WO2019061615A1 - 基于云监控的负载均衡优化方法及装置 - Google Patents

基于云监控的负载均衡优化方法及装置 Download PDF

Info

Publication number
WO2019061615A1
WO2019061615A1 PCT/CN2017/107480 CN2017107480W WO2019061615A1 WO 2019061615 A1 WO2019061615 A1 WO 2019061615A1 CN 2017107480 W CN2017107480 W CN 2017107480W WO 2019061615 A1 WO2019061615 A1 WO 2019061615A1
Authority
WO
WIPO (PCT)
Prior art keywords
load balancing
available area
recommended value
instance
available
Prior art date
Application number
PCT/CN2017/107480
Other languages
English (en)
French (fr)
Inventor
匡光彩
易仁杰
Original Assignee
平安科技(深圳)有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 平安科技(深圳)有限公司 filed Critical 平安科技(深圳)有限公司
Priority to JP2018517171A priority Critical patent/JP6671468B2/ja
Priority to EP17882263.1A priority patent/EP3490202B1/en
Priority to US16/075,327 priority patent/US10992581B2/en
Priority to KR1020187035327A priority patent/KR102157722B1/ko
Priority to AU2017404564A priority patent/AU2017404564B2/en
Publication of WO2019061615A1 publication Critical patent/WO2019061615A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/22Alternate routing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00Television systems
    • H04N7/18Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/08Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0823Errors, e.g. transmission errors
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/12Avoiding congestion; Recovering from congestion
    • H04L47/125Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/24Traffic characterised by specific attributes, e.g. priority or QoS
    • H04L47/2425Traffic characterised by specific attributes, e.g. priority or QoS for supporting services specification, e.g. SLA
    • H04L47/2433Allocation of priorities to traffic types
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/10Protocols in which an application is distributed across nodes in the network
    • H04L67/1001Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/1004Server selection for load balancing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/10Protocols in which an application is distributed across nodes in the network
    • H04L67/1001Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers

Definitions

  • the present application relates to the field of load balancing technologies, and in particular, to a load balancing optimization method and apparatus based on cloud monitoring.
  • load balancing scheduling technology is the basis for actual operations such as cloud computing and large-scale application systems.
  • Load balancing enables cloud computing platforms to run stably and efficiently, meet high concurrent requests, and play an important role in cloud computing and large-scale applications.
  • load balancing mainly refers to the ability to allocate client requests that arrive in large quantities in a short period of time to the back-end servers in a balanced and reasonable manner.
  • the present application provides a cloud monitoring-based load balancing optimization method and device, and the main purpose thereof is to solve the problem that the current routing priority of the load balancing device is manually adjusted to affect the switching efficiency of the active and standby Availability zones of the load balancing instance.
  • the priority of the load balancing device cannot be switched intelligently, and there is also a problem of misoperation, which affects the accuracy of the active and standby free zone switching.
  • a cloud monitoring-based load balancing optimization method comprising:
  • the primary available zone switch of the load balancing instance is performed according to the determined result.
  • a cloud monitoring-based load balancing optimization apparatus comprising:
  • An obtaining unit configured to obtain a traffic distribution statistical result in a control area of the load balancing system; and obtain a statistics result of the back-end server in the available area in the load balancing system and a network monitoring quality of the available area;
  • a determining unit configured to determine, according to the traffic distribution statistics result obtained by the acquiring unit, the back-end server statistics result, and the network monitoring quality, an available area that is suitable as a load-balancing instance corresponding to the primary available area;
  • a switching unit configured to perform a primary available area switch of the load balancing instance according to the determination result of the determining unit.
  • a storage device on which a computer program is stored, and the program is implemented by a processor to implement the cloud monitoring-based load balancing optimization method.
  • a cloud monitoring-based load balancing optimization physical device including a storage device, a processor, and a computer program stored on the storage device and executable on the processor, the processor executing The program implements the above cloud monitoring-based load balancing optimization method.
  • the cloud monitoring-based load balancing optimization method and apparatus are compared with the current method of manually adjusting the routing priority of the load balancing device, and the present application controls the area according to the load balancing system.
  • the traffic distribution statistics result the back-end server statistics in each available area, and the network monitoring quality of each available area are comprehensively analyzed, and the available area suitable for the primary available area of the load balancing instance is determined, and automatically according to the suitable available area.
  • the switchover between the active and standby zones of the load balancing instance can improve the switching efficiency of the active and standby zones of the load balancing instance.
  • the priority of the switchover device can be switched to meet the requirements of the service to migrate the available zone on demand. In the case of operation, the accuracy of the active/standby zone switching of the load balancing instance can be improved.
  • FIG. 1 is a schematic flowchart of a cloud monitoring-based load balancing optimization method provided by an embodiment of the present application
  • FIG. 2 is a schematic flowchart of another cloud monitoring-based load balancing optimization method provided by an embodiment of the present application
  • FIG. 3 is a schematic diagram showing an example structure of an available area provided by an embodiment of the present application.
  • FIG. 4 is a schematic diagram of an example of switching between active and standby Availability Zones provided by an embodiment of the present application
  • FIG. 5 is a schematic structural diagram of a cloud monitoring-based load balancing optimization apparatus according to an embodiment of the present disclosure
  • FIG. 6 is a schematic structural diagram of another entity device based on cloud monitoring and load balancing optimization provided by an embodiment of the present application.
  • the embodiment of the present application provides a load balancing optimization method based on the cloud monitoring, which can improve the switching efficiency of the active and standby available areas of the load balancing instance, and can implement the priority of intelligently switching the load balancing device to achieve the service on-demand migration of the available area.
  • the method includes:
  • the area controlled by the load balancing system can be an Internet data center of a city or a specific area (Internet Data Center, IDC)
  • IDC Internet Data Center
  • the traffic of the IDC machine room network load balancing system of the entire city or area can be aggregated and aggregated, and the traffic distribution statistics result can be obtained.
  • the statistics result can include information such as the traffic usage rate and total bandwidth of each available area.
  • the back-end server statistics in each available area can be obtained through the health check of the back-end servers in each available area of the load balancing system, and the back-end server statistics can include the back-end in the available area.
  • the number of servers; and the operational data of the network monitoring quality of each available area, the network quality of each available area is statistically obtained, which may include network quality such as delay, packet loss rate, jitter, and out-of-order.
  • the execution entity of the embodiment of the present application may be a cloud-based load balancing optimization device, which is used to implement the priority of the intelligent switching load balancing device, and may be combined with the load balancing system to control real-time traffic distribution statistics in the region, and each available area.
  • the real-time back-end server statistics and the real-time network monitoring quality of each available area are comprehensively analyzed, and the active and standby available area switching of the load balancing instance is automatically implemented, and the processes described in steps 101 to 103 are specifically performed.
  • the current traffic usage rate and total bandwidth of each available area determine the number of back-end servers currently in a normal state in each available area according to the back-end server statistics, and then The quality of the network monitoring determines the abnormal conditions in each available area.
  • the number of back-end servers with a lower current traffic usage rate and a higher total bandwidth and having a normal state is selected, and The available area N of the abnormal condition rarely occurs, and the available area N is determined to be an available area suitable as the primary available area of the load balancing instance.
  • the current available area N is suitable as the available area of the primary available area of the load balancing instance.
  • the load balancing instance advertises the route, the appropriate priority value is automatically filled in, so that the load balancing instance is distributed to the available area N for processing. Thereby, the active and standby Availability Zone switching of the load balancing instance is automatically implemented.
  • the cloud monitoring-based load balancing optimization method provided by the embodiment of the present application is compared with the current method for manually adjusting the routing priority of the load balancing device.
  • the embodiment of the present application controls the traffic distribution statistics in the region according to the load balancing system.
  • Comprehensive analysis of the back-end server statistics in each available area and the network monitoring quality of each available area determines that an available area suitable for the primary-available area of the load balancing instance is obtained, and the load balancing is automatically performed according to the suitable available area.
  • the switchover between the active and standby zones of the instance can improve the switching efficiency of the active and standby zones of the load balancing instance.
  • the priority of the switchover device can be switched to meet the requirements of the service to migrate the available zone on demand to avoid misoperation. In addition, the accuracy of the active/standby zone switching of the load balancing instance can be improved.
  • the method includes:
  • the step of obtaining the statistics of the back-end server in the available area in the load balancing system may include: determining, by the health check of the back-end server, the number of the back-end servers in the available normal state in the available area; The number determines the back-end server statistics in the Availability Zone in the load balancing system.
  • the preset normal condition state may be pre-configured according to the actual needs of the health check.
  • Health check is a very important function in the load balancing service.
  • the load balancing device forwards traffic to the back-end server and needs to check whether the back-end server is providing services normally through health check. If the health check of the back-end server in the abnormal state reaches the health check threshold, the back-end server is restored to the normal state and traffic is forwarded. If the traffic is normal, If the health check of the server fails to reach the Health Check Abnormal Threshold, the backend server is considered to be in an abnormal state and traffic forwarding is stopped.
  • the number of backend servers in the normal state existing in each available zone can be determined, and the backend server statistics in the available zone in the load balancing system are determined according to the number.
  • control area is city A
  • traffic of the IDC room public network load balancing system of the city A is uniformly aggregated and counted, and it is assumed that the area has N available areas, namely AZ1, AZ2, ..., AZn.
  • the bandwidth of each available area (Bandwidth) is B1, B2, ..., Bn.
  • the current average of the public network egress bandwidth used by the current statistics is B_Ave1, B_Ave2, ..., B_AveN.
  • the step of calculating, according to the obtained egress bandwidth and the average value of the egress of the public network, the first recommended value of each of the available areas in the load balancing system as the load balancing instance corresponding to the main available area which may include: corresponding to the available area
  • the average value of the bandwidth used by the public network is divided by the egress bandwidth of the available area to obtain the traffic usage rate of the available area.
  • the total bandwidth of the available area is multiplied by the preset coordination factor coefficient to obtain the weight corresponding to the available area. The lower the usage rate is.
  • the respective recommended areas in the load balancing system are determined as the first recommended value of the primary available area of the load balancing instance.
  • the preset coordination factor coefficient may be preset by a technician according to actual conditions.
  • the traffic usage rate of each available area is calculated by the following formula.
  • the available areas are sorted according to the size of the traffic usage. If the traffic usage is the same, the total bandwidth is ranked higher. According to the ranking result, the corresponding reference value is configured according to the principle that the larger the reference value is. Then, according to the weight of each available area, the total bandwidth of AZ1 is 1G, and the total bandwidth of AZ2 is 2G. If the total bandwidth is larger, the weight is larger, and the coordination factor is r, then AZ1 The weight is r*1, and the weight of AZ2 is r*2. Finally, the recommended reference value of each available area configuration is multiplied by the corresponding weight to obtain the recommended value Ra.
  • the current traffic usage rate and total bandwidth of each available area can be fully combined, and the preferred primary available area is recommended for the load balancing instance according to the principle of preferential recommendation with lower priority of traffic usage and higher priority of total bandwidth. .
  • the step 203 may include: determining, according to the principle that the number of recommended values is higher, determining that each available area in the load balancing system is used as the second recommended value of the primary available area of the load balancing instance.
  • the recommended value Rb of each available area switch is configured according to the principle that the higher the recommended value is, the normal back-end server is available in AZ1. 6 sets, there are 2 back-end servers available in AZ2, then the recommended value of AZ1 free zone switch will be greater than the recommended value of AZ2 free zone switch.
  • the step 204 may include: calculating, according to the operation data obtained by the network monitoring quality, the number of times the network quality parameter of the available area is abnormal during the operation time; determining the each of the load balancing system according to the principle that the smaller the number of abnormal times is, the smaller the recommended value is.
  • the available areas are respectively used as the third recommended value of the primary available area of the load balancing instance.
  • the network quality of each AZ is statistically obtained, including network delay parameters such as network delay, DNS resolution delay, packet loss rate, jitter, and out-of-order.
  • network delay parameters such as network delay, DNS resolution delay, packet loss rate, jitter, and out-of-order.
  • time of operation historical data statistics are performed. For example, the network delay in the available area is long, the DNS resolution delay is long, the packet loss rate is high, and the number of abnormal conditions such as jitter and disorder is increased. The smaller the value, the configuration gets the recommended value Rc for each available zone switch.
  • the abnormal situation in each available area can be fully combined, and the appropriate primary available area is recommended for the load balancing instance according to the principle of reducing the occurrence of abnormal conditions.
  • the step 205 may specifically include: multiplying the calculated first recommended value by the weight of the corresponding item, adding the second recommended value by the weight of the corresponding item, and adding the third recommended value by the weight of the corresponding item.
  • the obtained sum value is determined as the recommended value result of each available area in the load balancing system as the load-available instance corresponding to the main available area.
  • the formula Rating Ra*wa + Rb*wb + is used.
  • Rc*wc the final recommended value result is obtained, where Rating represents the recommended value result of the available area handover, wa is the weight corresponding to the first item, wb is the weight corresponding to the second item, and wc is the corresponding item of the third item Weight.
  • the step 206 may include: determining an available area with the highest recommended value result, or an available area with a recommended value result greater than a preset recommended value threshold, and determining that the available area is suitable as the load-available instance corresponding to the primary available area.
  • the available area A with the highest recommended value result may be determined as the available area corresponding to the primary available area of the load balancing instance; a certain recommended value threshold may be preset, and the available area larger than the recommended value threshold is eligible. Selecting an available area from it is determined to be suitable as an available area corresponding to the primary available area of the load balancing instance.
  • the score may be scored according to the recommended value result, and the user may be recommended as the available area of the primary available area according to the scored result, and the user selects which available area as the load.
  • Balance the primary Availability Zone of the instance For example, a score of 1-100 points is scored according to the recommended value result, and the scoring area is segmented, and the recommended strength and explanation of each segment are explained. For example, AZ of 80 points or more, the main AZ setting of the load balancing instance is strongly recommended. For this AZ.
  • the load balancing system uses a dynamic routing protocol to distribute 32-bit host routes. For example, a load balancing instance configured on a load balancing device is advertised to the router through a 32-bit route. The router needs to maintain a 32-bit route. Solutions consume too much system resources. As shown in Figure 3, it is the deployment structure of the dual-availability zone in the same city. Data Processing Center Internet (Datacenter Interconnection, DCI) channel, used to connect two available rooms to the machine room. Router 3 and Router 4 belong to the backbone network of the local domain and are used to directly connect to the Internet. Router 3 and router 4 are connected through a DCI channel. Router 1 and Router 2 belong to the data center network inside AZ1 and AZ2 respectively.
  • DCI Data Processing Center Internet
  • the load balancing instance (a VIP) configured on the load balancing device is advertised to Router 1 and Router 2 through 32-bit routing. Therefore, each new load balancing instance needs to issue a new 32-bit route to Router 1 and Router 2. The routers 1 and 2 need to maintain the 32-bit route. Therefore, this scheme consumes too many systems. Resources.
  • the method further includes: pre-posting the network segment route to the load balancing device to which the load balancing instance belongs by configuring the static equal-cost route The corresponding device on the load balancing device; for example, in conjunction with the example shown in FIG. 3, the network segment will be routed from the load balancing device to the router 1 and the router 2. Then, through the network controller, Router 1 and Router 2 are directly scheduled, and 32-bit routes are issued to Router 3 and Router 4. In this way, the 32-bit routes that need to be maintained on Router 1 and Router 2 will disappear, thus saving system resources.
  • AZ1, LB1, LB2, and LB3 are three load balancing devices.
  • the device forms a load balancing resource pool.
  • the load balancing IP address segment is configured on Router 1 as three static equal-cost routes, pointing to three devices.
  • the 32-bit route of the load balancing instance is released as needed, with the priority value a, to the router 3; the same thing is done in the AZ2, according to the configuration requirements of the load balancing instance, as needed Publish the 32-bit route of the load balancing instance with the priority value b to Router 4.
  • the step 207 may include: switching the primary available area of the load balancing instance to the determination result by adjusting a priority value attached to the router by issuing a 32-bit route corresponding to the load balancing instance to the backbone network router.
  • the available area where the backbone network routers of different available areas are connected through the DCI channel.
  • the action of the primary available area switching of the specific load balancing instance may be completed by the centralized controller scheduling router 1 and the router 2, that is, the priority value a is automatically adjusted.
  • the value of b causes router 1 to issue a 32-bit route with the adjusted priority value a to router 3, and router 2 issues a 32-bit route with the adjusted priority value b to router 4 to implement the load.
  • the primary free area of the equalization instance is automatically switched.
  • the time for switching the main available area may also be given according to the operation experience. For example, in general, the time suitable for switching is from 11:00 to the next day. At 6 o'clock in the morning, but some systems may have some special recommendation time, which is a more suitable recommendation time.
  • the operation of switching the available area of the load balancing instance is automatically entered through the cloud platform timing operating system, that is, the load balancing is determined through the above implementation process.
  • the primary Availability Zone of the instance is set to which Availability Zone, and the operation of switching the primary Availability Zone is automatically completed at the specified time.
  • Another cloud monitoring-based load balancing optimization method can improve the switching efficiency of the active and standby available areas of the load balancing instance, and can implement the priority of intelligently switching the load balancing device to achieve the service-on-demand migration of the available area.
  • the purpose is to avoid the occurrence of misoperations, thereby improving the accuracy of the active and standby availability zone switching of the load balancing instance; and saving system resources.
  • the embodiment of the present application provides a load balancing optimization device based on cloud monitoring.
  • the device includes: an acquiring unit 31, and a determining unit. 32. Switching unit 33.
  • the obtaining unit 31 may be configured to obtain a traffic distribution statistical result in a load balancing system control region; and obtain a back-end server statistical result in the available area of the load balancing system and a network monitoring quality of the available area;
  • the determining unit 32 may be configured to determine, according to the traffic distribution statistics result obtained by the obtaining unit 31, the back-end server statistics result, and the network monitoring quality, an available area that is suitable as a load-balancing instance corresponding to the primary available area;
  • the switching unit 33 may be configured to perform the primary available area switching of the load balancing instance according to the determination result of the determining unit 32.
  • the determining unit 32 includes: an obtaining module 321, a calculating module 322, and a determining module 323.
  • the obtaining module 321 is configured to obtain, according to the statistics of the traffic distribution statistics, an egress bandwidth of each available area in the load balancing system and an average value of a public network egress use bandwidth corresponding to each available area;
  • the calculating module 322 is configured to calculate, according to the egress bandwidth and the average value of the used bandwidth of the public network, a first recommended value of each available area in the load balancing system as a main available area of the load balancing instance;
  • the calculation module 322 is further configured to calculate, according to the statistics result of the backend server, a second recommended value of each available area in the load balancing system as a primary available area of the load balancing instance.
  • the calculating module 322 is further configured to calculate, according to the network monitoring quality, a third recommended value of each available area in the load balancing system as a load-available instance corresponding to the primary available area;
  • the determining module 323 is configured to determine, according to the first recommended value, the second recommended value, the third recommended value, and the weights corresponding to the three, each available area in the load balancing system as a load
  • the equalization instance corresponds to the recommended value result of the main available area
  • the determining module 323 is further configured to determine, according to the recommended value result, an available area that is suitable as a primary available area of the load balancing instance.
  • the calculation module 322 may be specifically configured to divide the average bandwidth of the public network outlet corresponding to the available area by the outlet bandwidth of the available area to obtain the traffic usage rate of the available area; multiply the total bandwidth of the available area. Determining the weight corresponding to the available area by using the preset coordination factor coefficient; determining that each of the load balancing systems is available according to the principle that the lower the recommended usage value is, the higher the recommended value is, the higher the recommended value is.
  • the areas respectively serve as the first recommended value of the primary available area corresponding to the load balancing instance.
  • the obtaining unit 31 may be specifically configured to determine, by using a health check on the backend server, the number of backend servers in the available normal state in the available area; determining the load balancing according to the quantity.
  • the calculation module 322 may be further configured to determine, according to the principle that the number of the recommended values is higher, the respective available areas in the load balancing system are respectively used as the second recommended value of the primary available area of the load balancing instance.
  • the calculation module 322 may be further configured to: according to the operation data of the network monitoring quality, the number of times that the network quality parameter of the available area is abnormal during the operation time; According to a small principle, each available area in the load balancing system is determined as a third recommended value corresponding to the primary available area of the load balancing instance.
  • the determining module 323 may be specifically configured to multiply the first recommended value by the weight of the corresponding item, and add the second recommended value by the weight of the corresponding item, plus The third recommended value is multiplied by the weight of the corresponding item, and the obtained sum value is determined as the recommended value result of each available area in the load balancing system as the load-available instance corresponding to the main available area.
  • the determining module 323 may be further configured to use an available area with the highest recommended value result or an available area with a recommended value result that is greater than a preset recommended value threshold, and is determined to be suitable as a load balancing instance corresponding to the primary available area. Availability area.
  • the device further includes: a publishing unit 34;
  • the issuing unit 34 may be configured to pre-advertise the network segment route to the router corresponding to the load balancing device by using the load balancing device to which the load balancing instance belongs by configuring the static equal-cost route.
  • the switching unit 33 is specifically configured to switch the primary available area of the load balancing instance to be adjusted by adjusting a priority value attached to the router by issuing a 32-bit route corresponding to the load balancing instance to the backbone network router.
  • the available area in the result is determined, wherein the backbone network routers of different available areas are connected by the data processing center interconnection network DCI channel.
  • the embodiment of the present application further provides a storage device, where a computer program is stored, and when the program is executed by the processor, the foregoing FIG. 1 and FIG. 2 are implemented.
  • a cloud-based load balancing optimization method is shown.
  • the embodiment of the present application further provides an entity based on cloud monitoring and load balancing optimization.
  • the physical device comprising a storage device and a processor; the storage device for storing a computer program; the processor for executing the computer program to implement the cloud-based monitoring as shown in FIG. 1 and FIG. 2 above Load balancing optimization method.
  • the switching efficiency of the active and standby Availability Zones of the load balancing instance can be improved, and the priority of the intelligent load balancing device can be switched to achieve the purpose of the service to migrate the available area on demand, thereby avoiding the occurrence of misoperation.
  • the accuracy of the active/standby zone switching of the load balancing instance can be improved; and system resources can be saved.
  • the present application can be implemented by hardware, or by software plus a necessary general hardware platform.
  • the technical solution of the present application may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including several The instructions are for causing a computer device (which may be a personal computer, server, or network device, etc.) to perform the methods described in various implementation scenarios of the present application.
  • modules in the apparatus in the implementation scenario may be distributed in the apparatus for implementing the scenario according to the implementation scenario description, or may be correspondingly changed in one or more devices different from the implementation scenario.
  • the modules of the above implementation scenarios may be combined into one module, or may be further split into multiple sub-modules.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Multimedia (AREA)
  • Environmental & Geological Engineering (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本申请公开了一种基于云监控的负载均衡优化方法及装置,涉及负载均衡技术领域,可以提高负载均衡实例的主备可用区切换效率,可以实现智能切换负载均衡设备的优先级。所述方法包括:获取负载均衡系统控制地域内的流量分布统计结果;及获取所述负载均衡系统中可用区内的后端服务器统计结果和可用区的网络监控质量;根据所述流量分布统计结果、所述后端服务器统计结果和所述网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区;按照确定结果进行所述负载均衡实例的主可用区切换。本申请适用于负载均衡的优化。

Description

基于云监控的负载均衡优化方法及装置
技术领域
本申请涉及负载均衡技术领域,特别是涉及一种基于云监控的负载均衡优化方法及装置。
背景技术
高效、准确的负载均衡调度技术是云计算及大型应用系统等实际运营的基础。负载均衡可以使云计算平台能稳定、高效地运行,满足高并发请求,在云计算及大型应用系统中扮演着重要角色。在云计算领域,所谓负载均衡主要是指能够均衡、合理地将短时间内大量到达的客户端请求分配给后端各服务器之上。
目前大部分公有云的负载均衡系统仅仅支持主/备HA技术。主备可用区切换时,需要手动调整负载均衡设备的路由优先级,例如,手动修改每个可用区(Available Zone,AZ)中负载均衡实例发布路由时填写的优先级数值,进而通过路由优先级决定这个负载均衡实例被分发到哪一个可用区中,从而达到切换可用区的目的。
然而,通过手动调整负载均衡设备的路由优先级的方式会影响负载均衡实例的主备可用区切换效率,无法智能切换负载均衡设备的优先级,并且还会存在误操作的情况,影响主备可用区切换的准确性。
发明内容
有鉴于此,本申请提供了一种基于云监控的负载均衡优化方法及装置,主要目的在于解决目前通过手动调整负载均衡设备的路由优先级的方式会影响负载均衡实例的主备可用区切换效率,无法智能切换负载均衡设备的优先级,并且还会存在误操作的情况,影响主备可用区切换的准确性的问题。
依据本申请一个方面,提供了一种基于云监控的负载均衡优化方法,该方法包括:
获取负载均衡系统控制地域内的流量分布统计结果;及
获取所述负载均衡系统中可用区内的后端服务器统计结果和可用区的网络监控质量;
根据所述流量分布统计结果、所述后端服务器统计结果和所述网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区;
按照确定结果进行所述负载均衡实例的主可用区切换。
依据本申请另一个方面,提供了一种基于云监控的负载均衡优化装置,该装置包括:
获取单元,用于获取负载均衡系统控制地域内的流量分布统计结果;及获取所述负载均衡系统中可用区内的后端服务器统计结果和可用区的网络监控质量;
确定单元,用于根据所述获取单元获取的流量分布统计结果、所述后端服务器统计结果和所述网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区;
切换单元,用于按照所述确定单元的确定结果进行所述负载均衡实例的主可用区切换。
依据本申请又一个方面,提供了一种存储设备,其上存储有计算机程序,所述程序被处理器执行时实现上述基于云监控的负载均衡优化方法。
依据本申请再一个方面,提供了一种基于云监控的负载均衡优化的实体装置,包括存储设备、处理器及存储在存储设备上并可在处理器上运行的计算机程序,所述处理器执行所述程序时实现上述基于云监控的负载均衡优化方法。
借由上述技术方案,本申请提供的一种基于云监控的负载均衡优化方法及装置,与目前通过手动调整负载均衡设备的路由优先级的方式相比,本申请根据负载均衡系统控制地域内的流量分布统计结果、各个可用区内的后端服务器统计结果和各个可用区的网络监控质量进行综合分析,确定得到适合作为负载均衡实例对应主可用区的可用区,并按照这个合适的可用区自动进行该负载均衡实例的主备可用区切换,可以提高负载均衡实例的主备可用区切换效率,可以实现智能切换负载均衡设备的优先级,以达到业务按需迁移可用区的目的,避免出现误操作的情况,进而可以提高负载均衡实例的主备可用区切换的准确性。
上述说明仅是本申请技术方案的概述,为了能够更清楚了解本申请的技术手段,而可依照说明书的内容予以实施,并且为了让本申请的上述和其它目的、特征和优点能够更明显易懂,以下特举本申请的具体实施方式。
附图说明
通过阅读下文优选实施方式的详细描述,各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。附图仅用于示出优选实施方式的目的,而并不认为是对本申请的限制。而且在整个附图中,用相同的参考符号表示相同的部件。在附图中:
图1示出了本申请实施例提供的一种基于云监控的负载均衡优化方法流程示意图;
图2示出了本申请实施例提供的另一种基于云监控的负载均衡优化方法流程示意图;
图3示出了本申请实施例提供的一种可用区实例架构示意图;
图4示出了本申请实施例提供的一种主备可用区切换实例示意图;
图5示出了本申请实施例提供的一种基于云监控的负载均衡优化装置的结构示意图;
图6示出了本申请实施例提供的另一种基于云监控的负载均衡优化的实体装置结构示意图。
具体实施方式
下面将参照附图更详细地描述本公开的示例性实施例。虽然附图中显示了本公开的示例性实施例,然而应当理解,可以以各种形式实现本公开而不应被这里阐述的实施例所限制。相反,提供这些实施例是为了能够更透彻地理解本公开,并且能够将本公开的范围完整的传达给本领域的技术人员。
本申请实施例提供了一种基于云监控的负载均衡优化方法,可以提高负载均衡实例的主备可用区切换效率,可以实现智能切换负载均衡设备的优先级,以达到业务按需迁移可用区的目的,如图1所示,该方法包括:
101、获取负载均衡系统控制地域内的流量分布统计结果,以及获取负载均衡系统中可用区内的后端服务器统计结果和可用区的网络监控质量。
其中,负载均衡系统控制的地域可以为一个城市或一个特定区域的互联网数据中心(Internet Data Center,IDC)机房集合,由同城或者同区域的多个IDC机房组成。
例如,可以将整个城市或者区域的IDC机房公网负载均衡系统的流量进行统一汇总统计,得到流量分布统计结果,该统计结果中可以包含每个可用区的流量使用率、总带宽大小等信息;可以通过对该负载均衡系统中各个可用区内的后端服务器的健康检查,得到各个可用区内的后端服务器统计结果,该后端服务器统计结果中可以包含可用区内处于正常状态的后端服务器数量;以及根据每个可用区的网络监控质量的运营数据,统计得到每个可用区的网络质量情况,具体可以包括时延、丢包率、抖动、乱序等网络质量情况。
对于本申请实施例的执行主体可以为基于云监控的负载均衡优化装置,用于实现智能切换负载均衡设备的优先级,可以结合负载均衡系统控制地域内实时的流量分布统计结果、各个可用区内实时的后端服务器统计结果和各个可用区实时的网络监控质量进行综合分析,自动实现负载均衡实例的主备可用区切换,具体执行步骤101至103所述的过程。
102、根据获取到的流量分布统计结果、后端服务器统计结果和网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区。
例如,首先根据获取到的流量分布统计结果,确定各个可用区当前的流量使用率和总带宽,再根据后端服务器统计结果,确定各个可用区中当前处于正常状态的后端服务器数量,然后根据网络监控质量,确定各个可用区中出现的异常状况,最后根据这些信息,选择当前流量使用率较低的、且总带宽较高的、且拥有处于正常状态的后端服务器数量较多的、且很少出现异常状况的可用区N,将可用区N确定为适合作为负载均衡实例对应主可用区的可用区。
103、按照确定结果进行负载均衡实例的主备可用区切换。
例如,当前可用区N适合作为负载均衡实例对应主可用区的可用区,在该负载均衡实例发布路由时自动填写合适的优先级数值,使得将该负载均衡实例分发到可用区N中进行处理,从而自动实现该负载均衡实例的主备可用区切换。
本申请实施例提供的一种基于云监控的负载均衡优化方法,与目前通过手动调整负载均衡设备的路由优先级的方式相比,本申请实施例根据负载均衡系统控制地域内的流量分布统计结果、各个可用区内的后端服务器统计结果和各个可用区的网络监控质量进行综合分析,确定得到适合作为负载均衡实例对应主可用区的可用区,并按照这个合适的可用区自动进行该负载均衡实例的主备可用区切换,可以提高负载均衡实例的主备可用区切换效率,可以实现智能切换负载均衡设备的优先级,以达到业务按需迁移可用区的目的,避免出现误操作的情况,进而可以提高负载均衡实例的主备可用区切换的准确性。
进一步地,作为上述实施例具体实施方式的细化和扩展,提供了另一种基于云监控的负载均衡优化方法,如图2所示,该方法包括:
201、获取负载均衡系统控制地域内的流量分布统计结果,以及获取负载均衡系统中可用区内的后端服务器统计结果和可用区的网络监控质量。
对于获取负载均衡系统中可用区内的后端服务器统计结果的步骤,具体可以包括:通过对后端服务器的健康检查,确定可用区中处于预设正常条件状态的后端服务器的数量;根据该数量确定负载均衡系统中可用区内的后端服务器统计结果。其中,预设正常条件状态可以根据健康检查的实际需求预先进行配置。
健康检查是负载均衡服务中非常重要的功能,负载均衡设备将流量转发到后端服务器上,需要通过健康检查来探测后端服务器是否在正常提供服务。若对处于异常状态的后端服务器的健康检查连续成功的次数达到“健康检查正常阈值”,则认为该后端服务器恢复到正常状态,并开始对它进行流量转发;若对处于正常状态的后端服务器的健康检查连续失败的次数达到“健康检查异常阈值”,则认为该后端服务器处于异常状态,并停止对它的流量转发。
通过对后端服务器的健康检查可以确定每个可用区中存在的处于正常状态的后端服务器的数量,根据该数量确定负载均衡系统中可用区内的后端服务器统计结果。
202、根据获取到的流量分布统计结果,获取负载均衡系统中各个可用区的出口带宽和各个可用区分别对应的公网出口使用带宽平均值,并依据获取到的出口带宽和公网出口使用带宽平均值,计算负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值。
例如,控制地域为城市A,将城市A的IDC机房公网负载均衡系统的流量进行统一汇总统计,假设这个地域拥有N个可用区,分别为AZ1、AZ2、…、AZn。每个可用区的出口带宽(Bandwidth)分别为B1、B2、…、Bn。当前统计到的公网出口使用带宽每天的平均值为B_Ave1、B_Ave2、…、B_AveN。
对于依据获取到的出口带宽和公网出口使用带宽平均值,计算负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值的步骤,具体可以包括:将可用区对应的公网出口使用带宽平均值除以可用区的出口带宽,得到可用区的流量使用率;将可用区的总带宽乘以预设协调因子系数得到可用区对应的权值;按照流量使用率越低推荐值越高、且权值越大推荐值越高的原则,确定负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值。其中,预设协调因子系数可以由技术人员根据实际情况预先进行设置。
例如,结合上述实例,通过如下公式计算得到每个可用区的流量使用率,
然后根据流量使用率的大小对各个可用区进行排序,如果流量使用率相同,总带宽大的靠前排序,根据排序结果,按照越靠前参考值越大的原则,分别配置相应的推荐参考值;然后根据每个可用区的权值进行评分,如AZ1的总带宽为1G,而AZ2的总带宽为2G,如果总带宽越大那么权值就越大,加入协调因子为r,那么AZ1的权值为r*1,而AZ2的权值为r*2;最后将每个可用区配置的推荐参考值乘以各自对应的权值,得到推荐值Ra。
通过上述方式可以充分结合各个可用区当前的流量使用率和总带宽的情况,按照流量使用率越低的优先推荐、总带宽越高的优先推荐的原则,为负载均衡实例推荐合适的主可用区。
203、根据获取到的后端服务器统计结果,计算负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值。
步骤203具体可以包括:按照数量越多推荐值越高的原则,确定负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值。
例如,依据每个可用区中处于正常状态的后端服务器的数量,按照数量越多推荐值越高的原则,配置每个可用区切换的推荐值Rb,如AZ1中有正常可用的后端服务器6台,AZ2中有正常可用的后端服务器2台,那么AZ1可用区切换的推荐值将会大于AZ2可用区切换的推荐值。
通过上述方式可以充分结合各个可用区中正常可用的后端服务器数量,即可用区分布比例,按照正常可用的后端服务器越多、处理能力越强的原则,为负载均衡实例推荐合适的主可用区。
204、根据获取到的网络监控质量,计算负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值。
步骤204具体可以包括:根据获取到网络监控质量的运营数据,统计可用区的网络质量参数在运营时间内出现异常的次数;按照异常次数越大推荐值越小的原则,确定负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值。
例如,根据网络监控质量的运营数据,统计得到每个AZ的网络质量情况,具体包括网络时延、DNS解析时延、丢包率、抖动、乱序等网络质量参数。根据运营的时间,进行历史数据统计,如可用区内出现网络时延较长,DNS解析时延较长、丢包率较多、抖动、乱序等异常情况的次数,按照异常次数越大推荐值越小的原则,配置得到每个可用区切换的推荐值Rc。
通过上述方式可以充分结合各个可用区中异常情况,按照减少出现异常状况的原则,为负载均衡实例推荐合适的主可用区。
205、根据计算得到的第一推荐值、第二推荐值、第三推荐值以及三者各自对应的权值,确定负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果。
步骤205具体可以包括:将计算得到的第一推荐值乘以相应项的权值,加上第二推荐值乘以相应项的权值,再加上第三推荐值乘以相应项的权值,得到的和值确定为负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果。
例如,结合上述实例,参照步骤202至204中得到的Ra、Rb、Rc三个推荐值,利用公式Rating=Ra*wa + Rb*wb + Rc*wc,得到最终的推荐值结果,其中,Rating代表可用区切换的推荐值结果,wa为第一项对应的权值,wb为第二项对应的权值,wc为第三项对应的权值。
通过上述方式可以根据实际情况,对不同因素实现不同的侧重考虑,准确的计算出各个可用区切换的推荐值结果。
206、按照推荐值结果,确定适合作为负载均衡实例对应主可用区的可用区。
步骤206具体可以包括:将推荐值结果最高的可用区、或推荐值结果大于预置推荐值阈值的可用区,确定为适合作为负载均衡实例对应主可用区的可用区。
例如,可以将推荐值结果最高的可用区A,确定为适合作为负载均衡实例对应主可用区的可用区;还可以预先设定一定推荐值阈值,大于该推荐值阈值的可用区都符合条件,从中选择一个可用区确定为适合作为负载均衡实例对应主可用区的可用区。
为了满足不同的业务需求,在本申请的一个可选实施例中,还可以根据推荐值结果进行打分,按照打分结果为用户推荐合适作为主可用区的可用区,由用户选择哪个可用区作为负载均衡实例的主可用区。例如,根据推荐值结果进行1-100分的打分,并进行打分区域分段,解释每个分段的推荐力度和解释说明,如80分以上的AZ,强力推荐将负载均衡实例的主AZ设置为这个AZ。
207、按照确定结果进行负载均衡实例的主备可用区切换。
目前负载均衡系统大部分使用动态路由协议进行32位主机路由的发布,例如,负载均衡设备上配置的负载均衡实例通过32位路由的方式发布到路由器上,该路由器上需要维护32位路由,这种方案会消耗过多的系统资源。如图3所示,为同城异地双可用区的部署结构。资料处理中心互联网络(Datacenter Interconnection,DCI)通道,用于连接两个可用区的机房。路由器3和路由器4属于本地域的骨干网络,用于直接对接互联网。路由器3和路由器4之间通过DCI通道连接。路由器1和路由器2分别属于AZ1、AZ2内部的数据中心网络,在传统的方案中,负载均衡设备上配置的负载均衡实例(某个VIP)通过32位路由的方式发布到路由器1和路由器2上,这样每新增一个负载均衡实例就需要发布一条新的32位路由发布到路由器1和路由器2上,路由器1和路由器2上需要维护该32位路由,因此这种方案会消耗过多的系统资源。
为了解决上述问题,提高部署的可用性,对于本申请实施例,在步骤207之前,还可以包括:通过配置静态等价路由的方式,从与负载均衡实例所属的负载均衡设备预先发布网段路由到负载均衡设备相应的路由器上;例如,结合图3所示的实例进行说明,将从负载均衡设备发布网段路由到路由器1和路由器2。然后通过网络控制器,直接调度路由器1和路由器2,发布32位路由到路由器3和路由器4。这样路由器1和路由器2上原先需要维护的32位路由就会消失,从而节省了系统资源,如图4所示,在AZ1中,存在LB1、LB2和LB3是三个负载均衡设备,由这三台设备形成负载均衡资源池。此时在路由器1上将负载均衡IP地址段配置为三条静态等价路由,指向三台设备。然后,按照负载均衡实例的配置需求,按需发布负载均衡实例的32位路由,附带优先级数值a,到路由器3;在AZ2中也做相同的事情,按照负载均衡实例的配置需求,按需发布负载均衡实例的32位路由,附带优先级数值b,到路由器4。
相应的,步骤207具体可以包括:通过调整该路由器在发布与负载均衡实例对应的32位路由到骨干网络路由器时附带的优先级数值,实现将负载均衡实例的主可用区切换为确定结果中的可用区,其中不同可用区的骨干网络路由器之间通过DCI通道进行连接。
例如,结合图3和图4所示,并结合上述实施过程,具体的负载均衡实例的主可用区切换的动作可以通过集中控制器调度路由器1和路由器2完成,即自动调整上述优先级数值a、b的数值,使得路由器1发布32位路由,附带调整后的优先级数值a,到路由器3,并且路由器2发布32位路由,附带调整后的优先级数值b,到路由器4,进而实现负载均衡实例的主可用区自动切换。
进一步的,为了实现帮助用户自动在指定时间完成切换主可用区,还可以根据运营经验,给出切换主可用区的时间,例如,一般来说,适合切换的时间为晚上11点到第二天早上6点,但是有些系统可能有一些特殊的推荐时间是更加合适的推荐时间,通过云平台定时操作系统,将负载均衡实例切换可用区的操作自动录入,即通过上述实施过程确定需要将负载均衡实例的主可用区设置为哪个可用区,并在指定时间自动完成该切换主可用区的操作。
本申请实施例提供的另一种基于云监控的负载均衡优化方法,可以提高负载均衡实例的主备可用区切换效率,可以实现智能切换负载均衡设备的优先级,以达到业务按需迁移可用区的目的,避免出现误操作的情况,进而可以提高负载均衡实例的主备可用区切换的准确性;并且可以节省系统资源。
进一步地,作为图1和图2所述方法的具体实现,本申请实施例提供了一种基于云监控的负载均衡优化装置,如图5所示,所述装置包括:获取单元31、确定单元32、切换单元33。
获取单元31,可以用于获取负载均衡系统控制地域内的流量分布统计结果;及获取所述负载均衡系统中可用区内的后端服务器统计结果和可用区的网络监控质量;
确定单元32,可以用于根据所述获取单元31获取的流量分布统计结果、所述后端服务器统计结果和所述网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区;
切换单元33,可以用于按照所述确定单元32的确定结果进行所述负载均衡实例的主可用区切换。
在具体的应用场景中,如图6所示,确定单元32,具体包括:获取模块321、计算模块322、确定模块323。
获取模块321,可以用于根据所述流量分布统计结果,获取所述负载均衡系统中各个可用区的出口带宽和所述各个可用区分别对应的公网出口使用带宽平均值;
计算模块322,可以用于依据所述出口带宽和所述公网出口使用带宽平均值,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值;
计算模块322,还可以用于根据所述后端服务器统计结果,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值;
计算模块322,还可以用于根据所述网络监控质量,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值;
确定模块323,可以用于根据所述第一推荐值、所述第二推荐值、所述第三推荐值以及三者各自对应的权值,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果;
确定模块323,还可以用于按照所述推荐值结果,确定适合作为负载均衡实例对应主可用区的可用区。
在具体的应用场景中,计算模块322,具体可以用于将可用区对应的公网出口使用带宽平均值除以可用区的出口带宽,得到可用区的流量使用率;将可用区的总带宽乘以预设协调因子系数得到可用区对应的权值;按照所述流量使用率越低推荐值越高、且所述权值越大推荐值越高的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值。
在具体的应用场景中,获取单元31,具体可以用于通过对后端服务器的健康检查,确定可用区中处于预设正常条件状态的后端服务器的数量;根据所述数量确定所述负载均衡系统中可用区内的后端服务器统计结果;
计算模块322,具体还可以用于按照所述数量越多推荐值越高的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值。
在具体的应用场景中,计算模块322,具体还可以用于根据所述网络监控质量的运营数据,统计可用区的网络质量参数在运营时间内出现异常的次数;按照异常次数越大推荐值越小的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值。
在具体的应用场景中,确定模块323,具体可以用于将所述第一推荐值乘以相应项的权值,加上所述第二推荐值乘以相应项的权值,再加上所述第三推荐值乘以相应项的权值,得到的和值确定为所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果。
在具体的应用场景中,确定模块323,具体还可以用于将推荐值结果最高的可用区、或推荐值结果大于预置推荐值阈值的可用区,确定为适合作为负载均衡实例对应主可用区的可用区。
在具体的应用场景中,如图6所示,所述装置还包括:发布单元34;
发布单元34,可以用于通过配置静态等价路由的方式,从与所述负载均衡实例所属的负载均衡设备预先发布网段路由到所述负载均衡设备相应的路由器上;
切换单元33,具体可以用于通过调整所述路由器在发布与所述负载均衡实例对应的32位路由到骨干网络路由器时附带的优先级数值,实现将所述负载均衡实例的主可用区切换为确定结果中的可用区,其中不同可用区的骨干网络路由器之间通过资料处理中心互联网络DCI通道进行连接。
需要说明的是,本申请实施例提供的一种基于云监控的负载均衡优化装置所涉及各功能单元的其他相应描述,可以参考图1和图2中的对应描述,在此不再赘述。
基于上述如图1和图2所示方法,相应的,本申请实施例还提供了一种存储设备,其上存储有计算机程序,该程序被处理器执行时实现上述如图1和图2所示的基于云监控的负载均衡优化方法。
基于上述如图1和图2所示方法和如图5和如图6所示虚拟装置的实施例,为了实现上述目的,本申请实施例还提供了一种基于云监控的负载均衡优化的实体装置,该实体装置包括存储设备和处理器;所述存储设备,用于存储计算机程序;所述处理器,用于执行所述计算机程序以实现上述如图1和图2所示的基于云监控的负载均衡优化方法。
通过应用本申请的技术方案,可以提高负载均衡实例的主备可用区切换效率,可以实现智能切换负载均衡设备的优先级,以达到业务按需迁移可用区的目的,避免出现误操作的情况,进而可以提高负载均衡实例的主备可用区切换的准确性;并且可以节省系统资源。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到本申请可以通过硬件实现,也可以借助软件加必要的通用硬件平台的方式来实现。基于这样的理解,本申请的技术方案可以以软件产品的形式体现出来,该软件产品可以存储在一个非易失性存储介质(可以是CD-ROM,U盘,移动硬盘等)中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本申请各个实施场景所述的方法。
本领域技术人员可以理解附图只是一个优选实施场景的示意图,附图中的模块或流程并不一定是实施本申请所必须的。
本领域技术人员可以理解实施场景中的装置中的模块可以按照实施场景描述进行分布于实施场景的装置中,也可以进行相应变化位于不同于本实施场景的一个或多个装置中。上述实施场景的模块可以合并为一个模块,也可以进一步拆分成多个子模块。
上述本申请序号仅仅为了描述,不代表实施场景的优劣。
以上公开的仅为本申请的几个具体实施场景,但是,本申请并非局限于此,任何本领域的技术人员能思之的变化都应落入本申请的保护范围。

Claims (32)

  1. 一种基于云监控的负载均衡优化方法,其特征在于,包括:
    获取负载均衡系统控制地域内的流量分布统计结果;及
    获取所述负载均衡系统中可用区内的后端服务器统计结果和可用区的网络监控质量;
    根据所述流量分布统计结果、所述后端服务器统计结果和所述网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区;
    按照确定结果进行所述负载均衡实例的主备可用区切换。
  2. 根据权利要求1所述的方法,其特征在于,所述根据所述流量分布统计结果、所述后端服务器统计结果和所述网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区,具体包括:
    根据所述流量分布统计结果,获取所述负载均衡系统中各个可用区的出口带宽和所述各个可用区分别对应的公网出口使用带宽平均值,并依据所述出口带宽和所述公网出口使用带宽平均值,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值;
    根据所述后端服务器统计结果,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值;
    根据所述网络监控质量,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值;
    根据所述第一推荐值、所述第二推荐值、所述第三推荐值以及三者各自对应的权值,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果;
    按照所述推荐值结果,确定适合作为负载均衡实例对应主可用区的可用区。
  3. 根据权利要求2所述的方法,其特征在于,所述依据所述出口带宽和所述公网出口使用带宽平均值,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值,具体包括:
    将可用区对应的公网出口使用带宽平均值除以可用区的出口带宽,得到可用区的流量使用率;
    将可用区的总带宽乘以预设协调因子系数得到可用区对应的权值;
    按照所述流量使用率越低推荐值越高、且所述权值越大推荐值越高的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值。
  4. 根据权利要求2所述的方法,其特征在于,所述获取所述负载均衡系统中可用区内的后端服务器统计结果,具体包括:
    通过对后端服务器的健康检查,确定可用区中处于预设正常条件状态的后端服务器的数量;
    根据所述数量确定所述负载均衡系统中可用区内的后端服务器统计结果;
    所述根据所述后端服务器统计结果,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值,具体包括:
    按照所述数量越多推荐值越高的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值。
  5. 根据权利要求2所述的方法,其特征在于,所述根据所述网络监控质量,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值,具体包括:
    根据所述网络监控质量的运营数据,统计可用区的网络质量参数在运营时间内出现异常的次数;
    按照异常次数越大推荐值越小的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值。
  6. 根据权利要求2所述的方法,其特征在于,所述根据所述第一推荐值、所述第二推荐值、所述第三推荐值以及三者各自对应的权值,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果,具体包括:
    将所述第一推荐值乘以相应项的权值,加上所述第二推荐值乘以相应项的权值,再加上所述第三推荐值乘以相应项的权值,得到的和值确定为所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果。
  7. 根据权利要求2所述的方法,其特征在于,所述按照所述推荐值结果,确定适合作为负载均衡实例对应主可用区的可用区,具体包括:
    将推荐值结果最高的可用区、或推荐值结果大于预置推荐值阈值的可用区,确定为适合作为负载均衡实例对应主可用区的可用区。
  8. 根据权利要求1所述的方法,其特征在于,所述按照确定结果进行所述负载均衡实例的主备可用区切换之前,所述方法还包括:
    通过配置静态等价路由的方式,从与所述负载均衡实例所属的负载均衡设备预先发布网段路由到所述负载均衡设备相应的路由器上;
    所述按照确定结果进行所述负载均衡实例的主备可用区切换,具体包括:
    通过调整所述路由器在发布与所述负载均衡实例对应的32位路由到骨干网络路由器时附带的优先级数值,实现将所述负载均衡实例的主可用区切换为确定结果中的可用区,其中不同可用区的骨干网络路由器之间通过资料处理中心互联网络DCI通道进行连接。
  9. 一种基于云监控的负载均衡优化装置,其特征在于,包括:
    获取单元,用于获取负载均衡系统控制地域内的流量分布统计结果;及
    获取所述负载均衡系统中可用区内的后端服务器统计结果和可用区的网络监控质量;
    确定单元,用于根据所述流量分布统计结果、所述后端服务器统计结果和所述网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区;
    切换单元,用于按照确定结果进行所述负载均衡实例的主备可用区切换。
  10. 根据权利要求9所述的装置,其特征在于,所述确定单元,具体包括:获取模块、计算模块、确定模块;
    获取模块,用于根据所述流量分布统计结果,获取所述负载均衡系统中各个可用区的出口带宽和所述各个可用区分别对应的公网出口使用带宽平均值,并依据所述出口带宽和所述公网出口使用带宽平均值,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值;
    计算模块,用于根据所述后端服务器统计结果,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值;
    根据所述网络监控质量,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值;
    确定模块,用于根据所述第一推荐值、所述第二推荐值、所述第三推荐值以及三者各自对应的权值,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果;
    按照所述推荐值结果,确定适合作为负载均衡实例对应主可用区的可用区。
  11. 根据权利要求10所述的装置,其特征在于,
    所述计算模块,具体用于将可用区对应的公网出口使用带宽平均值除以可用区的出口带宽,得到可用区的流量使用率;
    将可用区的总带宽乘以预设协调因子系数得到可用区对应的权值;
    按照所述流量使用率越低推荐值越高、且所述权值越大推荐值越高的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值。
  12. 根据权利要求10所述的装置,其特征在于,
    所述获取单元,具体用于通过对后端服务器的健康检查,确定可用区中处于预设正常条件状态的后端服务器的数量;
    根据所述数量确定所述负载均衡系统中可用区内的后端服务器统计结果;
    所述计算模块,具体还用于按照所述数量越多推荐值越高的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值。
  13. 根据权利要求10所述的装置,其特征在于,
    所述计算模块,具体还用于根据所述网络监控质量的运营数据,统计可用区的网络质量参数在运营时间内出现异常的次数;
    按照异常次数越大推荐值越小的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值。
  14. 根据权利要求10所述的装置,其特征在于,
    所述确定模块,具体用于将所述第一推荐值乘以相应项的权值,加上所述第二推荐值乘以相应项的权值,再加上所述第三推荐值乘以相应项的权值,得到的和值确定为所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果。
  15. 根据权利要求10所述的装置,其特征在于,
    所述确定模块,具体用于将推荐值结果最高的可用区、或推荐值结果大于预置推荐值阈值的可用区,确定为适合作为负载均衡实例对应主可用区的可用区。
  16. 根据权利要求9所述的装置,其特征在于,所述装置还包括:发布单元;
    所述发布单元,用于通过配置静态等价路由的方式,从与所述负载均衡实例所属的负载均衡设备预先发布网段路由到所述负载均衡设备相应的路由器上;
    所述切换单元,具体用于通过调整所述路由器在发布与所述负载均衡实例对应的32位路由到骨干网络路由器时附带的优先级数值,实现将所述负载均衡实例的主可用区切换为确定结果中的可用区,其中不同可用区的骨干网络路由器之间通过资料处理中心互联网络DCI通道进行连接。
  17. 一种存储设备,其上存储有计算机程序,其特征在于,所述程序被处理器执行时实现基于云监控的负载均衡优化方法,包括:
    获取负载均衡系统控制地域内的流量分布统计结果;及
    获取所述负载均衡系统中可用区内的后端服务器统计结果和可用区的网络监控质量;
    根据所述流量分布统计结果、所述后端服务器统计结果和所述网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区;
    按照确定结果进行所述负载均衡实例的主备可用区切换。
  18. 根据权利要求17所述的存储设备,其特征在于,所述程序被处理器执行时实现所述根据所述流量分布统计结果、所述后端服务器统计结果和所述网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区,具体包括:
    根据所述流量分布统计结果,获取所述负载均衡系统中各个可用区的出口带宽和所述各个可用区分别对应的公网出口使用带宽平均值,并依据所述出口带宽和所述公网出口使用带宽平均值,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值;
    根据所述后端服务器统计结果,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值;
    根据所述网络监控质量,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值;
    根据所述第一推荐值、所述第二推荐值、所述第三推荐值以及三者各自对应的权值,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果;
    按照所述推荐值结果,确定适合作为负载均衡实例对应主可用区的可用区。
  19. 根据权利要求18所述的存储设备,其特征在于,所述程序被处理器执行时实现所述依据所述出口带宽和所述公网出口使用带宽平均值,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值,具体包括:
    将可用区对应的公网出口使用带宽平均值除以可用区的出口带宽,得到可用区的流量使用率;
    将可用区的总带宽乘以预设协调因子系数得到可用区对应的权值;
    按照所述流量使用率越低推荐值越高、且所述权值越大推荐值越高的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值。
  20. 根据权利要求18所述的存储设备,其特征在于,所述程序被处理器执行时实现所述获取所述负载均衡系统中可用区内的后端服务器统计结果,具体包括:
    通过对后端服务器的健康检查,确定可用区中处于预设正常条件状态的后端服务器的数量;
    根据所述数量确定所述负载均衡系统中可用区内的后端服务器统计结果;
    所述根据所述后端服务器统计结果,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值,具体包括:
    按照所述数量越多推荐值越高的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值。
  21. 根据权利要求18所述的存储设备,其特征在于,所述程序被处理器执行时实现所述根据所述网络监控质量,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值,具体包括:
    根据所述网络监控质量的运营数据,统计可用区的网络质量参数在运营时间内出现异常的次数;
    按照异常次数越大推荐值越小的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值。
  22. 根据权利要求18所述的存储设备,其特征在于,所述程序被处理器执行时实现所述根据所述第一推荐值、所述第二推荐值、所述第三推荐值以及三者各自对应的权值,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果,具体包括:
    将所述第一推荐值乘以相应项的权值,加上所述第二推荐值乘以相应项的权值,再加上所述第三推荐值乘以相应项的权值,得到的和值确定为所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果。
  23. 根据权利要求18所述的存储设备,其特征在于,所述程序被处理器执行时实现所述按照所述推荐值结果,确定适合作为负载均衡实例对应主可用区的可用区,具体包括:
    将推荐值结果最高的可用区、或推荐值结果大于预置推荐值阈值的可用区,确定为适合作为负载均衡实例对应主可用区的可用区。
  24. 根据权利要求17所述的存储设备,其特征在于,所述程序被处理器执行时实现所述按照确定结果进行所述负载均衡实例的主备可用区切换之前,还包括:
    通过配置静态等价路由的方式,从与所述负载均衡实例所属的负载均衡设备预先发布网段路由到所述负载均衡设备相应的路由器上;
    所述程序被处理器执行时实现所述按照确定结果进行所述负载均衡实例的主备可用区切换,具体包括:
    通过调整所述路由器在发布与所述负载均衡实例对应的32位路由到骨干网络路由器时附带的优先级数值,实现将所述负载均衡实例的主可用区切换为确定结果中的可用区,其中不同可用区的骨干网络路由器之间通过资料处理中心互联网络DCI通道进行连接。
  25. 一种基于云监控的负载均衡优化装置,包括存储设备、处理器及存储在存储设备上并可在处理器上运行的计算机程序,其特征在于,所述处理器执行所述程序时实现基于云监控的负载均衡优化方法,包括:
    获取负载均衡系统控制地域内的流量分布统计结果;及
    获取所述负载均衡系统中可用区内的后端服务器统计结果和可用区的网络监控质量;
    根据所述流量分布统计结果、所述后端服务器统计结果和所述网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区;
    按照确定结果进行所述负载均衡实例的主备可用区切换。
  26. 根据权利要求25所述的装置,其特征在于,所述处理器执行所述程序时实现所述根据所述流量分布统计结果、所述后端服务器统计结果和所述网络监控质量,确定适合作为负载均衡实例对应主可用区的可用区,具体包括:
    根据所述流量分布统计结果,获取所述负载均衡系统中各个可用区的出口带宽和所述各个可用区分别对应的公网出口使用带宽平均值,并依据所述出口带宽和所述公网出口使用带宽平均值,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值;
    根据所述后端服务器统计结果,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值;
    根据所述网络监控质量,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值;
    根据所述第一推荐值、所述第二推荐值、所述第三推荐值以及三者各自对应的权值,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果;
    按照所述推荐值结果,确定适合作为负载均衡实例对应主可用区的可用区。
  27. 根据权利要求26所述的装置,其特征在于,所述处理器执行所述程序时实现所述依据所述出口带宽和所述公网出口使用带宽平均值,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值,具体包括:
    将可用区对应的公网出口使用带宽平均值除以可用区的出口带宽,得到可用区的流量使用率;
    将可用区的总带宽乘以预设协调因子系数得到可用区对应的权值;
    按照所述流量使用率越低推荐值越高、且所述权值越大推荐值越高的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第一推荐值。
  28. 根据权利要求26所述的装置,其特征在于,所述处理器执行所述程序时实现所述获取所述负载均衡系统中可用区内的后端服务器统计结果,具体包括:
    通过对后端服务器的健康检查,确定可用区中处于预设正常条件状态的后端服务器的数量;
    根据所述数量确定所述负载均衡系统中可用区内的后端服务器统计结果;
    所述根据所述后端服务器统计结果,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值,具体包括:
    按照所述数量越多推荐值越高的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第二推荐值。
  29. 根据权利要求26所述的装置,其特征在于,所述处理器执行所述程序时实现所述根据所述网络监控质量,计算所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值,具体包括:
    根据所述网络监控质量的运营数据,统计可用区的网络质量参数在运营时间内出现异常的次数;
    按照异常次数越大推荐值越小的原则,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的第三推荐值。
  30. 根据权利要求26所述的装置,其特征在于,所述处理器执行所述程序时实现所述根据所述第一推荐值、所述第二推荐值、所述第三推荐值以及三者各自对应的权值,确定所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果,具体包括:
    将所述第一推荐值乘以相应项的权值,加上所述第二推荐值乘以相应项的权值,再加上所述第三推荐值乘以相应项的权值,得到的和值确定为所述负载均衡系统中各个可用区分别作为负载均衡实例对应主可用区的推荐值结果。
  31. 根据权利要求26所述的装置,其特征在于,所述处理器执行所述程序时实现所述按照所述推荐值结果,确定适合作为负载均衡实例对应主可用区的可用区,具体包括:
    将推荐值结果最高的可用区、或推荐值结果大于预置推荐值阈值的可用区,确定为适合作为负载均衡实例对应主可用区的可用区。
  32. 根据权利要求25所述的装置,其特征在于,所述处理器执行所述程序时实现所述按照确定结果进行所述负载均衡实例的主备可用区切换之前,还包括:
    通过配置静态等价路由的方式,从与所述负载均衡实例所属的负载均衡设备预先发布网段路由到所述负载均衡设备相应的路由器上;
    所述处理器执行所述程序时实现所述按照确定结果进行所述负载均衡实例的主备可用区切换,具体包括:
    通过调整所述路由器在发布与所述负载均衡实例对应的32位路由到骨干网络路由器时附带的优先级数值,实现将所述负载均衡实例的主可用区切换为确定结果中的可用区,其中不同可用区的骨干网络路由器之间通过资料处理中心互联网络DCI通道进行连接。
PCT/CN2017/107480 2017-09-28 2017-10-24 基于云监控的负载均衡优化方法及装置 WO2019061615A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
JP2018517171A JP6671468B2 (ja) 2017-09-28 2017-10-24 クラウドモニタリングに基づく負荷分散の最適化方法および装置
EP17882263.1A EP3490202B1 (en) 2017-09-28 2017-10-24 Cloud monitoring-based load balancing optimization method and device
US16/075,327 US10992581B2 (en) 2017-09-28 2017-10-24 Methods and devices for optimizing load balancing based on cloud monitoring
KR1020187035327A KR102157722B1 (ko) 2017-09-28 2017-10-24 클라우드 모니터링을 기반으로 한 부하 분산 최적화 방법 및 장치
AU2017404564A AU2017404564B2 (en) 2017-09-28 2017-10-24 Methods and devices for optimizing load balancing based on cloud monitoring

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710903413.4 2017-09-28
CN201710903413.4A CN107689925B (zh) 2017-09-28 2017-09-28 基于云监控的负载均衡优化方法及装置

Publications (1)

Publication Number Publication Date
WO2019061615A1 true WO2019061615A1 (zh) 2019-04-04

Family

ID=61155453

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/107480 WO2019061615A1 (zh) 2017-09-28 2017-10-24 基于云监控的负载均衡优化方法及装置

Country Status (7)

Country Link
US (1) US10992581B2 (zh)
EP (1) EP3490202B1 (zh)
JP (1) JP6671468B2 (zh)
KR (1) KR102157722B1 (zh)
CN (1) CN107689925B (zh)
AU (1) AU2017404564B2 (zh)
WO (1) WO2019061615A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113766032A (zh) * 2021-09-14 2021-12-07 烽火通信科技股份有限公司 转控分离架构vbras的业务保护方法及系统
CN114615177A (zh) * 2022-03-03 2022-06-10 腾讯科技(深圳)有限公司 一种云平台的负载检测方法、装置、电子设备和存储介质

Families Citing this family (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11010233B1 (en) 2018-01-18 2021-05-18 Pure Storage, Inc Hardware-based system monitoring
CN110572453A (zh) * 2019-09-06 2019-12-13 杭州涂鸦信息技术有限公司 一种同可用区加权负载均衡的方法
US11657155B2 (en) * 2019-11-22 2023-05-23 Pure Storage, Inc Snapshot delta metric based determination of a possible ransomware attack against data maintained by a storage system
US11720714B2 (en) * 2019-11-22 2023-08-08 Pure Storage, Inc. Inter-I/O relationship based detection of a security threat to a storage system
US11625481B2 (en) * 2019-11-22 2023-04-11 Pure Storage, Inc. Selective throttling of operations potentially related to a security threat to a storage system
US11687418B2 (en) 2019-11-22 2023-06-27 Pure Storage, Inc. Automatic generation of recovery plans specific to individual storage elements
US11675898B2 (en) * 2019-11-22 2023-06-13 Pure Storage, Inc. Recovery dataset management for security threat monitoring
US11720692B2 (en) * 2019-11-22 2023-08-08 Pure Storage, Inc. Hardware token based management of recovery datasets for a storage system
US11651075B2 (en) * 2019-11-22 2023-05-16 Pure Storage, Inc. Extensible attack monitoring by a storage system
US11341236B2 (en) * 2019-11-22 2022-05-24 Pure Storage, Inc. Traffic-based detection of a security threat to a storage system
US11615185B2 (en) * 2019-11-22 2023-03-28 Pure Storage, Inc. Multi-layer security threat detection for a storage system
US11520907B1 (en) * 2019-11-22 2022-12-06 Pure Storage, Inc. Storage system snapshot retention based on encrypted data
US11645162B2 (en) * 2019-11-22 2023-05-09 Pure Storage, Inc. Recovery point determination for data restoration in a storage system
US11755751B2 (en) * 2019-11-22 2023-09-12 Pure Storage, Inc. Modify access restrictions in response to a possible attack against data stored by a storage system
US20210382992A1 (en) * 2019-11-22 2021-12-09 Pure Storage, Inc. Remote Analysis of Potentially Corrupt Data Written to a Storage System
US11941116B2 (en) * 2019-11-22 2024-03-26 Pure Storage, Inc. Ransomware-based data protection parameter modification
CN110995868A (zh) * 2019-12-20 2020-04-10 紫光云(南京)数字技术有限公司 一种负载均衡方法及其装置
CN111464607B (zh) * 2020-03-27 2023-04-07 中移雄安信息通信科技有限公司 数据备份方法、装置、设备及介质
CN113254205B (zh) * 2021-05-24 2023-08-15 北京百度网讯科技有限公司 负载均衡系统、方法、装置、电子设备及存储介质
US11671486B1 (en) * 2021-08-02 2023-06-06 Amazon Technologies, Inc. Managing availability zone utilizing redundancy validation
US20230169355A1 (en) * 2021-12-01 2023-06-01 Capital One Services, Llc Replica reliability
CN114844951B (zh) * 2022-04-22 2024-03-19 百果园技术(新加坡)有限公司 请求处理方法、系统、设备、存储介质及产品
US11971791B2 (en) * 2022-08-12 2024-04-30 Capital One Services, Llc Automated regional failover

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103401948A (zh) * 2013-08-21 2013-11-20 中广有线信息网络有限公司 一种云宽带智能缓存平台
CN106330743A (zh) * 2015-06-29 2017-01-11 中兴通讯股份有限公司 一种流量均衡度度量的方法及装置
CN106488507A (zh) * 2015-08-27 2017-03-08 四川师范大学 路由负载分流的非均匀分簇传感器网络能量空洞避免方法

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002019131A1 (en) * 2000-09-01 2002-03-07 Op40, Inc. System and method for collaboration using web browsers
US8607035B2 (en) * 2008-08-29 2013-12-10 Texas Instruments Incorporated Multi-core processing utilizing prioritized interrupts for optimization
US20100223364A1 (en) * 2009-02-27 2010-09-02 Yottaa Inc System and method for network traffic management and load balancing
US8819701B2 (en) * 2009-12-12 2014-08-26 Microsoft Corporation Cloud computing monitoring and management system
US8606938B1 (en) * 2012-09-27 2013-12-10 Ringcentral, Inc. High availability for cloud-based services
US20160149763A1 (en) * 2013-11-27 2016-05-26 Carl B Ingram Systems and Methods for Providing an Administrative Framework in a Cloud Architecture
JP6097235B2 (ja) * 2014-02-19 2017-03-15 西日本電信電話株式会社 負荷分散システム、負荷分散装置及び負荷分散方法
JP6237318B2 (ja) * 2014-02-19 2017-11-29 富士通株式会社 管理装置、業務負荷分散管理方法および業務負荷分散管理プログラム
JP5778815B1 (ja) 2014-03-24 2015-09-16 株式会社野村総合研究所 基盤運用管理システムおよび基盤運用管理方法
WO2015197564A1 (en) * 2014-06-23 2015-12-30 Getclouder Ltd. Cloud hosting systems featuring scaling and load balancing with containers
CN104283948B (zh) * 2014-09-26 2018-12-07 东软集团股份有限公司 服务器集群系统及其负载均衡实现方法
EP3234774B1 (en) * 2014-11-13 2021-01-27 Telefonaktiebolaget LM Ericsson (publ) Elasticity for highly availabile applications
PL3941106T3 (pl) * 2016-03-18 2023-11-20 Plume Design, Inc. Oparte na chmurze sterowanie sieci wi-fi
EP3497843A4 (en) * 2016-08-11 2019-12-18 Hopzero, Inc. METHOD AND SYSTEM FOR LIMITING THE RANGE OF DATA TRANSMISSIONS
CN106469091B (zh) * 2016-09-05 2018-05-18 北京百度网讯科技有限公司 用于分配可用区的方法和装置
CN107196869B (zh) * 2017-07-14 2019-11-15 北京知道创宇信息技术股份有限公司 基于主机实际负载的自适应负载均衡方法、设备和系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103401948A (zh) * 2013-08-21 2013-11-20 中广有线信息网络有限公司 一种云宽带智能缓存平台
CN106330743A (zh) * 2015-06-29 2017-01-11 中兴通讯股份有限公司 一种流量均衡度度量的方法及装置
CN106488507A (zh) * 2015-08-27 2017-03-08 四川师范大学 路由负载分流的非均匀分簇传感器网络能量空洞避免方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3490202A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113766032A (zh) * 2021-09-14 2021-12-07 烽火通信科技股份有限公司 转控分离架构vbras的业务保护方法及系统
CN114615177A (zh) * 2022-03-03 2022-06-10 腾讯科技(深圳)有限公司 一种云平台的负载检测方法、装置、电子设备和存储介质
CN114615177B (zh) * 2022-03-03 2023-10-13 腾讯科技(深圳)有限公司 一种云平台的负载检测方法、装置、电子设备和存储介质

Also Published As

Publication number Publication date
AU2017404564B2 (en) 2019-11-14
AU2017404564A1 (en) 2019-04-11
CN107689925A (zh) 2018-02-13
US20210036957A1 (en) 2021-02-04
JP6671468B2 (ja) 2020-03-25
JP2019533913A (ja) 2019-11-21
US10992581B2 (en) 2021-04-27
KR20190038760A (ko) 2019-04-09
EP3490202B1 (en) 2020-05-13
CN107689925B (zh) 2020-01-14
EP3490202A4 (en) 2019-09-18
KR102157722B1 (ko) 2020-09-21
EP3490202A1 (en) 2019-05-29

Similar Documents

Publication Publication Date Title
WO2019061615A1 (zh) 基于云监控的负载均衡优化方法及装置
US11606699B2 (en) Scalable evolved packet core
Ghaznavi et al. Elastic virtual network function placement
JP5537600B2 (ja) 制御ノード及び通信制御方法
Yang et al. Seamless support of low latency mobile applications with nfv-enabled mobile edge-cloud
WO2014104634A1 (en) System and method for dynamically expanding virtual cluster and recording medium on which program for executing the method is recorded
US20160359628A1 (en) Unique id generation for sensors
WO2014032262A1 (zh) 一种平台间迁移应用的实现方法及系统
WO2016121973A1 (ja) ノードシステム、サーバ装置、スケールリング制御方法及びプログラム
US9705704B2 (en) Method and system of forming a mobile virtual network
Beshley et al. Experimental performance analysis of software-defined network switch and controller
Ma et al. Profit maximization for admitting requests with network function services in distributed clouds
WO2015192584A1 (zh) 虚拟路由系统及方法
Zhang et al. Tuning the aggressive TCP behavior for highly concurrent HTTP connections in intra-datacenter
WO2012095053A9 (zh) 数据迁移方法和设备
EP3082013A1 (en) Method for extending hold time of power supply units
WO2019221377A1 (ko) 블록체인 기반 트랜잭션 관리 시스템 및 그 동작 방법
JPWO2012133300A1 (ja) 仮想デスクトップシステム、ネットワーク処理装置、管理方法、及び管理プログラム
WO2018233372A1 (zh) 支持哈希调度策略的lvs调度方法、设备、装置及存储介质
WO2012065440A1 (zh) 虚拟路由器冗余协议备份组中设备优先级实现方法及装置
US20130054493A1 (en) Routing traffic after power failure
WO2018120822A1 (zh) 一种用户数据锚点迁移的方法、设备和系统
JP2014041414A (ja) 仮想マシンのマイグレーションシステム、制御装置およびマイグレーション方法
JP6835683B2 (ja) 分散方法
WO2016153119A1 (ko) 폐쇄형 데이터 공유 서비스 데이터 관리 방법 및 시스템

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2018517171

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 2017882263

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2017882263

Country of ref document: EP

Effective date: 20180626

ENP Entry into the national phase

Ref document number: 20187035327

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2017404564

Country of ref document: AU

Date of ref document: 20171024

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE