CN110515720B - Cloud computing system service cost and reliability driven job scheduling method - Google Patents
Cloud computing system service cost and reliability driven job scheduling method Download PDFInfo
- Publication number
- CN110515720B CN110515720B CN201910826099.3A CN201910826099A CN110515720B CN 110515720 B CN110515720 B CN 110515720B CN 201910826099 A CN201910826099 A CN 201910826099A CN 110515720 B CN110515720 B CN 110515720B
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- job
- reliability
- cloud
- service
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45591—Monitoring or debugging support
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45595—Network integration; Enabling network access in virtual machine instances
Abstract
The method comprises the following steps that 1), a cloud computing system is established, wherein the cloud computing system is composed of a user module, an operation queue module, a resource management module, an operation scheduling module, a reliability analysis module, a physical resource layer module and a virtual machine; initializing the virtual machine, and receiving the operation submitted by a cloud computing system user to an operation queue; 2) And inquiring whether the job waiting queue has the job to be scheduled or not by the job scheduling method. If not, the job scheduling algorithm is ended; if so, each job in the job waiting queue performs step 3), step 4), step 5), step 6), step 7); and so on. The method can comprehensively compromise and optimize the service cost and the computing reliability under the constraint of the cloud service response time to realize the efficient scheduling of the service request of the cloud computing system, thereby improving the system performance.
Description
Technical Field
The invention relates to the technical field of job scheduling methods driven by job adjustment cloud computing system service cost and reliability of a cloud computing system, in particular to a job scheduling method driven by the cloud computing system service cost and reliability.
Background
With the rapid deployment of on-demand virtualized IT infrastructures, such as servers, storage and databases, more and more cloud services are being offered to individuals and organizations in a pay-as-you-go mode over wired and wireless internet. This is due to the most prominent resource delivery model of cloud computing, where all resources, such as machines, networks, applications, software, platforms, can be delivered as services to users. Similar to socioeconomic services, cloud quality of service (QoS), such as security, sustainability, reliability, energy consumption, cost, etc., have become important challenges for cloud service providers and are receiving more and more attention from enterprises and academic circles. Among all these challenges, reliability and cost are two core issues that must be addressed by both cloud users and providers.
There are many incidents that affect the reliability of cloud services, including service delays, service outages, service errors, data loss, long response times, etc. These events not only affect the service reliability of the cloud computing provider, but even cause significant economic loss and some detrimental social impact. For example, recent interruptions in Netfix, AWS, hua ye, twitter, tencent, and Facebook all cause significant losses to users and cloud service providers. One well-known technique for addressing the reliability of cloud services is fault tolerance, which attempts to ensure that a cloud service will be sustainable through one or more backups when it fails. Cloud computing system fault tolerance may be achieved by scheduling services (or jobs) to different virtual machines, which may have multiple copies, some of which failures may be tolerated. The most common fault tolerance mechanism is a main/backup (PB) cloud computing system service cost and reliability driven job scheduling method, where each cloud service is arranged on a main virtual machine and a backup virtual machine. However, these multi-copy reliability enhancement techniques require the use of more redundant system virtual machines, which necessarily results in high cloud service costs.
Disclosure of Invention
The technical problem to be solved by the invention is to overcome the defects in the prior art and provide a cloud computing system service cost and reliability driven job scheduling method for realizing efficient scheduling of cloud computing system service requests by comprehensively compromising and optimizing service cost and computing reliability.
The technical scheme adopted by the invention for solving the technical problem is as follows: the job scheduling method driven by the service cost and the reliability of the cloud computing system comprises the following steps:
1) Establishing a cloud computing system, wherein the cloud computing system is composed of a user module, an operation queue module, a resource management module, an operation scheduling module, a reliability analysis module, a physical resource layer module and a virtual machine; initializing the virtual machine, and receiving the operation submitted by a cloud computing system user to an operation queue;
2) Is job scheduling method query job waiting queue for job scheduling required? If not, the job scheduling algorithm is ended; if so, each job in the job waiting queue performs step 3), step 4), step 5), step 6), step 7);
3) Calculating the response time satisfaction condition of each job, and cloud service s i In a virtual machine vm j Above, its start execution time ES(s) i ，vm j ) Calculating the time st i，j Virtual machine vm j Time delay sd (vm) of j ) Whether the sum of (A) and (B) is less than or equal to the cloud service s i Ar(s) of i ) And a response time rt(s) i ) Summing; if not, the virtual machine cannot be used as the cloud service s i And (6) candidate virtual machines. For the conditions satisfied, the method proposed by this patent will perform step 4);
4) Calculating the real time RT(s) that the virtual machine has executed i ，vm j ) Computing cloud service requests s i Reliability of (2)
Judging cloud services s i In virtual machine vm j Whether upper reliability is less than or equal to virtual machine vm j Reliability threshold prTr (vm) j ) As shown in the following formula: PR si，vmj ≤prTr(vm j ) If not, then the cloud service request s is not required i Providing a backup virtual machine, directly jumping to the step 6), and if the backup virtual machine is met, entering the step 5) to request s for cloud service i Searching for an optimal backup virtual machine except the virtual machine;
5) And for each other virtual machine, if the response time of the operation in the step 3) is met, calculating a cloud service operation s i In virtual machine vm j Cost above, find virtual machine vm with minimum service cost k The virtual machine is taken as a cloud service request s i The backup virtual machine of (1);
6) Taking the cloud service cost calculation method in the step 5) as a core, and performing service composition calculation on the main virtual machine or the cloud service with the main/backup virtual machine;
7) For each job and each virtual machine, find the minimum cost of service c(s) i ) And assigning the job to the corresponding primary virtual machine or primary/backup virtual machine; updating virtual machine parameters, such as virtual machine availability time avail (vm) j ) And deleting the job from the queue waiting queue; returning to the step 2) to continue the method.
Further, virtualVirtual machine is provided with a physical server set PS = { PS = { PS } 1 ，ps 2 ，…，ps A And the device comprises a memory pt and a communication network system pn.
Further, the service request is described as a job set S with mutual independence, and the cloud service job S is calculated according to the following formula i The computing time st of the virtual machine i，j ：
Wherein s is i Representing a cloud service job request, w(s) i ) Representation cloud service s i Calculated amount of (c), w (vm) j ) Representing virtual machines vm j The computing processing power of (1).
Further, the failure rate of the physical resources of the physical server set obeys Poisson distribution, and the probability density function of the physical resources is calculated according to the following formula
Calculating a reliability function of the physical resource according to the following formula
Wherein psa denotes a physical server, λ pt Presentation memory, λ pn Representing a network.
Further, in the step 3), scheduling to the virtual machine vm j The cloud service job s processed i Calculating a start execution time ES(s) of the cloud service job according to the following formula i ，vm j )：
ES(s i ，vm j )＝Max{ar(s i )，avail(vm j ) Therein avail (vm) j ) Representing virtual machines vm j The available time of (c);
the response time of each job is calculated according to the following formula:
ES(s i ，vm j )+st i，j +sd(vm j )≤ar(s i )+rt(s i ).
further, in the step 4), the real-time that the virtual machine has executed is calculated as RT(s) according to the following formula i ，vm j )＝ES(s i ，vm j )+st i，j +runTime(vm j ) Said virtual machine vm j Handling cloud service requests s i Reliability of (2)
For the product of all physical resource reliabilities, the reliability is calculated according to the following formula
Further, in the step 5), the cloud service job request s is calculated according to the following formula i In virtual machine vm j Cost c(s) of i ，vm j )：c(s i ，vm j )＝st i，j ×cc(vm j )+ss(s i )×sc(vm j )。
Further, in the step 6), the service cost c(s) is calculated according to the following formula i )：
The working principle is as follows:
1) Establishing a cloud computing system, wherein the cloud computing system is composed of a user module, an operation queue module, a resource management module, an operation scheduling module, a reliability analysis module, a physical resource layer module and a virtual machine;
2) The user puts forward a service request to the cloud computing system, the service request enters the operation queue module and is managed by the resource management module and the operation scheduling module in a unified mode;
3) The resource management module calls a service reliability analysis function to obtain the reliability of cloud service jobs executed on each virtual machine of the cloud computing, and the job scheduling module makes a scheduling scheme for compromising and optimizing service cost and computing reliability according to job execution reliability, job computing cost and job response time and schedules jobs to the corresponding virtual machines for execution;
4) And the cloud computing system feeds back the execution result of each virtual machine to the cloud service user.
The method has the advantages that the method can comprehensively compromise and optimize the service cost and the computing reliability under the constraint of the cloud service response time to realize the efficient scheduling of the cloud computing system service request, thereby improving the system performance.
