write speed Avg_S
rwcarry out equivalent weighted sum and obtain the load value of server, divide default space size as simulated environment at the server of load value minimum;
S404, in default simulated environment, test randomized policy group, in the time that the operation of randomized policy group causes the direct fault of simulated environment, directly delete the randomized policy group that causes fault;
In the present embodiment, Figure 7 shows that example, in first evolution cycle, strategy group 2 is deleted because causing simulated environment operation troubles, and in second evolution cycle, strategy organizes that b, c and g cause simulated environment operation troubles and deleted.
S405, according to the default time interval, obtain optimal policy group, and use the factor in optimal policy group to replace the factor in initial policy group.
In the present embodiment, default certain time interval, carries out once strategy group a cycle in the time interval and relatively, selects optimal policy group.The strategy of selecting optimal policy group is first to determine the maximum randomized policy group of normal strategy group of the same clan, the performance parameter optimum of system after operation be chosen in simulated environment in the maximum randomized policy group of strategy group of the same clan in, described performance parameter can be calculated and obtain according to the failure rate of operation, load balance, also can obtain according to the calculation of parameter of other operations, performance parameter computational methods specifically realize according to actual conditions demand;
Still Figure 7 shows that example, in first evolution cycle, the random variation of initial policy group has generated tactful group 1, strategy group 2 and strategy group 3, in simulated environment, strategy group 2 is eliminated because of operation troubles, strategy group 1 and strategy group 3 are all the variations of initial policy group, therefore their strategy group of the same clan is all the other side, therefore cannot find optimal policy group by quantity more of the same clan, suppose that the performance parameter that tactful group 1 is moved rear system is better than the tactful performance parameter of organizing the rear system of 3 operation, strategy group 1 is optimal policy group in the time that first evolution cycle finishes so, replace the factor of initial policy group by the factor of strategy group 1,
In the second evolution cycle, the strategy group 1 of survival is by random variation generation strategy group a, strategy group b and strategy group c, the strategy group 3 of survival is by random variation generation strategy group e, strategy group f and strategy group g, by the operation in simulated environment, strategy group a, strategy group e and strategy group f survive, strategy group e has strategy group f of the same clan, strategy group f has strategy group e of the same clan, and tactful tuple a is without tactful group of the same clan, therefore select optimal policy group can compare by comparison strategy group e and the postrun performance parameter of tactful ancestral f, in strategy group e and strategy group f, select optimal policy group, and replace the factor of initial policy group by the factor of optimal policy group,
Make to use the same method in the follow-up cycle and select optimal policy group, and replace the factor of initial policy group by the factor of optimal policy group, also have much for the method for optimal policy group selection, do not repeat one by one at this.
In the present embodiment, carry out the random mode generating by the factor to default, in the environment of simulation, implement and compare result of implementation, guaranteed the more and more optimization of the factor; The factor is to affect the variable that efficiency value calculates, and the optimization of the factor can cause efficiency value optimization, and final cloud system is more and more reasonable to the distribution of resource.
The modular structure figure that Figure 5 shows that the system of the cloud scheduling of resource that another embodiment of the present invention provides, for convenience of explanation, only illustrates the part relevant to the embodiment of the present invention.
The system of this cloud scheduling of resource comprises:
Receiving element 51, for receiving the resource request of submission;
The first computing unit 53, for calculating external data value, internal state value and the O&M data value of the server that meets constraints
The second computing unit 54, for calculating the efficiency value of the server that meets constraints;
The modular structure figure that Figure 6 shows that the system of the cloud scheduling of resource that another embodiment of the present invention provides, for convenience of explanation, only illustrates the part relevant to the embodiment of the present invention.
The system of this cloud scheduling of resource comprises:
Described replacement unit 61 specifically comprises:
Simulated environment arranges module 613, for the average occupancy Avg_O of CPU obtaining according to collection
cpu, the average occupancy Avg_O of internal memory
memorywith hard disk average read-write speed Avg_S
rwcarry out equivalent weighted sum and obtain the load value of server, divide default space size as simulated environment at the server of load value minimum;
One of ordinary skill in the art will appreciate that and just divide according to function logic for above-described embodiment included modules, but be not limited to above-mentioned division, as long as can realize corresponding function; In addition, the concrete title of each functional module also, just for the ease of mutual differentiation, is not limited to protection scope of the present invention.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any modifications of doing within the spirit and principles in the present invention, be equal to and replace and improvement etc., within all should being included in protection scope of the present invention.
