CN107948330A - Load balancing based on dynamic priority under a kind of cloud environment - Google Patents
Load balancing based on dynamic priority under a kind of cloud environment Download PDFInfo
- Publication number
- CN107948330A CN107948330A CN201810008891.3A CN201810008891A CN107948330A CN 107948330 A CN107948330 A CN 107948330A CN 201810008891 A CN201810008891 A CN 201810008891A CN 107948330 A CN107948330 A CN 107948330A
- Authority
- CN
- China
- Prior art keywords
- service node
- priority
- time
- monitor
- dynamic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/101—Server selection for load balancing based on network conditions
Abstract
The present invention discloses the load balancing based on dynamic priority under a kind of cloud environment, comprises the following steps：Obtain each service node status information；Service node status information includes cpu busy percentage, space utilisation and the network utilization that the service node includes each cloud host；Load equalizer establishes service node priority query according to service node status information；Load equalizer dynamically adjusts the priority update cycle of each service node according to service node priority situation of change；Load equalizer adjusts service node priority query according to the update cycle of each service node.The present invention by consider the calculating of service node, storage, network real time load situation, the foundation for calculating its priority as the service that provides is more reasonable efficient.Meanwhile its update cycle is obtained according to the load change frequency of each node, the pressure of load equalizer is not only reduced, and ensure that the timeliness of load equalizer.
Description
Technical Field
The invention relates to a load balancing strategy, in particular to a load balancing strategy based on dynamic priority in a cloud environment.
Background
The cloud computing technology utilizes a virtualization technology to pool resources such as computing, storage, networks and the like, and provides shared software and hardware to users in a service mode through the internet. The cloud computing technology has a series of advantages of virtualization technology, parallel computing technology, grid computing, distributed computing technology and transparent computing technology. Resource scheduling of cloud computing is in multiple levels, and generally includes three levels, namely an application layer, a virtual layer and a physical layer, and objects and granularities scheduled by the layers are different. However, the scheduling of resources of any layer is to achieve effective and reasonable utilization of cloud computing resources. Load balancing is an important factor affecting the efficient use of resources. The unbalanced load can seriously affect the performance of the cloud computing system, cause system blockage and reduce the throughput rate and the resource utilization rate of the system.
Disclosure of Invention
In order to solve the above problems, the present invention provides a load balancing strategy based on dynamic priority in a cloud environment for optimizing load balancing.
The technical scheme of the invention is that a load balancing strategy based on dynamic priority in a cloud environment comprises the following steps:
acquiring state information of each service node; the service node state information comprises the CPU utilization rate, the storage utilization rate and the network utilization rate of each cloud host contained in the service node;
the load balancer establishes a service node priority queue according to the service node state information;
the load balancer dynamically adjusts the priority updating period of each service node according to the change condition of the priority of the service node;
and the load balancer adjusts the service node priority queue according to the update period of each service node.
Further, the load balancer establishes the service node priority queue according to the service node state information means that the load balancer calculates the dynamic priority score of the service node according to the following model, and the higher the dynamic priority score is, the higher the dynamic priority is:
wherein Host _ Score (i) represents a dynamic priority Score of the ith service node; CPU _ Avg (i), storage _ Avg (i) and NetWork _ Avg (i) respectively represent the average utilization rate of the CPU, storage and NetWork of the ith service node in the current monitoring period; and the Host _ CPU (j), the Host _ Storage (j) and the Host _ NetWork (j) respectively represent the CPU, storage and NetWork utilization rate of the jth cloud Host of the ith service node.
Further, the specific implementation method for dynamically adjusting the priority update period of each service node by the load balancer according to the change condition of the priority of the service node is as follows:
the load balancer calculates the priority deviation delta of the service node with a period default _ monitor _ time (i)
Wherein, the Host _ Score (i) is the dynamic priority Score of the ith service node, and the Host _ Score is the average value of the dynamic priority scores of the n service nodes;
defining the upper limit of the priority deviation of the ith service node as threshold (i), judging the size of the upper limit of the priority deviation of the ith service node as well as the priority deviation delta, and if delta is less than the threshold (i), the priority update period monitor (i) of the ith service node is as follows:
monitor_time(i)＝default_monitor_time(i) (1)
monitor_time(i)＝monitor_time(i)+atomic_time(i)*f(δ，threshold(i)) (2)
the formula (1) represents the initialization setting of the update period of the ith service node, and the formula (2) represents the dynamic adjustment of the update period of the ith service node;
atomic _ time (i) is an atomic time of an update cycle interval, and f (δ, threshold (i)) represents a corresponding functional relationship between the priority deviation δ and a priority deviation upper limit threshold (i);
if δ > = threshold (i), the priority update period monitor (i) of the ith serving node is:
monitor_time(i)＝max(monitor_time(i)/2,atomic_time(i))。
furthermore, the policy acquires the state information of each service node through the agent system.