Drawings
FIG. 1 is a flowchart of a cloud computing system service cost and reliability driven job scheduling method provided by the present invention;
FIG. 2 is a cloud computing system architecture diagram provided by an implementation of the present invention;
fig. 3 is a comparison diagram of experiments performed on 20 virtual machines according to the present invention.
Detailed Description
The following examples are given to further illustrate the embodiments of the present invention:
as shown in fig. 1 and fig. 2, an embodiment of a job scheduling method driven by service cost and reliability of a cloud computing system is mainly composed of modules such as a user, a job queue, resource management and job scheduling, reliability analysis, a physical resource layer and a virtual machine layer of the cloud computing system. The user puts service requests to the cloud computing system, the services firstly enter a working queue and are uniformly managed by a resource management and job scheduling module. The resource management module calls a service reliability analysis function to acquire the reliability of the operation executed on each virtual machine of the cloud computing, and the operation scheduling module makes a scheduling scheme for compromising the optimization service cost and the computation reliability according to the operation execution reliability, the operation computation cost and the operation response time and schedules the operation to the corresponding virtual machine for execution. And finally, the cloud computing system feeds back the execution result of each virtual machine to the cloud service user.
The cloud service request is described as a set of jobs S = { S } having mutual independence 1 ，s 2 ，…，s n H, where w(s) i ) Representing cloud services s i Calculated amount of(s), ss(s) i ) G denotes cloud service s i The required amount of memory, ar(s) i ) Representing cloud services s i System time to arrival and representation cloud service s i Response time rt(s) i ). The description of the cloud computing system virtual machine set is omega = { vm 1 ，vm 2 ，…，vm m Where w (vm) j ) Representing virtual machines vm j Computing processing power of, thus, jobs i The computation time st of the virtual machine i，j Comprises the following steps:
cc(vm j ) Representing virtual machines vm j Calculated cost per hour, sc (vm) j ) Representing virtual machines vm j Storage cost per G, sd (vm) j ) Is a virtual machine vm j Time delay, runTime (vm) j ) Is a virtual machine vm j Run time and prTr (vm) j ) Representing virtual machines vm j And (4) a reliability threshold value.
Table 1 is some cloud computing system virtual machine configuration examples thereof.
Table 1 cloud computing system virtual machine configuration table
In this embodiment, each cloud virtual machine PS = { PS ] by physical server set 1 ，ps 2 ，…，ps A And the device comprises a memory pt and a communication network system pn. This patent assumes that all physical resources fail to efficiently serve the poisson distribution, and thus its failure parameters are, for example, physical servers
Memory lambda pt And network lambda pn . Here, the physical resource index probability density function is f, e.g., physical server ps a A probability density function of
Thus, the reliability function of the resource
Can be expressed as
For scheduling to virtual machine vm j On-processing cloud service s i It starts execution time ES(s) i ，vm j ) Is composed of
ES(s i ，vm j )＝Max{ar(s i )，avail(vm j )}. (3)
Here avail (vm) j ) Is a virtual machine vm j The available time of (c). Thus, the real-time that the virtual machine has been executing is
RT(s i ，vm j )＝ES(s i ，vm j )+st i，j +runTime(vm j ). (4)
Therefore, the virtual machine vm j Processing cloud service requests s i Reliability of (2)
Is the product of all physical resource reliabilities, i.e.
The operation scheduling method firstly initializes the virtual machines, wherein 10 virtual machines are established, the computing capacity of each virtual machine is 10000, and the price is 0.75$/hr; each virtual machine has 2 physical servers, and the failure parameter lambda =0.00002; the storage space of the virtual machine is 1000GB, the price is 0.23$/G, the failure rate is lambda =0.0001, and the failure rate of the communication network is lambda =0.00005. The second type of virtual machine has 5 virtual machines, each virtual machine has 25000 computing power and 0.98$/hr price; each virtual machine has 1 physical server, and the failure parameter lambda =0.00001; the virtual machine storage space is 1000GB, the price is 0.45$/G, the failure rate is lambda =0.00015, and the communication network failure rate is lambda =0.00003. The third type of virtual machine has 3 virtual machines, each computing power is 8000, and the price is 0.3$/hr; each virtual machine has 1 physical server, and the failure parameter lambda =0.0001; the virtual machine storage space is 2000GB, the price is 0.15$/G, the failure rate is lambda =0.00008, and the communication network failure rate is lambda =0.0002. The fourth type of virtual machine has 2 virtual machines, each of which has the calculation capacity of 40000 and the price of 1.1$/hr; each virtual machine has 5 physical servers, and the failure parameter lambda =0.00001; the virtual machine storage space is 500GB, the price is 0.32$/G, the failure rate is lambda =0.00052, and the communication network failure rate is lambda =0.00001.