Claims (8)
1. a dispatching method for cloud resource, is characterized in that, described method comprises:
Receive the resource request of submitting to;
According to the constraints of resource request, filter out the server that meets described constraints;
Calculating meets external data value, internal state value and the O&M data value of the server of described constraints;
According to default weights, outside data value, internal state value and O&M data value are weighted to summation, calculate the efficiency value of the server that meets constraints;
In the server that meets described constraints, select the server of efficiency value maximum, and give described resource request the server-assignment of described efficiency value maximum.
2. the method for claim 1, is characterized in that, external data value, internal state value and O&M data value that described calculating meets the server of described constraints are specially:
According to the each assembly combination property of server P
prodcut, server failure rate R
error, brand public credibility S
brandwith product testing score P
evaluate, and external data in the initial policy group input EDI factor
with breakdown loss factor sigma, calculate the external data value of the server that meets constraints,
and
The average occupancy Avg_O of CPU obtaining according to collection
cpu, the average occupancy Avg_O of internal memory
memory, hard disk average read-write speed Avg_S
rw, switch average delay Avg_D
switch, the comprehensive caloric value H of server
serverwith server energy consumption C
server, and cpu load factor ω, internal memory load factor θ and input and output I/O raising efficiency factor mu in initial policy group, calculates the internal state value of the server that meets constraints,
and
According to the server O&M failure rate E in historical record
history, server uses duration T, server status record S
history, server performance parameter P
history, year power consumption C
power, year Operational Visit amount V, maximum concurrent number N and maximum application load N
apps,, and ODI factor set δ, server decay factor ε and concurrent load factor ρ in initial policy group, calculates the O&M data value of the server that meets constraints,
3. method as claimed in claim 1 or 2, is characterized in that, described method also comprises:
By the default time interval, obtain optimal policy group, use the factor in optimal policy group to replace the factor in initial policy group.
4. method as claimed in claim 3, described pressing the default time interval, obtains optimal policy group, and the factor that uses the factor in optimal policy group to replace in initial policy group is specially:
By the default EDI factor
breakdown loss factor sigma, cpu load factor ω, internal memory load factor θ, I/O raising efficiency factor mu, ODI factor delta, server decay factor ε and an initial policy group of concurrent load factor ρ composition;
Use random function to generate M group randomized policy group, described M is greater than 1 integer, the EDI factor in randomized policy group
breakdown loss factor sigma, cpu load factor ω, internal memory load factor θ, I/O raising efficiency factor mu, ODI factor delta, server decay factor ε and concurrent load factor ρ are in threshold range;
The average occupancy Avg_O of CPU obtaining according to collection
cpu, the average occupancy Avg_O of internal memory
memorywith hard disk average read-write speed Avg_S
rwcarry out equivalent weighted sum and obtain the load value of server, divide default space size as simulated environment at the server of load value minimum.
In default simulated environment, test randomized policy group, in the time that the operation of randomized policy group causes the direct fault of simulated environment, directly delete the randomized policy group that causes fault;
According to the default time interval, the performance parameter of the rear system of randomized policy group operation that randomized policy group more of the same clan is maximum, the optimum corresponding randomized policy group of performance parameter of getting the rear system of operation is optimal policy group, and uses the factor in optimal policy group to replace the factor in initial policy group.
5. a dispatching patcher for cloud resource, is characterized in that, described system comprises:
Receiving element, for receiving the resource request of submission;
Screening unit, for according to the constraints of resource request, filters out the server that meets described constraints;
The first computing unit, for calculating external data value, internal state value and the O&M data value of the server that meets described constraints;
The second computing unit, for according to default weights, is weighted summation to outside data value, internal state value and O&M data value, calculates the efficiency value of the server that meets constraints;
Allocation units, for select the server of efficiency value maximum at the server that meets described constraints, and give described resource request the server-assignment of described efficiency value maximum.
6. system as claimed in claim 5, is characterized in that, described the first computing unit specifically for:
According to the each assembly combination property of server P
prodcut, server failure rate R
error, brand public credibility S
brandwith product testing score P
evaluate, and external data in the initial policy group input EDI factor
with breakdown loss factor sigma, calculate the external data value of the server that meets constraints,
and
The average occupancy Avg_O of CPU obtaining according to collection
cpu, the average occupancy Avg_O of internal memory
memory, hard disk average read-write speed Avg_S
rw, switch average delay Avg_D
switch, the comprehensive caloric value H of server
serverwith server energy consumption C
server, and cpu load factor ω, internal memory load factor θ and input and output I/O raising efficiency factor mu in initial policy group, calculates the internal state value of the server that meets constraints,
and
According to the server O&M failure rate E in historical record
history, server uses duration T, server status record S
history, server performance parameter P
history, year power consumption C
power, year Operational Visit amount V, maximum concurrent number N and maximum application load N
apps,, and ODI factor set δ, server decay factor ε and concurrent load factor ρ in initial policy group, calculates the O&M data value of the server that meets constraints,
7. the system as described in claim 5 or 6, is characterized in that, described system further comprises:
Replacement unit, for according to the default time interval, obtains optimal policy group, uses the factor in optimal policy group to replace the factor in initial policy group.
8. system as claimed in claim 7, is characterized in that, described replacement unit further specifically for:
By the default EDI factor
breakdown loss factor sigma, cpu load factor ω, internal memory load factor θ, I/O raising efficiency factor mu, ODI factor delta, server decay factor ε and an initial policy group of concurrent load factor ρ composition;
Use random function to generate M group randomized policy group, described M is greater than 1 integer, the EDI factor in randomized policy group
breakdown loss factor sigma, cpu load factor ω, internal memory load factor θ, I/O raising efficiency factor mu, ODI factor delta, server decay factor ε and concurrent load factor ρ are in threshold range;
The average occupancy Avg_O of CPU obtaining according to collection
cpu, the average occupancy Avg_O of internal memory
memorywith hard disk average read-write speed Avg_S
rwcarry out equivalent weighted sum and obtain the load value of server, dividing default space size as simulated environment at the server of load value minimum.
In default simulated environment, test randomized policy group, in the time that the operation of randomized policy group causes the direct fault of simulated environment, directly delete the randomized policy group that causes fault;
According to the default time interval, the performance parameter of the rear system of randomized policy group operation that randomized policy group more of the same clan is maximum, the optimum corresponding randomized policy group of performance parameter of getting the rear system of operation is optimal policy group, and uses the factor in optimal policy group to replace the factor in initial policy group.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210404063.4A CN103780646B (en) | 2012-10-22 | 2012-10-22 | Cloud resource scheduling method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210404063.4A CN103780646B (en) | 2012-10-22 | 2012-10-22 | Cloud resource scheduling method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103780646A true CN103780646A (en) | 2014-05-07 |
| CN103780646B CN103780646B (en) | 2017-04-12 |
Family
ID=50572471
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210404063.4A Active CN103780646B (en) | 2012-10-22 | 2012-10-22 | Cloud resource scheduling method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103780646B (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105099782A (en) * | 2015-08-18 | 2015-11-25 | 北京京东世纪贸易有限公司 | Method and system for controlling big data resource of cloud environment cluster |
| CN105162840A (en) * | 2015-07-31 | 2015-12-16 | 清华大学 | Online distributed request scheduling and service management method and agent system |
| CN105335229A (en) * | 2014-07-25 | 2016-02-17 | 杭州华三通信技术有限公司 | Business resource scheduling method and apparatus |
| CN105763636A (en) * | 2016-04-15 | 2016-07-13 | 北京思特奇信息技术股份有限公司 | Optimal host selection method and system in distributed system |
| WO2017128980A1 (en) * | 2016-01-29 | 2017-08-03 | 华为技术有限公司 | Method and device for managing resources in cloud platform |
| CN107277143A (en) * | 2017-06-21 | 2017-10-20 | 郑州云海信息技术有限公司 | A kind of resource matched management method and device |
| CN107645397A (en) * | 2016-07-21 | 2018-01-30 | 阿里巴巴集团控股有限公司 | A kind of system, device and method that fault simulation is carried out in distributed system |
| WO2018072619A1 (en) * | 2016-10-20 | 2018-04-26 | 阿里巴巴集团控股有限公司 | Resource scheduling method, device, and system |
| CN108780403A (en) * | 2016-03-10 | 2018-11-09 | 沃拉斯堤技术解决方案公司 | To the system and method being managed for the cloud computing resources of information system |
| CN105306597B (en) * | 2015-11-19 | 2019-05-14 | 国家电网公司 | A kind of cloud computing service method of quality control |
| CN110019313A (en) * | 2017-12-30 | 2019-07-16 | 中国移动通信集团四川有限公司 | Method for optimizing resources, device, equipment and medium |
| CN110691118A (en) * | 2019-08-30 | 2020-01-14 | 许昌许继软件技术有限公司 | Service selection method and device in micro-service cluster |
| CN110764904A (en) * | 2019-09-23 | 2020-02-07 | 合肥中科类脑智能技术有限公司 | Resource scheduling method and system based on FPGA chip |
| CN113392988A (en) * | 2021-05-10 | 2021-09-14 | 贵州乌江水电开发有限责任公司乌江渡发电厂 | Maintenance file management method for paperless operation of hydraulic power plant |
| CN116483013A (en) * | 2023-06-19 | 2023-07-25 | 成都实时技术股份有限公司 | High-speed signal acquisition system and method based on multichannel collector |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1300988A (en) * | 1999-06-17 | 2001-06-27 | 国际商业机器公司 | System and method for comprehensive load distribution and source management in Internet network |
| US20020059429A1 (en) * | 1999-01-20 | 2002-05-16 | Geoffrey C. Carpenter | Client-side method and apparatus for improving the availability and performance of network mediated services |
| US20020198995A1 (en) * | 2001-04-10 | 2002-12-26 | International Business Machines Corporation | Apparatus and methods for maximizing service-level-agreement profits |
| US20040230317A1 (en) * | 2003-05-15 | 2004-11-18 | Sun Microsystems, Inc. | Method, system, and program for allocating storage resources |
| CN101026617A (en) * | 2006-02-18 | 2007-08-29 | 华为技术有限公司 | Media resource scheduling method for IMS network |
| CN101067823A (en) * | 2006-05-02 | 2007-11-07 | 国际商业机器公司 | System and method for optimizing federated databases magagement |
| CN101610433A (en) * | 2009-07-10 | 2009-12-23 | 北京邮电大学 | The multi-constraint condition routing selection method that a kind of support policy is resolved |
| CN101841526A (en) * | 2010-03-04 | 2010-09-22 | 清华大学 | Cluster streaming media server system applied to large-scale user demand |
| CN101909196A (en) * | 2009-06-05 | 2010-12-08 | 华为技术有限公司 | Channel-switching handling method, system and related equipment |
| EP2357561A1 (en) * | 2004-03-13 | 2011-08-17 | Adaptive Computing Enterprises, Inc. | System and method for providing advanced reservations in a compute environment |
| CN102624922A (en) * | 2012-04-11 | 2012-08-01 | 武汉大学 | Method for balancing load of network GIS heterogeneous cluster server |
| CN102681889A (en) * | 2012-04-27 | 2012-09-19 | 电子科技大学 | Scheduling method of cloud computing open platform |
| CN102739770A (en) * | 2012-04-18 | 2012-10-17 | 上海和辰信息技术有限公司 | Method and system for scheduling resource based on cloud computing |
-
2012
- 2012-10-22 CN CN201210404063.4A patent/CN103780646B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020059429A1 (en) * | 1999-01-20 | 2002-05-16 | Geoffrey C. Carpenter | Client-side method and apparatus for improving the availability and performance of network mediated services |
| CN1300988A (en) * | 1999-06-17 | 2001-06-27 | 国际商业机器公司 | System and method for comprehensive load distribution and source management in Internet network |
| US20020198995A1 (en) * | 2001-04-10 | 2002-12-26 | International Business Machines Corporation | Apparatus and methods for maximizing service-level-agreement profits |
| US20040230317A1 (en) * | 2003-05-15 | 2004-11-18 | Sun Microsystems, Inc. | Method, system, and program for allocating storage resources |
| EP2357561A1 (en) * | 2004-03-13 | 2011-08-17 | Adaptive Computing Enterprises, Inc. | System and method for providing advanced reservations in a compute environment |
| CN101026617A (en) * | 2006-02-18 | 2007-08-29 | 华为技术有限公司 | Media resource scheduling method for IMS network |
| CN101067823A (en) * | 2006-05-02 | 2007-11-07 | 国际商业机器公司 | System and method for optimizing federated databases magagement |
| CN101909196A (en) * | 2009-06-05 | 2010-12-08 | 华为技术有限公司 | Channel-switching handling method, system and related equipment |
| CN101610433A (en) * | 2009-07-10 | 2009-12-23 | 北京邮电大学 | The multi-constraint condition routing selection method that a kind of support policy is resolved |
| CN101841526A (en) * | 2010-03-04 | 2010-09-22 | 清华大学 | Cluster streaming media server system applied to large-scale user demand |
| CN102624922A (en) * | 2012-04-11 | 2012-08-01 | 武汉大学 | Method for balancing load of network GIS heterogeneous cluster server |
| CN102739770A (en) * | 2012-04-18 | 2012-10-17 | 上海和辰信息技术有限公司 | Method and system for scheduling resource based on cloud computing |
| CN102681889A (en) * | 2012-04-27 | 2012-09-19 | 电子科技大学 | Scheduling method of cloud computing open platform |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105335229A (en) * | 2014-07-25 | 2016-02-17 | 杭州华三通信技术有限公司 | Business resource scheduling method and apparatus |
| CN105335229B (en) * | 2014-07-25 | 2020-07-07 | 新华三技术有限公司 | Scheduling method and device of service resources |
| CN105162840A (en) * | 2015-07-31 | 2015-12-16 | 清华大学 | Online distributed request scheduling and service management method and agent system |
| CN105162840B (en) * | 2015-07-31 | 2018-12-28 | 清华大学 | Distribution on line formula request scheduling and service management and agency plant |
| CN105099782A (en) * | 2015-08-18 | 2015-11-25 | 北京京东世纪贸易有限公司 | Method and system for controlling big data resource of cloud environment cluster |
| CN105306597B (en) * | 2015-11-19 | 2019-05-14 | 国家电网公司 | A kind of cloud computing service method of quality control |
| WO2017128980A1 (en) * | 2016-01-29 | 2017-08-03 | 华为技术有限公司 | Method and device for managing resources in cloud platform |
| US10805385B2 (en) | 2016-01-29 | 2020-10-13 | Huawei Technologies Co., Ltd. | Method and apparatus for managing resource on cloud platform |
| CN108780403A (en) * | 2016-03-10 | 2018-11-09 | 沃拉斯堤技术解决方案公司 | To the system and method being managed for the cloud computing resources of information system |
| CN105763636B (en) * | 2016-04-15 | 2019-01-15 | 北京思特奇信息技术股份有限公司 | The selection method and system of optimal host in a kind of distributed system |
| CN105763636A (en) * | 2016-04-15 | 2016-07-13 | 北京思特奇信息技术股份有限公司 | Optimal host selection method and system in distributed system |
| CN107645397A (en) * | 2016-07-21 | 2018-01-30 | 阿里巴巴集团控股有限公司 | A kind of system, device and method that fault simulation is carried out in distributed system |
| CN107645397B (en) * | 2016-07-21 | 2021-01-05 | 阿里巴巴集团控股有限公司 | System, device and method for fault simulation in distributed system |
| TWI780065B (en) * | 2016-10-20 | 2022-10-11 | 香港商阿里巴巴集團服務有限公司 | Method, apparatus and system for resource scheduling |
| CN107967173A (en) * | 2016-10-20 | 2018-04-27 | 阿里巴巴集团控股有限公司 | A kind of methods, devices and systems of scheduling of resource |
| WO2018072619A1 (en) * | 2016-10-20 | 2018-04-26 | 阿里巴巴集团控股有限公司 | Resource scheduling method, device, and system |
| CN107277143A (en) * | 2017-06-21 | 2017-10-20 | 郑州云海信息技术有限公司 | A kind of resource matched management method and device |
| CN110019313A (en) * | 2017-12-30 | 2019-07-16 | 中国移动通信集团四川有限公司 | Method for optimizing resources, device, equipment and medium |
| CN110691118A (en) * | 2019-08-30 | 2020-01-14 | 许昌许继软件技术有限公司 | Service selection method and device in micro-service cluster |
| CN110764904A (en) * | 2019-09-23 | 2020-02-07 | 合肥中科类脑智能技术有限公司 | Resource scheduling method and system based on FPGA chip |
| CN110764904B (en) * | 2019-09-23 | 2022-10-28 | 北京中科云脑智能技术有限公司 | Resource scheduling method and system based on FPGA chip |
| CN113392988A (en) * | 2021-05-10 | 2021-09-14 | 贵州乌江水电开发有限责任公司乌江渡发电厂 | Maintenance file management method for paperless operation of hydraulic power plant |
| CN113392988B (en) * | 2021-05-10 | 2023-06-09 | 贵州乌江水电开发有限责任公司乌江渡发电厂 | Maintenance file management method for paperless operation of hydropower plant |
| CN116483013A (en) * | 2023-06-19 | 2023-07-25 | 成都实时技术股份有限公司 | High-speed signal acquisition system and method based on multichannel collector |
| CN116483013B (en) * | 2023-06-19 | 2023-09-05 | 成都实时技术股份有限公司 | High-speed signal acquisition system and method based on multichannel collector |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103780646B (en) | 2017-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103780646B (en) | Cloud resource scheduling method and system | |
| CN110022226B (en) | Object-oriented data acquisition system and acquisition method | |
| CN110225074B (en) | Communication message distribution system and method based on equipment address domain | |
| CN104038540B (en) | Method and system for automatically selecting application proxy server | |
| CN109617099B (en) | Virtual energy storage coordination control system and method thereof | |
| CN103248659A (en) | Method and system for dispatching cloud computed resources | |
| CN102694868A (en) | Cluster system implementation and task dynamic distribution method | |
| CN104133727A (en) | Load distribution method based on real-time resources | |
| CN102929961A (en) | Data processing method and device thereof based on building quick data staging channel | |
| CN102025753B (en) | Load balancing method and equipment for data resources of servers | |
| CN105071994B (en) | A kind of mass data monitoring system | |
| CN104283959A (en) | Performance-grading-based storage mechanism suitable for cloud platform | |
| CN102637138A (en) | Method for computing and scheduling virtual machine | |
| CN103577337B (en) | Space allocation method and device for storage system | |
| CN106681823A (en) | Load balancing method for processing MapReduce data skew | |
| Mylavarapu et al. | An optimized capacity planning approach for virtual infrastructure exhibiting stochastic workload | |
| CN106326012A (en) | Web application cluster buffer utilization method and system | |
| CN104111875A (en) | Device, system and method for dynamically controlling number of newly-increased tasks at cloud data center | |
| CN103501509B (en) | A kind of method and device equalizing radio network controller load | |
| CN104378239A (en) | Rapid reliability index statistical system and method based on cluster frame | |
| CN103902310A (en) | Scheduling system and method for starting of virtual machines | |
| CN107277143A (en) | A kind of resource matched management method and device | |
| CN112968962B (en) | Cloud platform storage resource scheduling method based on distributed computer cluster architecture | |
| CN109522129A (en) | A kind of resource method for dynamically balancing, device and relevant device | |
| CN112612579B (en) | Virtual machine deployment method, storage medium and computer equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 518057 computer building of the Great Wall, Nanshan District science and Technology Park, Shenzhen, Guangdong Patentee after: China the Great Wall science and technology group Limited by Share Ltd Address before: 518057 computer building of the Great Wall, Nanshan District science and Technology Park, Shenzhen, Guangdong Patentee before: China Changcheng Computer Shenzhen Co., Ltd. |
|
| CP01 | Change in the name or title of a patent holder |