Further, the policy further comprises the steps of:
the load balancer establishes a buffer queue for request servicing.
Further, the load balancer establishes a buffer queue for the request service by using a first-come first-serve strategy.
Further, the service node with the higher priority in the service node priority queue is used as the service node to be served requesting service.
According to the load balancing strategy based on the dynamic priority under the cloud environment, the priority is calculated to be used as the basis for providing the service more reasonably and efficiently by comprehensively considering the calculation, storage and real-time load conditions of the network of the service node. Meanwhile, the updating period is obtained according to the load change frequency of each node, so that the pressure of the load balancer is reduced, and the timeliness of the load balancer is guaranteed.
Drawings
Fig. 1 is a schematic diagram of an implementation process of a specific embodiment of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings by way of specific examples, which are illustrative of the present invention and are not limited to the following embodiments.
As shown in fig. 1, the core idea of the present invention is to deploy an intelligent agent to a service node of a platform through a cloud platform, and collect state information of the service node through the agent. And simultaneously, a buffer queue is established for the request service according to the first-come-first-serve strategy. And then the load balancer establishes a priority queue of the service node according to the state information collected by the agent, and the service node with high priority is defined as the service node to be served. And finally, the load balancer dynamically adjusts the priority updating period of each node according to the priority change condition of the node.
Specifically, the strategy comprises the following steps:
acquiring state information of each service node; the service node state information comprises the CPU utilization rate, the storage utilization rate and the network utilization rate of each cloud host contained in the service node;
the load balancer establishes a service node priority queue according to the service node state information;
the load balancer dynamically adjusts the priority updating period of each service node according to the change condition of the priority of the service node;
and the load balancer adjusts the priority queue of the service nodes according to the update period of each service node.
It should be noted that, the state information of each service node is obtained through the agent system deployed on the service node. And the load balancer also establishes a buffer queue for the request service according to the first-come first-serve strategy, and takes the service node with the higher priority in the service node priority queue as the service node to be served for the request service according to the task sequence in the buffer queue.
In this embodiment, the load balancer establishes the service node priority queue according to the service node state information, that is, the load balancer calculates a dynamic priority score of the service node according to the following model, where the higher the dynamic priority score is, the higher the dynamic priority is:
wherein Host _ Score (i) represents the dynamic priority Score of the ith service node; CPU _ Avg (i), storage _ Avg (i) and NetWork _ Avg (i) respectively represent the average utilization rate of the CPU, storage and NetWork of the ith service node in the current monitoring period; and the Host _ CPU (j), the Host _ Storage (j) and the Host _ NetWork (j) respectively represent the CPU, storage and NetWork utilization rate of the jth cloud Host of the ith service node. It should be noted that the cloud platform has its own monitoring period, and in the default monitoring period, the computing, storage and network data of each service node are collected.
The specific implementation method for dynamically adjusting the priority updating period of each service node by the load balancer according to the change condition of the priority of the service node comprises the following steps:
the load balancer calculates the priority deviation delta of the service node with a period default _ monitor _ time (i)
Where, host _ Score (i) is the dynamic priority Score of the ith service node, and Host _ Score is the average of the dynamic priority scores of the n service nodes.
Defining the upper limit of the priority deviation of the ith service node as threshold (i), judging the sizes of the upper limit of the priority deviation of the ith service node and the priority deviation delta, if delta is less than threshold (i), indicating that the variation trend of the acquired data is relatively stable, the priority update period monitor (i) of the ith service node is as follows:
monitor_time(i)＝default_monitor_time(i) (1)
monitor_time(i)＝monitor_time(i)+atomic_time(i)*f(δ，threshold(i)) (2)
wherein, the formula (1) represents the initialization setting of the update period of the ith service node, and the formula (2) represents the dynamic adjustment of the update period of the ith service node.
atomic _ time (i) is an atomic time of an update cycle interval, and f (δ, threshold (i)) represents a corresponding functional relationship between the priority deviation δ and the upper limit of the priority deviation threshold (i). f (delta, threshold (i)) is used to control the pace of the increase of the monitoring period, typically 2-fold and 4-fold increases. That is, the updated time interval can be adjusted only to an integer multiple of the atomic time atomic _ time (i).
If δ > = threshold (i), which indicates that the variation trend of the update period is large, the priority update period monitor (i) of the ith service node needs to be reduced as:
monitor_time(i)＝max(monitor_time(i)/2,atomic_time(i))。
the above strategy is reasonable and efficient by comprehensively considering the calculation, storage and real-time load conditions of the service nodes and calculating the priority as the basis for providing the service. Meanwhile, the updating period is obtained according to the load change frequency of each service node, so that the pressure of the load balancer is reduced, and the timeliness of the load balancer is guaranteed.
The above disclosure is only for the preferred embodiments of the present invention, but the present invention is not limited thereto, and any non-inventive changes that can be made by those skilled in the art and several modifications and amendments made without departing from the principle of the present invention shall fall within the protection scope of the present invention.
Claims (7)
1. A load balancing strategy based on dynamic priority in cloud environment is characterized in that,
the method comprises the following steps:
acquiring state information of each service node; the service node state information comprises the CPU utilization rate, the storage utilization rate and the network utilization rate of each cloud host contained in the service node;
the load balancer establishes a service node priority queue according to the service node state information;
the load balancer dynamically adjusts the priority updating period of each service node according to the change condition of the priority of the service node;
and the load balancer adjusts the service node priority queue according to the update period of each service node.
2. The load balancing strategy based on dynamic priority under the cloud environment according to claim 1, wherein the load balancer establishes the service node priority queue according to the service node status information, that is, the load balancer calculates the dynamic priority score of the service node according to the following model, and the higher the dynamic priority score is, the higher the dynamic priority is:
wherein Host _ Score (i) represents the dynamic priority Score of the ith service node;
CPU _ Avg (i), storage _ Avg (i) and NetWork _ Avg (i) respectively represent the average utilization rate of the CPU, storage and NetWork of the ith service node in the current monitoring period; and the Host _ CPU (j), the Host _ Storage (j) and the Host _ NetWork (j) respectively represent the CPU, storage and NetWork utilization rate of the jth cloud Host of the ith service node.
3. The load balancing strategy based on dynamic priority in cloud environment according to claim 2, wherein the load balancer is based on the priority change of the service node,
the specific implementation method for dynamically adjusting the priority updating period of each service node comprises the following steps:
the load balancer calculates the service node with a period default _ monitor _ time (i)
Priority deviation δ
Wherein Host _ Score (i) is the dynamic priority Score of the ith service node,
host _ Score is the average of the n serving node dynamic priority scores;
defining the upper limit of the priority deviation of the ith service node as threshold (i), judging the size of the upper limit of the priority deviation of the ith service node as the priority deviation delta, if delta is less than threshold (i), the priority update period monitor (i) of the ith service node is as follows:
monitor_time(i)＝default_monitor_time(i) (1)
monitor_time(i)＝monitor_time(i)+atomic_time(i)*
f(δ，threshold(i)) (2)
wherein, the formula (1) represents the initialization setting of the update period of the ith service node,
formula (2) represents the dynamic adjustment of the update period of the ith service node;
atomic _ time (i) is the atomic time of the update cycle interval,
f (δ, threshold (i)) represents a corresponding functional relationship between the priority deviation δ and the upper limit of the priority deviation threshold (i);
if δ > = threshold (i), the priority update period monitor (i) of the ith serving node is:
monitor_time(i)＝max(monitor_time(i)/2,atomic_time(i))。
4. the load balancing strategy based on dynamic priority in the cloud environment according to claim 1, 2 or 3, wherein the strategy is to obtain the status information of each service node through an agent system.
5. A dynamic priority based load balancing policy in a cloud environment according to claim 1, 2 or 3, wherein the policy further comprises the steps of:
the load balancer establishes a buffer queue for request servicing.
6. The dynamic priority based load balancing strategy in the cloud environment of claim 5, wherein the load balancer establishes a buffer queue for the requested service according to a first-come-first-serve strategy.
7. The dynamic priority based load balancing strategy in the cloud environment according to claim 6, wherein the service node with higher priority in the service node priority queue is used as the service node to be serviced for requesting service.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810008891.3A CN107948330A (en) | 2018-01-04 | 2018-01-04 | Load balancing based on dynamic priority under a kind of cloud environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810008891.3A CN107948330A (en) | 2018-01-04 | 2018-01-04 | Load balancing based on dynamic priority under a kind of cloud environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107948330A true CN107948330A (en) | 2018-04-20 |
Family
ID=61938429
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810008891.3A Pending CN107948330A (en) | 2018-01-04 | 2018-01-04 | Load balancing based on dynamic priority under a kind of cloud environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107948330A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110716985A (en) * | 2019-10-16 | 2020-01-21 | 北京小米移动软件有限公司 | Node information processing method, device and medium |
| JP2022540785A (en) * | 2019-07-05 | 2022-09-20 | サービスナウ， インコーポレイテッド | Intelligent load balancer |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104317658A (en) * | 2014-10-17 | 2015-01-28 | 华中科技大学 | MapReduce based load self-adaptive task scheduling method |
| CN104702680A (en) * | 2015-02-13 | 2015-06-10 | 安科智慧城市技术（中国）有限公司 | Resource monitoring method and system |
| CN106250305A (en) * | 2016-08-17 | 2016-12-21 | 浪潮电子信息产业股份有限公司 | The self-adaptation control method of monitoring system data collection cycle under cloud computing environment |
| CN106878111A (en) * | 2017-03-15 | 2017-06-20 | 郑州云海信息技术有限公司 | The cloud monitoring system and monitoring method of a kind of High Availabitity |
| CN107483292A (en) * | 2017-09-11 | 2017-12-15 | 电子科技大学 | Dynamic monitoring and controlling method for cloud platform |
-
2018
- 2018-01-04 CN CN201810008891.3A patent/CN107948330A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104317658A (en) * | 2014-10-17 | 2015-01-28 | 华中科技大学 | MapReduce based load self-adaptive task scheduling method |
| CN104702680A (en) * | 2015-02-13 | 2015-06-10 | 安科智慧城市技术（中国）有限公司 | Resource monitoring method and system |
| CN106250305A (en) * | 2016-08-17 | 2016-12-21 | 浪潮电子信息产业股份有限公司 | The self-adaptation control method of monitoring system data collection cycle under cloud computing environment |
| CN106878111A (en) * | 2017-03-15 | 2017-06-20 | 郑州云海信息技术有限公司 | The cloud monitoring system and monitoring method of a kind of High Availabitity |
| CN107483292A (en) * | 2017-09-11 | 2017-12-15 | 电子科技大学 | Dynamic monitoring and controlling method for cloud platform |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2022540785A (en) * | 2019-07-05 | 2022-09-20 | サービスナウ， インコーポレイテッド | Intelligent load balancer |
| JP7330602B2 (en) | 2019-07-05 | 2023-08-22 | サービスナウ， インコーポレイテッド | Intelligent load balancer |
| CN110716985A (en) * | 2019-10-16 | 2020-01-21 | 北京小米移动软件有限公司 | Node information processing method, device and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110262899B (en) | Monitoring component elastic expansion method and device based on Kubernetes cluster and controlled terminal | |
| CN108829494B (en) | Container cloud platform intelligent resource optimization method based on load prediction | |
| CN109617826B (en) | Storm dynamic load balancing method based on cuckoo search | |
| CN110134495B (en) | Container cross-host online migration method, storage medium and terminal equipment | |
| US8856797B1 (en) | Reactive auto-scaling of capacity | |
| CN106133693B (en) | Moving method, device and the equipment of virtual machine | |
| CN103957237A (en) | Architecture of elastic cloud | |
| CN108416465B (en) | Workflow optimization method in mobile cloud environment | |
| WO2018000991A1 (en) | Data balancing method and device | |
| WO2021228103A1 (en) | Load balancing method and apparatus for cloud host cluster, and server | |
| CN105554102A (en) | Elastic expansion method based on container cluster and application system thereof | |
| CN110231976B (en) | Load prediction-based edge computing platform container deployment method and system | |
| CN109947574B (en) | Fog network-based vehicle big data calculation unloading method | |
| CN110046048B (en) | Load balancing method based on workload self-adaptive fast redistribution | |
| Megharaj et al. | A survey on load balancing techniques in cloud computing | |
| CN106534318A (en) | OpenStack cloud platform resource dynamic scheduling system and method based on flow affinity | |
| Singh et al. | Survey on various load balancing techniques in cloud computing | |
| WO2020134133A1 (en) | Resource allocation method, substation, and computer-readable storage medium | |
| CN109960579B (en) | Method and device for adjusting service container | |
| CN105607943A (en) | Dynamic deployment mechanism of virtual machine under cloud environment | |
| CN110597598B (en) | Control method for virtual machine migration in cloud environment | |
| CN107948330A (en) | Load balancing based on dynamic priority under a kind of cloud environment | |
| CN107203256B (en) | Energy-saving distribution method and device under network function virtualization scene | |
| CN113448714B (en) | Computing resource control system based on cloud platform | |
| CN114691372A (en) | Group intelligent control method of multimedia end edge cloud system |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200519 Address after: Building S01, Inspur Science Park, No. 1036, Inspur Road, high tech Zone, Jinan City, Shandong Province, 250000 Applicant after: Tidal Cloud Information Technology Co.,Ltd. Address before: 450000 Henan province Zheng Dong New District of Zhengzhou City Xinyi Road No. 278 16 floor room 1601 Applicant before: ZHENGZHOU YUNHAI INFORMATION TECHNOLOGY Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180420 |
|
| RJ01 | Rejection of invention patent application after publication |