Secondly, is the job scheduling method proposed in this patent inquiring whether there is a job waiting queue that needs to be scheduled? If not, the job scheduling algorithm is ended; if so, the scheduling method proposed by the patent performs the following steps for each job in the job waiting queue.
Third, for any job, the method calculates that the response time on each virtual machine satisfies the condition, as shown in the following formula:
ES(s i ，vm j )+st i，j +sd(vm j )≤ar(s i )+rt(s i ). (6)
i.e. cloud services s i In a virtual machine vm j Above, its start execution time ES(s) i ，vm j ) Calculating the time st i，j Virtual machine vm j Time delay sd (vm) of j ) Whether the sum of (A) and (B) is less than or equal to the cloud service s i Ar(s) of arrival i ) And response time rt(s) i ) And (4) summing. If not, the virtual machine cannot be used as the cloud service s i And (5) candidate virtual machines. For conditions to be met, the present method will perform the fourth step.
Fourth, calculate the virtual using equation 4Real time RT(s) that the machine has performed i ，vm j ) Computing cloud service request s using equation 5 i Reliability of (2)
Judging cloud services s i In virtual machine vm j Whether upper reliability is less than or equal to virtual machine vm j Reliability threshold prTr (vm) j ) As shown in the following formula:
if not, then the cloud service request s does not need to be requested i A backup virtual machine is provided. If yes, entering the fifth step to request s for cloud service i And searching the optimal backup virtual machine except the virtual machine.
Fifthly, for each of the other virtual machines, if equation (6) is satisfied, the cloud service s is calculated i In a virtual machine vm j Cost above:
c(s i ，vm j )＝st i，j ×cc(vm j )+ss(s i )×sc(vm j ) (8)
finding virtual machines vm with minimal service cost k The virtual machine is taken as a cloud service request s i The backup virtual machine of (1).
Sixthly, with the cloud service cost method of the formula (7) as a core, for the primary virtual machine or the cloud service with the primary/backup virtual machine, the service cost c(s) is performed by using the following formula i ) And (3) calculating:
seventh, for each job and each virtual machine, find has minimum cost of service c(s) i ) And assigns the job to the corresponding primary virtual machine or primary/backup virtual machine. Updating virtual machine parameters, such as virtual machine availability time avail (vm) j )，And removes the job from the queue waiting queue. Returning to the first step to continue the method.
Fig. 3 is a graph of the comparison result of the application of the embodiment of the present patent, wherein the number of cloud service jobs is from 28 to 52, and 4 in each step. As can be seen from fig. 3 (a), the average cost ratio of the RCJS scheduling method proposed by this patent is about 66.7% better than that of OPVMP, and about 34.9% better than that of M1N-M1N, respectively. For the reliability in FIG. 3 (b), both RCJS and OPVMP have very high reliability, both of which are significantly better than the MIN-MIN algorithm. From the rejection rate in FIG. 3 (c), it can be seen that RCJS is about 87.4% better than OPVMP method, and MIN-MIN is about 77.9% better. Therefore, the RCJS scheduling method provided by the patent has good balance of cloud computing system performance, and is very suitable for high-reliability and low-cost cloud service.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and improvements can be made without departing from the technical principle of the present invention, and these modifications and improvements should also be considered as being within the protection scope of the present invention.
Those not described in detail in the specification are well within the skill of the art.
Claims (8)
1. The cloud computing system service cost and reliability driven job scheduling method is characterized by comprising the following steps:
1) Establishing a cloud computing system, wherein the cloud computing system is composed of a user module, an operation queue module, a resource management module, an operation scheduling module, a reliability analysis module, a physical resource layer module and a virtual machine; initializing the virtual machine, and receiving jobs submitted by a cloud computing system user to a job queue;
2) The job scheduling method inquires whether the job waiting queue has the job to be scheduled or not, and if not, the job scheduling algorithm is ended; if yes, each job in the job waiting queue executes step 3), step 4), step 5), step 6), step 7);
3) Calculating the response time satisfaction condition of each job, and cloud service s i In virtual machine vm j At the start of execution time ES(s) i ，vm j ) Calculating the time st i，j Vm of virtual machine j Time delay sd (vm) of j ) Whether the sum of (A) and (B) is less than or equal to the cloud service s i Ar(s) of i ) And a response time rt(s) i ) Summing; if not, the virtual machine cannot be used as the cloud service s i A candidate virtual machine; for the conditions satisfied, the method proposed by this patent will perform step 4);
4) Calculating real time RT(s) that the virtual machine has executed i ，vm j ) Computing cloud service requests s i Reliability of (2)
Judging cloud services s i In a virtual machine vm j Whether upper reliability is less than or equal to virtual machine vm j Reliability threshold prTr (vm) j ) As shown in the following formula: PR si，vm j ≤prTr(vm j ) If not, then the cloud service request s is not required i Providing a backup virtual machine, directly jumping to the step 6), and if the backup virtual machine is met, entering the step 5) for requesting s) for cloud service i Searching for an optimal backup virtual machine except the virtual machine;
5) And for each other virtual machine, if the response time of the operation in the step 3) is met, calculating a cloud service operation s i In a virtual machine vm j Cost above, find virtual machine vm with minimum service cost k The virtual machine is taken as a cloud service request s i The backup virtual machine of (1);
6) Taking the cloud service cost calculation method in the step 5) as a core, and performing service formation calculation on the main virtual machine or the cloud service with the main/backup virtual machine;
7) For each job and each virtual machine, find the minimum cost of service c(s) i ) And assigning the job to the corresponding primary virtual machine or primary/backup virtual machine; updating virtual machine parameters, such as virtual machine availability time avail (vm) j ) And deleting the job from the queue waiting queue; returning to the step 2) to continue the methodA method is provided.
2. The cloud computing system service cost and reliability driven job scheduling method of claim 1, wherein the virtual machine is composed of a set of physical servers PS = { PS = } 1 ，ps 2 ，…，ps A And the device comprises a memory pt and a communication network system pn.
3. The cloud computing system service cost and reliability driven job scheduling method of claim 1, wherein the service request is described as a job set S having mutual independence, and the cloud service job S is calculated according to the following formula i Where the computing time of the virtual machine
Wherein s is i Representing a cloud service job request, w(s) i ) Representing cloud services s i Calculated amount of (c), w (vm) j ) Representing virtual machines vm j The computing processing power of (1).
4. The cloud computing system service cost and reliability driven job scheduling method of claim 2, wherein the physical resource failure rate of the set of physical servers obeys a poisson distribution, and the probability density function of a physical resource is calculated according to the following formula
Calculating a reliability function of the physical resource according to the following formula
Wherein ps a Denotes a physical server, λ pt Presentation memory, λ pn Representing a network.
5. The cloud computing system service cost of claim 1The job scheduling method driven by reliability is characterized in that in the step 3), the job is scheduled to the virtual machine vm j The cloud service job s processed i Calculating a start execution time ES(s) of the cloud service job according to the following formula i ，vm j )：ES(s i ，vm j )＝Max{ar(s i )，avail(vm j ) Therein avail (vm) j ) Representing virtual machines vm j The available time of (c); the response time for each job is calculated according to the following formula:
ES(s i ，vm j )+st i，j +sd(vm j )≤ar(s i )+rt(s i )。
6. the cloud computing system service cost and reliability driven job scheduling method according to claim 1, wherein in the step 4), the real-time that the virtual machine has executed is calculated as RT(s) according to the following formula i ，vm j )＝ES(s i ，vm j )+st i，j +runTime(vm j ) Said virtual machine vm j Processing cloud service requests s i Reliability of (2)
For the product of all physical resource reliabilities, the reliability is calculated according to the following formula
7. The cloud computing system service cost and reliability driven job scheduling method according to claim 1, wherein in the step 5), the cloud service job request s is calculated according to the following formula i In virtual machine vm j Cost c(s) of i ，vm j )：c(s i ，vm j )＝st i，j ×cc(vm j )+ss(s i )×sc(vm j )。
8. The cloud computing system service cost and reliability driven job scheduling method according to claim 1, wherein in the step 6), the service cost c(s) is calculated according to the following formula i )：
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910826099.3A CN110515720B (en) | 2019-09-04 | 2019-09-04 | Cloud computing system service cost and reliability driven job scheduling method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910826099.3A CN110515720B (en) | 2019-09-04 | 2019-09-04 | Cloud computing system service cost and reliability driven job scheduling method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110515720A CN110515720A (en) | 2019-11-29 |
| CN110515720B true CN110515720B (en) | 2022-11-22 |
Family
ID=68629560
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910826099.3A Active CN110515720B (en) | 2019-09-04 | 2019-09-04 | Cloud computing system service cost and reliability driven job scheduling method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110515720B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111405072B (en) * | 2020-06-03 | 2021-04-02 | 杭州朗澈科技有限公司 | Hybrid cloud optimization method based on cloud manufacturer cost scheduling |
| CN111882134A (en) * | 2020-08-04 | 2020-11-03 | 山东大学 | Cloud computing service scheduling method, system, medium and electronic device |
| CN113422740B (en) * | 2021-06-22 | 2022-12-06 | 广东电网有限责任公司计量中心 | Packet scheduling method, device, terminal and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8352941B1 (en) * | 2009-06-29 | 2013-01-08 | Emc Corporation | Scalable and secure high-level storage access for cloud computing platforms |
| WO2013174178A1 (en) * | 2012-05-25 | 2013-11-28 | 中兴通讯股份有限公司 | Method, apparatus and system for managing virtual network service |
| CN105389206A (en) * | 2015-11-02 | 2016-03-09 | 广东石油化工学院 | Method for rapidly configuring virtual machine resources in cloud computing data center |
| CN109254836A (en) * | 2018-07-23 | 2019-01-22 | 湖南农业大学 | Time limit committed cost Optimization Scheduling towards the preferential dependence task of cloud computing system |
-
2019
- 2019-09-04 CN CN201910826099.3A patent/CN110515720B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8352941B1 (en) * | 2009-06-29 | 2013-01-08 | Emc Corporation | Scalable and secure high-level storage access for cloud computing platforms |
| WO2013174178A1 (en) * | 2012-05-25 | 2013-11-28 | 中兴通讯股份有限公司 | Method, apparatus and system for managing virtual network service |
| CN105389206A (en) * | 2015-11-02 | 2016-03-09 | 广东石油化工学院 | Method for rapidly configuring virtual machine resources in cloud computing data center |
| CN109254836A (en) * | 2018-07-23 | 2019-01-22 | 湖南农业大学 | Time limit committed cost Optimization Scheduling towards the preferential dependence task of cloud computing system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110515720A (en) | 2019-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110515720B (en) | Cloud computing system service cost and reliability driven job scheduling method | |
| Zhou et al. | On cloud service reliability enhancement with optimal resource usage | |
| US8566835B2 (en) | Dynamically resizing a virtual machine container | |
| US20120066020A1 (en) | Multi-tenant database management for sla profit maximization | |
| US9940162B2 (en) | Realtime optimization of compute infrastructure in a virtualized environment | |
| Xie et al. | Quantitative fault-tolerance for reliable workflows on heterogeneous IaaS clouds | |
| US7953729B2 (en) | Resource optimizations in computing utilities | |
| CN108021435B (en) | Cloud computing task flow scheduling method with fault tolerance capability based on deadline | |
| US20200151025A1 (en) | Job scheduling based on job execution history | |
| CN107547595B (en) | Cloud resource scheduling system, method and device | |
| EP3125468B1 (en) | Cloud application processing method and application deployment method and relevant apparatus and system | |
| CN107464151B (en) | Order data processing method and device for high-concurrency service | |
| Khazaei et al. | Availability analysis of cloud computing centers | |
| WO2014194869A1 (en) | Request processing method, device and system | |
| Ashouraei et al. | A new SLA-aware load balancing method in the cloud using an improved parallel task scheduling algorithm | |
| Zhu et al. | Scaling service-oriented applications into geo-distributed clouds | |
| CN106257893A (en) | Storage server task response method, client, server and system | |
| US10871988B1 (en) | Methods for feedback-based optimal workload scheduling and devices thereof | |
| Amoon | A framework for providing a hybrid fault tolerance in cloud computing | |
| WO2013082742A1 (en) | Resource scheduling method, device and system | |
| Lawrance et al. | Efficient QoS based resource scheduling using PAPRIKA method for cloud computing | |
| CN104735134B (en) | A kind of method and apparatus serviced for providing calculating | |
| Wei et al. | AIMING: resource allocation with latency awareness for federated-cloud applications | |
| Niyato | Optimization-based virtual machine manager for private cloud computing | |
| JP6098167B2 (en) | Virtual machine management program and method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |