CN114640598B - Container placement method based on WOA algorithm in multi-tenant environment - Google Patents
Container placement method based on WOA algorithm in multi-tenant environment Download PDFInfo
- Publication number
- CN114640598B CN114640598B CN202210263975.8A CN202210263975A CN114640598B CN 114640598 B CN114640598 B CN 114640598B CN 202210263975 A CN202210263975 A CN 202210263975A CN 114640598 B CN114640598 B CN 114640598B
- Authority
- CN
- China
- Prior art keywords
- tenant
- container
- task
- cost
- representing
- 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
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000013507 mapping Methods 0.000 claims abstract description 16
- 238000013508 migration Methods 0.000 claims description 36
- 230000005012 migration Effects 0.000 claims description 36
- 230000006870 function Effects 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 6
- 239000004065 semiconductor Substances 0.000 claims description 3
- 230000008569 process Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5003—Managing SLA; Interaction between SLA and QoS
Abstract
The invention belongs to the field of cloud computing, and particularly relates to a container placement method based on a WOA algorithm in a multi-tenant environment, which comprises the following steps: acquiring tenant information, task information, container resource information and node resource information; defining tenant grades and constructing a tenant QoS model; constructing a constraint minimum cost model, and calculating the execution cost of each task in various containers; calculating the execution cost of each task of each tenant by adopting a greedy algorithm, and collecting all containers with the minimum cost; calculating an optimal mapping relation between the container and the node by adopting a WOA algorithm, and taking the calculated mapping relation as an optimal container placement method; according to the invention, from the perspective of tenants, the optimal mapping relation between the container and the nodes is calculated by using the WOA algorithm, and the mapping relation is used as an optimal container placement scheme, so that good resource utilization rate can be obtained while the QoS of the tenants is ensured, and the SLA default rate and the cloud service center operation cost are reduced.
Description
Technical Field
The invention belongs to the field of cloud computing, and particularly relates to a container placement method based on a WOA algorithm in a multi-tenant environment.
Background
Cloud computing is a computing mode for realizing on-demand distribution and convenient access to a shared resource pool by utilizing the Internet, and provides a basic support for the development of the fields of big data, artificial intelligence, the Internet of things and the like. At present, the domestic cloud computing market is in a high-speed growth stage, and the cloud prospect of small and medium enterprises in China is 68.6%, wherein 59.3% of small and medium enterprises have higher acceptance on cloud services, the proportion is gradually increasing along with the development of enterprises and cloud service providers, and the container placement in a multi-tenant environment is particularly important along with the increasing number of tenants of the cloud service providers.
In a multi-tenant environment, a cloud service center needs to improve the resource utilization rate while meeting the QoS of all tenants as much as possible, and the purpose of meeting the QoS of the tenants is to reduce the SLA default rate, reduce the default cost, and improve the resource utilization rate and reduce the resource idle cost. In the existing container placement research, most researchers aim at the maximum utilization rate of resources of a cloud service center, and usually only pay attention to the optimal allocation of single-class resource dimensions, multi-dimensional resources are rarely considered, in addition, most researchers usually build models only from the task or container angles, and rarely from the tenant angles, while in the multi-tenant scenario, the tenant is a paid entity, and the tenant should be considered in addition to the resource factors. Therefore, in the multi-tenant environment, it is expected that not only the resource utilization rate can be improved, but also the operation cost of the cloud service center can be reduced on the premise of meeting the QoS of the tenant.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a container placement method based on a WOA algorithm in a multi-tenant environment, which comprises the following steps:
s1: acquiring tenant information, task information, container resource information and node resource information; defining tenant grades according to tenant information, and constructing a tenant QoS model;
s2: constructing a constraint minimum cost model, and calculating the execution cost of each task in various containers according to the constraint minimum cost model;
s3: calculating the execution cost of each task of each tenant by adopting a greedy algorithm, selecting the minimum container according to the execution cost, and collecting all containers with the minimum execution cost to obtain a container set to be placed;
s4: and calculating the optimal mapping relation between the container and the node by adopting a WOA algorithm according to the container set to be placed, and taking the calculated mapping relation as an optimal container placing method.
Preferably, the obtained tenant information comprises a task set, task expected cost and task expected completion time of the tenant; the task information comprises a task length, a task size and a affiliated tenant of each task; the container resource information comprises the computing capacity, the memory size and the bandwidth size of the container; the node resource information includes computing power, memory size, bandwidth size of the node, and the assigned container set.
Preferably, defining the tenant level includes:
wherein ,Lr Representing Tenant r Tenant represents the Tenant set on the container cloud platform, expCOST r Tenant for Tenant r ExpTime, an expense expected on a container cloud platform r Tenant for Tenant r Desired task completion time, expco s Representing Tenant s ExpTime, an expense expected on a container cloud platform s Representing Tenant s The desired task completion time.
Preferably, the process of constructing the tenant QoS model includes:
step 1: normalizing the tenant task execution time to obtain normalized tenant task execution time
Step 2: normalizing the tenant task execution cost to obtain normalized tenant task execution cost
Step 3: establishing a tenant QoS model according to the normalized tenant task execution time and tenant task execution cost, wherein the expression of the model is as follows:
wherein ,Lr Representing Tenant r Is a class of the tenant of the (a) in the (c),representing Tenant r Normalized task execution time,/->Representing Tenant r Normalized task execution cost, α is the preference of the tenant.
Preferably, the constrained minimum cost model is constructed as follows:
wherein VCost represents the total cost of container deployment, V free Representing the idle cost of resources, V sla Representing the cost of SLA violations to the tenant,respectively represent node N k CPU unit price, memory unit price, bandwidth unit price, R k,cpu 、R k,ram 、R k,bw Respectively represent node N k Idle CPU resource, memory resource, bandwidth resource, TTET r Representing Tenant r ExpTime is performed by the task execution time of (1) r Representing Tenant r Desired task execution time, TCOST r Representing Tenant r Task execution cost of a) r The value of 0 or 1 indicates that the cloud platform is not specific to the Tenant when the value of 0 is 0 r When the value of the default is 1, the default is generated for the tenant; constraint C2, C3 and C4 are resource capacity constraints and represent node N k The resources allocated above cannot exceed the resource capacity thereof; constraint C5 represents the QoS constraint of the tenant.
Preferably, calculating the execution cost of each task in each type of container according to the constraint minimum cost model includes:
Cost i，j ＝(CPU_Cost j +RAM_Cost j +BW_Cost j )*TET i，j
wherein, cost i,j Representing Task i Dispensing to container C j In (3) execution Cost, CPU_cost j Representing container C j RAM_cost j Representing container C j BW_cost j Representing container C j Bandwidth cost, TET i,j Representing when Task i Dispensing to container C j In execution time, TET i,j Representing the execution time of task i on container j, length i For Task i Length of Size of (2) i For Task i MIPS (metal-oxide-semiconductor package) size j Representing container C j Is (are) BW j Representing container C j Is not limited to the bandwidth of the (c).
Preferably, calculating the optimal mapping relation between the container and the node by using the WOA algorithm includes:
s41: initializing parameters including iteration number, population number, upper and lower boundaries of a solution space and fitness function;
s42: calculating the fitness function value of each element in the container set to be placed by adopting the fitness function, and calculating an initial solution according to the fitness function value; judging whether each node meets each resource capacity constraint of the cost model, if so, executing a step S43, otherwise, executing a step S44;
s43: updating the fitness function value by using a WOA algorithm, and updating the solution according to the updated fitness function value;
s44: migrating nodes which do not meet all resource capacity constraints of the cost model by using a greedy migration strategy until all nodes meet constraint conditions;
s45: and repeating the steps S43-S44 until the algorithm iteration is completed, and taking the solution corresponding to the minimum fitness value as the mapping relation between the optimal container and the node.
Further, the process of migrating nodes which do not meet all the resource capacity constraints of the cost model by using a greedy migration strategy comprises the following steps:
s441: traversing all the currently distributed nodes, and screening and storing all the nodes exceeding the resource constraint as an migrating node set N out The remaining nodes are taken as an ingress node set N in ；
S442: for the migration node set N out Each node traverses all containers on the node, calculates the migration distance, screens out the containers according to the minimum migration distance until all the nodes meet the constraint of the resource capacity, and obtains a container set C to be allocated mig ；
S443: traversing an ingress node set N in For each migrating node, traversing the set of containers C to be allocated mig Calculating migration distance, screening out a proper container according to the minimum migration distance under the condition of meeting the constraint of resource capacity, and migrating the container to the current node until a container set C to be distributed mig Is empty.
Further, the migration distance calculation formula and the migration distance calculation formula are as follows:
wherein ,representing the migration distance, < > of container j on node k>Representing the migration distance of container j onto node k, +.>Respectively represent the allocated computing power and memory on node kBandwidth, MIPS j 、RAM j 、BW j Representing the computing power, memory, bandwidth, MIPS of container j, respectively k 、RAM k 、BW k Respectively representing the computing power, the memory and the bandwidth of the node k.
The invention has the beneficial effects that:
according to the container placement method based on the WOA algorithm in the multi-tenant environment, the migration distance and the migration distance calculation mode are provided, multi-dimensional resource constraint is considered, good resource utilization rate can be obtained, and the operation cost of the cloud service center is reduced.
Drawings
FIG. 1 is a general flow chart of container placement in an embodiment of the present invention;
FIG. 2 is a flowchart of calculating an optimal mapping relationship between a container and a node using WOA algorithm according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
An embodiment of a container placement method based on a WOA algorithm in a multi-tenant environment is shown in fig. 1, and the method includes:
s1: acquiring tenant information, task information, container resource information and node resource information; defining tenant grades according to tenant information, and constructing a tenant QoS model;
s2: constructing a constraint minimum cost model, and calculating the execution cost of each task in various containers according to the constraint minimum cost model;
s3: calculating the execution cost of each task of each tenant by adopting a greedy algorithm, selecting the minimum container according to the execution cost, and collecting all containers with the minimum execution cost to obtain a container set to be placed;
s4: and calculating the optimal mapping relation between the container and the node by adopting a WOA algorithm according to the container set to be placed, and taking the calculated mapping relation as an optimal container placing method.
The tenant information comprises a task set of the tenant, task expected cost and task expected completion time, the task information comprises task length, task size and affiliated tenant of each task, the container resource information comprises computing capacity, memory size and bandwidth size of a container, and the node resource information comprises computing capacity, memory size and bandwidth size of a node and the allocated container set.
Defining tenant levels includes:
wherein ,Lr Representing Tenant r Tenant represents the Tenant set on the container cloud platform, expCOST r Tenant for Tenant r ExpTime, an expense expected on a container cloud platform r Tenant for Tenant r Desired task completion time, expco s Representing Tenant s ExpTime, an expense expected on a container cloud platform s Representing Tenant s The desired task completion time.
The process of constructing the tenant QoS model according to the tenant level comprises the following steps:
step 1: normalizing the tenant task execution time to obtain normalized tenant task execution timeThe expression is as follows:
wherein TTET is r Representing Tenant r ExpTime is performed by the task execution time of (1) r Representing Tenant r Desired task execution time.
Step 2: normalizing the tenant task execution cost to obtain normalized tenant task execution costThe expression is as follows:
ExpCost r tenant for Tenant r Expected spending on container cloud platform, TCost r Representing Tenant r Is a task execution cost of (a).
Step 3: establishing a tenant QoS model according to the normalized tenant task execution time and tenant task execution cost, wherein the expression of the model is as follows:
wherein ,Lr Representing Tenant r Is a class of the tenant of the (a) in the (c),representing Tenant r Normalized task execution time,/->Representing Tenant r Normalized task execution cost, alpha is the preference of the tenant, and the larger the value of alpha is, the more concerned the tenant is about the execution time of the task, and the more concerned the execution cost is otherwise.
And constructing a multi-constraint minimum cost model according to the collected information, and calculating the execution cost of each task in various containers.
The built multi-constraint minimum cost model is as follows:
wherein VCost represents the total cost of container deployment, V free Representing the idle cost of resources, V sla Representing the cost of SLA violations to the tenant,respectively represent node N k CPU unit price, memory unit price, bandwidth unit price, R k,cpu 、R k,ram 、R k,bw Respectively represent node N k Idle CPU resource, memory resource, bandwidth resource, TTET r Representing Tenant r ExpTime is performed by the task execution time of (1) r Representing Tenant r Desired task execution time, TCOST r Representing Tenant r Task execution cost of a) r The value of 0 or 1 indicates that the cloud platform is not specific to the Tenant when the value of 0 is 0 r When the value of the default is 1, the default is generated for the tenant; constraint C2, C3 and C4 are resource capacity constraints and represent node N k The resources allocated above cannot exceed the resource capacity thereof; constraint C5 represents the QoS constraint of the tenant, and when its value is less than 1, SLA breach costs are incurred.
Calculating the execution cost of each task in various containers according to the constraint minimum cost model comprises:
Cost i，j ＝(CPU_Cost j +RAM_Cost j +BW_Cost j )*TET i，j
wherein, cost i , j Representing Task i Dispensing to container C j In (3) execution Cost, CPU_cost j Representing container C j RAM_cost j Representing container C j BW_cost j Representing container C j Bandwidth cost, TET i , j Representing when Task i Dispensing to container C j Is performed in the same manner as the execution time of the program.
Task execution time TET i , j The calculation formula of (2) is as follows:
wherein ,TETi , j Representing the execution time of task i on container j, length i For Task i Length of Size of (2) i For Task i MIPS (metal-oxide-semiconductor package) size j Representing container C j Is (are) BW j Representing container C j Is not limited to the bandwidth of the (c).
The process for obtaining the container set to be placed by using the greedy algorithm comprises the following steps: and (3) traversing the execution cost of each task under different containers according to the execution cost of each task calculated in the step (S2), selecting a container with the minimum execution cost, distributing the task under the container, and placing the container into a container set to be placed.
In a specific embodiment of a container placement method based on a WOA algorithm in a multi-tenant environment, an optimal mapping relationship between a container and a node is calculated by using the WOA algorithm as shown in fig. 2, and the method includes the following steps:
s41, initializing WOA algorithm parameters including iteration number, population number, upper and lower boundaries of a solution space and fitness function, and inputting a container set to be placed;
s42, initializing an adaptability value to obtain an initial solution, judging whether each node meets each resource capacity constraint of the cost model, if so, continuing the step, otherwise, jumping to the step S44;
s43, updating the fitness function value according to the WOA algorithm to obtain an updated solution;
s44, migrating the nodes which do not meet the constraint of each resource capacity of the cost model by using a greedy migration strategy until all the nodes meet the constraint condition;
s45: and repeating the steps S43-S44 until the algorithm iteration is completed, and taking the solution corresponding to the minimum fitness value as the mapping relation between the optimal container and the node.
An embodiment of a greedy migration strategy, the method comprising the steps of:
s441, traversing all the currently allocated nodes, and screening out all nodes exceeding the resource constraint as an outgoing node set N out The remaining nodes are taken as an ingress node set N in ；
S442, for the migrating node set N out Each node traverses all containers on the node, calculates the migration distance, screens out the containers according to the minimum migration distance until all the nodes meet the constraint of the resource capacity, and obtains a container set C to be allocated mig ；
The calculation mode of the migration distance is as follows:
wherein ,representing the migration distance, < > of container j on node k>Representing the migration distance of container j onto node k, +.>Respectively representing the allocated computing power, memory and bandwidth on node k, MIPS j 、RAM j 、BW j Representing the computing power, memory, bandwidth, MIPS of container j, respectively k 、RAM k 、BW k Respectively representing the computing power, the memory and the bandwidth of the node k.
S443, traversing the migration node set N in For each migrating node, traversing the set of containers C to be allocated mig Calculating migration distance, screening out a proper container according to the minimum migration distance under the condition of meeting the constraint of resource capacity, and migrating the container to the current node until a container set C to be distributed mig Is empty.
The migration distance is calculated in the following way:
wherein ,representing the migration distance, < > of container j on node k>Representing the migration distance of container j onto node k, +.>Respectively representing the allocated computing power, memory and bandwidth on node k, MIPS j 、RAM j 、BW j Representing the computing power, memory, bandwidth, MIPS of container j, respectively k 、RAM k 、BW k Respectively representing the computing power, the memory and the bandwidth of the node k.
While the foregoing is directed to embodiments, aspects and advantages of the present invention, other and further details of the invention may be had by the foregoing description, it will be understood that the foregoing embodiments are merely exemplary of the invention, and that any changes, substitutions, alterations, etc. which may be made herein without departing from the spirit and principles of the invention.
Claims (3)
1. A container placement method based on WOA algorithm in a multi-tenant environment is characterized by comprising the following steps:
s1: acquiring tenant information, task information, container resource information and node resource information; defining tenant grades according to tenant information, and constructing a tenant QoS model; defining tenant levels includes:
wherein ,Lr Representing Tenant r Tenant represents the Tenant set on the container cloud platform, expCOST r Tenant for Tenant r ExpTime, an expense expected on a container cloud platform r Tenant for Tenant r Desired task completion time, expco s Representing Tenant s ExpTime, an expense expected on a container cloud platform s Representing Tenant s A desired task completion time;
building tenant QoS models includes:
step 1: normalizing the tenant task execution time to obtain normalized tenant task execution time
Step 2: normalizing the tenant task execution cost to obtain normalized tenant task execution cost
Step 3: establishing a tenant QoS model according to the normalized tenant task execution time and tenant task execution cost, wherein the expression of the model is as follows:
wherein ,Lr Representing Tenant r Is a class of the tenant of the (a) in the (c),representing Tenant r The normalized execution time of the task is calculated,representing Tenant r The normalized cost of execution of the task,alpha is the preference of the tenant;
s2: constructing a constraint minimum cost model, and calculating the execution cost of each task in various containers according to the constraint minimum cost model; the constrained minimum cost model is:
wherein VCost represents the total cost of container deployment, V free Representing the idle cost of resources, V sla Representing the cost of SLA violations to the tenant,respectively represent node N k CPU unit price, memory unit price, bandwidth unit price, R k,cpu 、R k,ram 、R k,bw Respectively represent node N k Idle CPU resource, memory resource, bandwidth resource, TTET r Representing Tenant r ExpTime is performed by the task execution time of (1) r Representing Tenant r Desired task execution time, TCOST r Representing Tenant r Task execution cost of a) r The value of 0 or 1 indicates that the cloud platform is not specific to the Tenant when the value of 0 is 0 r When the value of the default is 1, the default is generated for the tenant; constraint C2, C3 and C4 are resource capacity constraints and represent node N k The resources allocated above cannot exceed the resource capacity thereof; constraint C5 represents a QoS constraint of the tenant;
calculating the execution cost of each task in various containers according to the constraint minimum cost model comprises:
Cost i,j ＝(CPU_Cost j +RAM_Cost j +BW_Cost j )*TET i,j
wherein, cost i,j Representing Task i Dispensing to container C j In (3) execution Cost, CPU_cost j Representing container C j RAM_cost j Representing container C j BW_cost j Representing container C j Bandwidth cost, TET i,j Representing when Task i Dispensing to container C j In execution time, TET i,j Representing the execution time of task i on container j, length i For Task i Length of Size of (2) i For Task i MIPS (metal-oxide-semiconductor package) size j Representing container C j Is (are) BW j Representing container C j Is a bandwidth of (a);
s3: calculating the execution cost of each task of each tenant by adopting a greedy algorithm, selecting the minimum container according to the execution cost, and collecting all containers with the minimum execution cost to obtain a container set to be placed;
s4: calculating an optimal mapping relation between the container and the node by adopting a WOA algorithm according to the container set to be placed, and taking the calculated mapping relation as an optimal container placing method; the method specifically comprises the following steps:
s41: initializing parameters including iteration number, population number, upper and lower boundaries of a solution space and fitness function;
s42: calculating the fitness function value of each element in the container set to be placed by adopting the fitness function, and calculating an initial solution according to the fitness function value; judging whether each node meets each resource capacity constraint of the cost model, if so, executing a step S43, otherwise, executing a step S44;
s43: updating the fitness function value by using a WOA algorithm, and updating the solution according to the updated fitness function value;
s44: migrating nodes which do not meet all resource capacity constraints of the cost model by using a greedy migration strategy until all nodes meet constraint conditions; the method specifically comprises the following steps:
s441: traversing all the currently distributed nodes, and screening and storing all the nodes exceeding the resource constraint as an migrating node set N out The remaining nodes are taken as an ingress node set N in ；
S442: for the migration node set N out Each node traverses all containers on the node, calculates the migration distance, screens out the containers according to the minimum migration distance until all the nodes meet the constraint of the resource capacity, and obtains a container set C to be allocated mig ；
S443: traversing an ingress node set N in For each migrating node, traversing the set of containers C to be allocated mig Calculating migration distance, screening out a proper container according to the minimum migration distance under the condition of meeting the constraint of resource capacity, and migrating the container to the current node until a container set C to be distributed mig Is empty;
s45: and repeating the steps S43-S44 until the algorithm iteration is completed, and taking the solution corresponding to the minimum fitness value as the mapping relation between the optimal container and the node.
2. The method for placing a container based on a WOA algorithm in a multi-tenant environment according to claim 1, wherein the obtained tenant information includes a task set of the tenant, a task expected cost, and a task expected completion time; the task information comprises a task length, a task size and a affiliated tenant of each task; the container resource information comprises the computing capacity, the memory size and the bandwidth size of the container; the node resource information includes computing power, memory size, bandwidth size of the node, and the assigned container set.
3. The method for placing a container based on the WOA algorithm in a multi-tenant environment according to claim 1, wherein the migration distance calculation formula and the migration distance calculation formula are:
wherein ,representing the migration distance, < > of container j on node k>Representing the migration distance of container j onto node k, +.>Respectively representing the allocated computing power, memory and bandwidth on node k, MIPS j 、RAM j 、BW j Representing the computing power, memory, bandwidth, MIPS of container j, respectively k 、RAM k 、BW k Respectively representing the computing power, the memory and the bandwidth of the node k.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210263975.8A CN114640598B (en) | 2022-03-17 | 2022-03-17 | Container placement method based on WOA algorithm in multi-tenant environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210263975.8A CN114640598B (en) | 2022-03-17 | 2022-03-17 | Container placement method based on WOA algorithm in multi-tenant environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114640598A CN114640598A (en) | 2022-06-17 |
| CN114640598B true CN114640598B (en) | 2023-09-29 |
Family
ID=81950124
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210263975.8A Active CN114640598B (en) | 2022-03-17 | 2022-03-17 | Container placement method based on WOA algorithm in multi-tenant environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114640598B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115685875B (en) * | 2022-10-21 | 2023-08-01 | 广东工业大学 | Method and system for optimizing cutting path of irregular plate |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101783768A (en) * | 2010-03-08 | 2010-07-21 | 东南大学 | Quantity assurance method of grid service based on resource reservation |
| CN111124689A (en) * | 2019-12-31 | 2020-05-08 | 中国电子科技集团公司信息科学研究院 | Dynamic allocation method for container resources in cluster |
| CN112559135A (en) * | 2020-12-24 | 2021-03-26 | 重庆邮电大学 | QoS-based container cloud resource scheduling method |
| US11069082B1 (en) * | 2015-08-23 | 2021-07-20 | AI Incorporated | Remote distance estimation system and method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9183560B2 (en) * | 2010-05-28 | 2015-11-10 | Daniel H. Abelow | Reality alternate |
-
2022
- 2022-03-17 CN CN202210263975.8A patent/CN114640598B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101783768A (en) * | 2010-03-08 | 2010-07-21 | 东南大学 | Quantity assurance method of grid service based on resource reservation |
| US11069082B1 (en) * | 2015-08-23 | 2021-07-20 | AI Incorporated | Remote distance estimation system and method |
| CN111124689A (en) * | 2019-12-31 | 2020-05-08 | 中国电子科技集团公司信息科学研究院 | Dynamic allocation method for container resources in cluster |
| CN112559135A (en) * | 2020-12-24 | 2021-03-26 | 重庆邮电大学 | QoS-based container cloud resource scheduling method |
Non-Patent Citations (2)
| Title |
|---|
| Towards a container-based architecture for multi-tenant SaaS applications;Eddy Truyen;《ARM 2016: Proceedings of the 15th International Workshop on Adaptive and Reflective Middleware》;全文 * |
| 面向容器云平台的多目标资源调度方法的研究;张晓天;《硕士电子期刊》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114640598A (en) | 2022-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111953758B (en) | Edge network computing unloading and task migration method and device | |
| US9311159B2 (en) | Systems, methods, and articles of manufacture to provide cloud resource orchestration | |
| CN106095569B (en) | A kind of cloud workflow engine scheduling of resource and control method based on SLA | |
| CN103888371B (en) | A kind of method and apparatus of virtual network mapping | |
| CN107656799B (en) | Workflow scheduling method considering communication and computing cost under multi-cloud environment | |
| WO2021056787A1 (en) | Hybrid cloud service process scheduling method | |
| CN111614754B (en) | Fog-calculation-oriented cost-efficiency optimized dynamic self-adaptive task scheduling method | |
| CN114640598B (en) | Container placement method based on WOA algorithm in multi-tenant environment | |
| CN109901921A (en) | Task queue running time prediction method, apparatus and realization device | |
| CN109361750B (en) | Resource allocation method, device, electronic equipment and storage medium | |
| CN113254192B (en) | Resource allocation method, resource allocation device, electronic device and storage medium | |
| CN113163498B (en) | Virtual network resource allocation method and device based on genetic algorithm under 5G network slice | |
| CN108989122A (en) | Virtual network requests mapping method, device and realization device | |
| CN108108245A (en) | The mixed type dispatching method and system of a kind of cloud platform wide node scientific workflow | |
| CN107979545A (en) | A kind of mapping method of virtual network based on nodal community | |
| Tran et al. | A survey of autoscaling in kubernetes | |
| CN110167031B (en) | Resource allocation method, equipment and storage medium for centralized base station | |
| CN112165721B (en) | Multi-service task unloading and service migration method based on edge computing | |
| CN108664318A (en) | Computation migration method and server-side, boundary server-side for computation migration | |
| WO2023134403A1 (en) | Internet of things resource allocation method and system, terminal and storage medium | |
| CN108073449B (en) | Dynamic virtual machine placement method | |
| CN106357800A (en) | Cloud computing service architecture based on QoE | |
| WO2019169998A1 (en) | Method, system, and related apparatus for selecting data node | |
| CN116069473A (en) | Deep reinforcement learning-based Yarn cluster workflow scheduling method | |
| CN116009990A (en) | Cloud edge collaborative element reinforcement learning computing unloading method based on wide attention mechanism |
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 |