WO2020062439A1 - 一种监控带宽状态的方法和装置 - Google Patents

一种监控带宽状态的方法和装置 Download PDF

Info

Publication number
WO2020062439A1
WO2020062439A1 PCT/CN2018/113734 CN2018113734W WO2020062439A1 WO 2020062439 A1 WO2020062439 A1 WO 2020062439A1 CN 2018113734 W CN2018113734 W CN 2018113734W WO 2020062439 A1 WO2020062439 A1 WO 2020062439A1
Authority
WO
WIPO (PCT)
Prior art keywords
bandwidth
bandwidth data
domain name
target domain
reference period
Prior art date
Application number
PCT/CN2018/113734
Other languages
English (en)
French (fr)
Inventor
蒋龙威
夏东杰
Original Assignee
网宿科技股份有限公司
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 网宿科技股份有限公司 filed Critical 网宿科技股份有限公司
Priority to US16/959,128 priority Critical patent/US10965565B2/en
Priority to EP18935279.2A priority patent/EP3860052A4/en
Publication of WO2020062439A1 publication Critical patent/WO2020062439A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0876—Network utilisation, e.g. volume of load or congestion level
    • H04L43/0894—Packet rate
    • 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
    • H04L41/142—Network analysis or design using statistical or mathematical methods
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/12—Network monitoring probes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00—Network arrangements, protocols or services for addressing or naming
    • H04L61/30—Managing network names, e.g. use of aliases or nicknames
    • H04L61/3015—Name registration, generation or assignment
    • H04L61/3025—Domain name generation or assignment
    • 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
    • H04L41/5019—Ensuring fulfilment of SLA

Definitions

  • the present invention relates to the field of network communication technologies, and in particular, to a method and device for monitoring a bandwidth state.
  • CDN acceleration services In order to improve the feedback speed for user access, and at the same time ensure the stability and high availability of business back-end systems, service providers mostly choose CDN acceleration services to implement related business services. From the perspective of the CDN provider, the CDN provider often allocates CDN bandwidth resources to each CDN accelerated service provider (which can be called a customer) to implement CDN accelerated services.
  • embodiments of the present invention provide a method and device for monitoring a bandwidth state.
  • the technical solution is as follows:
  • a method for monitoring a bandwidth state includes:
  • the obtaining the average bandwidth data value of the target domain name in the monitoring reference period in the current statistical period and multiple historical statistical periods includes:
  • calculating and storing the average bandwidth data of the target domain name in the monitoring reference period based on the real-time bandwidth data includes:
  • the recalculating and storing the average value of the bandwidth data of the target domain name in the monitoring reference period according to the real-time bandwidth data remaining after culling further includes:
  • determining the bandwidth dynamic interval corresponding to the monitoring reference period according to the average value of the bandwidth data and a preset confidence level includes:
  • bandwidth data averages in the monitoring reference period excluding a bandwidth data average whose value is outside the range of the average fluctuation
  • a bandwidth dynamic interval corresponding to the monitoring reference period is determined according to the average value of the bandwidth data remaining after the culling and a preset confidence level.
  • determining the bandwidth dynamic interval corresponding to the monitoring reference period according to the average value of the bandwidth data remaining after the culling and a preset confidence level includes:
  • the bandwidth dynamic interval corresponding to the monitoring reference period is determined according to the average value of the bandwidth data remaining after culling, a preset period weight corresponding to each of the bandwidth data averages, and a preset confidence level.
  • the method before the obtaining the average bandwidth data of the target domain name in the monitoring reference period in the current statistical period and multiple historical statistical periods, the method further includes:
  • an apparatus for monitoring a bandwidth state includes:
  • a first determining module configured to determine a monitoring reference period of bandwidth data of a target domain name every preset monitoring interval duration
  • An obtaining module configured to obtain an average bandwidth data value of the target domain name in the monitoring reference period in the current statistical period and multiple historical statistical periods;
  • a second determining module configured to determine a bandwidth dynamic interval corresponding to the monitoring reference period according to the average value of the bandwidth data and a preset confidence level
  • a monitoring module is configured to monitor the current bandwidth status of the target domain name according to the current bandwidth data of the target domain name and the bandwidth dynamic interval.
  • the obtaining module is specifically configured to:
  • the obtaining module is specifically configured to:
  • the obtaining module is further configured to: recalculate and store the data in the monitoring reference period according to the real-time bandwidth data remaining after the elimination and a preset time weight corresponding to each of the real-time bandwidth data.
  • the average bandwidth data of the target domain name is further configured to: recalculate and store the data in the monitoring reference period according to the real-time bandwidth data remaining after the elimination and a preset time weight corresponding to each of the real-time bandwidth data.
  • the average bandwidth data of the target domain name is further configured to: recalculate and store the data in the monitoring reference period according to the real-time bandwidth data remaining after the elimination and a preset time weight corresponding to each of the real-time bandwidth data.
  • the second determining module is specifically configured to:
  • bandwidth data averages in the monitoring reference period excluding a bandwidth data average whose value is outside the range of the average fluctuation
  • a bandwidth dynamic interval corresponding to the monitoring reference period is determined according to the average value of the bandwidth data remaining after the culling and a preset confidence level.
  • the second determining module is specifically configured to:
  • the bandwidth dynamic interval corresponding to the monitoring reference period is determined according to the average value of the bandwidth data remaining after culling, a preset period weight corresponding to each of the bandwidth data averages, and a preset confidence level.
  • the apparatus further includes a third determining module, configured to:
  • a bandwidth monitoring device includes a processor and a memory, and the memory stores at least one instruction, at least one program, a code set, or an instruction set. The at least one piece of program, the code set or the instruction set is loaded and executed by the processor to implement the method for monitoring a bandwidth state according to the first aspect.
  • a computer-readable storage medium stores at least one instruction, at least one program, code set, or instruction set, the at least one instruction, the at least one program, and the code.
  • the set or instruction set is loaded and executed by the processor to implement the method of monitoring a bandwidth state as described in the first aspect.
  • the monitoring reference period of the bandwidth data of the target domain name is determined every preset monitoring interval duration; the average value of the bandwidth data of the target domain name in the monitoring reference period in the current statistical period and multiple historical statistical periods is obtained; according to the average of the bandwidth data And the preset confidence level, determine the bandwidth dynamic interval corresponding to the monitoring reference period; and monitor the bandwidth status of the current target domain name according to the current bandwidth data and bandwidth dynamic interval of the target domain name.
  • the historical bandwidth data is analyzed to determine a reasonable dynamic bandwidth interval, so that before the abnormal accumulation of bandwidth status affects the quality of service, abnormal bandwidth status can be found in time through the dynamic bandwidth interval and triggered.
  • the follow-up alarm or repair process helps to repair the abnormal state before the business party perceives the abnormal bandwidth, thereby ensuring the overall service quality of the customer.
  • FIG. 1 is a schematic diagram of a scenario architecture of a CDN system according to an embodiment of the present invention
  • FIG. 2 is a flowchart of a method for monitoring a bandwidth state according to an embodiment of the present invention
  • FIG. 3 is a logic principle diagram for determining a bandwidth dynamic interval according to an embodiment of the present invention.
  • FIG. 4 is a schematic structural diagram of an apparatus for monitoring a bandwidth state according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of an apparatus for monitoring a bandwidth status according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of a bandwidth monitoring device according to an embodiment of the present invention.
  • An embodiment of the present invention provides a method for monitoring a bandwidth state.
  • the method can be applied to a CDN system, and can be specifically implemented by a bandwidth monitoring device in the CDN system.
  • the bandwidth monitoring device may be a device in the CDN system for monitoring the bandwidth status of customers enjoying CDN accelerated services. It can monitor the bandwidth status of the customers through the obtained bandwidth data of the customers, and alarms based on the monitoring results on abnormal bandwidths. Or trigger preset exception handling.
  • the above-mentioned customers' bandwidth data collection and storage can be implemented by a bandwidth monitoring device, or can be separately completed by a data acquisition device and a data storage device in a CDN system.
  • the specific scenario architecture of the CDN system can be shown in Figure 1.
  • the above-mentioned bandwidth monitoring device may include a processor, a memory, and a transceiver.
  • the processor may be used to perform processing of monitoring the bandwidth status in the following flow.
  • the memory may be used to store the data required during the following processing and the generated data, and send and receive.
  • the device can be used to receive and send related data during the processing described below.
  • the above-mentioned processing of monitoring bandwidth status, collecting and storing bandwidth data can all be implemented by a distributed system deployed inside the CDN system. In this embodiment, "the process of monitoring the bandwidth status, collecting and storing the bandwidth data is all performed by the bandwidth monitoring device" as an example for description, and other situations are similar, and will not be described again.
  • Step 201 Determine a monitoring reference period of bandwidth data of a target domain name every preset monitoring interval.
  • the bandwidth monitoring device can monitor the bandwidth status of the target domain name every preset monitoring interval. Specifically, each time the monitoring is performed, the bandwidth monitoring device may first determine a monitoring reference period of the bandwidth data of the target domain name.
  • the monitoring reference period may be a time period covered by all bandwidth data that needs to be referred to when monitoring the bandwidth status.
  • the monitoring interval duration and the monitoring reference period may be preset by a technician of the CDN system according to the bandwidth monitoring requirements.
  • the monitoring interval can be 1 minute, that is, the bandwidth status of the target domain name is monitored every other minute.
  • the monitoring reference period can be 30 minutes before the current time, that is, the bandwidth data of the first 30 minutes must be referenced during each monitoring.
  • Step 202 Obtain the average bandwidth data of the target domain name in the monitoring reference period in the current statistical period and multiple historical statistical periods.
  • the bandwidth monitoring device when starting to monitor the bandwidth status, can determine the statistical period (that is, the current statistical period) to which the current moment belongs, as well as multiple historical statistical periods before the current statistical period. Then, after the monitoring reference period is determined, To obtain the average bandwidth data of the target domain name in the monitoring reference period in the current statistical period and multiple historical statistical periods.
  • the statistical period here may be preset by a technician of the CDN system, and a statistical period may be one day or one week. It is worth mentioning that for the case where the monitoring reference period spans the statistical period, it can be considered that the monitoring reference period belongs to the latter statistical period. For example, the monitoring reference period is 30 minutes and the statistical period is 1 day. Then on January 2 at 00 The monitoring reference period determined before: 30 spans two statistical periods, January 1 and January 2. These monitoring reference periods can still be considered as the monitoring reference period on January 2. The same applies to the monitoring disclosed in this embodiment. Method of bandwidth status.
  • the bandwidth monitoring device may reduce the calculation amount for each monitoring of the bandwidth status by processing the average value of the cached bandwidth data.
  • the processing in step 202 may be as follows: Obtain the data of each specified collection time in the monitoring reference period in the current statistical cycle. Real-time bandwidth data of the target domain name; Based on the real-time bandwidth data, calculate and store the average value of the bandwidth data of the target domain name in the monitoring reference period; Obtain the average value of the bandwidth data of the target domain name in the monitoring reference period in multiple historical statistical periods stored in advance.
  • the bandwidth monitoring device may collect real-time bandwidth data of the target domain name at the specified collection time at the interface between the business server corresponding to the target domain name and the CDN system, and store the collected real-time bandwidth data locally. In this way, after the monitoring reference period is determined, the bandwidth monitoring device can only obtain real-time bandwidth data of the target domain name at each specified collection time in the monitoring reference period in the current statistical cycle, and then calculate the target in the monitoring reference period based on these real-time bandwidth data
  • the average bandwidth data of the domain name, and the average of the bandwidth data is stored locally. It can be understood that for the average bandwidth data of the target domain name in the monitoring reference period in each historical statistical cycle, the bandwidth monitoring device has calculated and stored it locally during the monitoring process in the historical statistical cycle. Therefore, at this time, the bandwidth monitoring device does not need to be For calculation, the average bandwidth data of the target domain name during the reference period can be monitored directly from multiple historical statistical periods stored in advance.
  • an abnormal value in the real-time bandwidth data may be removed by a Gaussian function first, and the corresponding processing may be as follows: According to the real-time bandwidth data, calculate the average bandwidth data and bandwidth data of the target domain name in the monitoring reference period. Standard deviation; determine the bandwidth fluctuation range of the target domain name based on the average bandwidth data standard deviation, the bandwidth data standard deviation, and the preset bandwidth fluctuation weight; exclude real-time bandwidth data of the target domain name in the monitoring reference period, and the real-time values outside the bandwidth fluctuation range Bandwidth data; Based on the real-time bandwidth data remaining after the culling, recalculate and store the average bandwidth data of the target domain name in the monitoring reference period.
  • the bandwidth monitoring device can calculate the average bandwidth data standard deviation and the bandwidth data standard deviation of the target domain name in the monitoring reference period based on the real-time bandwidth data of the target domain name during the monitoring reference period in the current statistical cycle.
  • the data standard deviation and the preset bandwidth fluctuation weight determine the bandwidth fluctuation range of the target domain name.
  • the bandwidth monitoring device can exclude all real-time bandwidth data of the target domain name in the monitoring reference period from values that fall outside the above-mentioned bandwidth fluctuation range.
  • the bandwidth monitoring device may recalculate the average bandwidth data value of the target domain name in the monitoring reference period based on the real-time bandwidth data remaining after the elimination, and store the calculated average bandwidth data value locally.
  • the values are x 1-1 , x 1-2 , ... x 1-n , and the average bandwidth data of the n real-time bandwidth data is calculated.
  • the standard deviation is ⁇ x1 (n)
  • the preset bandwidth fluctuation weight is ⁇ 1 , so the bandwidth fluctuation range can be:
  • n real-time bandwidth data there are j real-time bandwidth data whose value is outside the bandwidth fluctuation range, and the average bandwidth data of the remaining nj real-time bandwidth data after excluding is calculated
  • the real-time bandwidth data at different times has different degrees of influence on the current bandwidth data.
  • the processing of calculating the average value of the bandwidth data can be as follows: according to the remaining real-time bandwidth data after culling and presets corresponding to each real-time bandwidth data Time weight, recalculate and store the average bandwidth data of the target domain name during the monitoring reference period.
  • the bandwidth monitoring device can obtain a preset time weight corresponding to each real-time bandwidth data.
  • the preset time weight can be set by a technician of the CDN system based on experience. It is used to determine the influence of the real-time bandwidth data at different times on monitoring the current bandwidth status. It can be understood that the preset time weight of the real-time bandwidth data that is closer to the current time at the time of collection is higher.
  • the bandwidth monitoring device can multiply each real-time bandwidth data remaining after culling and its corresponding preset time weight, and add all the multiplied accumulations to obtain the average value, thereby obtaining the monitoring reference period
  • the average of the bandwidth data of the target domain name is stored, and the average of the bandwidth data is stored.
  • step 202 determine the period type and Multiple historical statistical cycles of the same type as the current statistical cycle and closest to the current statistical cycle.
  • the technicians of the CDN system can divide the statistical cycle into multiple cycle types according to the bandwidth usage of the domain name.
  • the bandwidth usage in all statistical cycles under each cycle type should roughly conform to the same rule, and different cycle types Bandwidth usage during the next statistical week should have poor variance.
  • the statistical period can be divided into holidays and working days, the statistical period can be divided into the beginning of the month, the middle of the month, and the end of the month.
  • the statistical period can also be divided into the business update day, business maintenance day, and Business day and so on.
  • a single statistical cycle can have multiple cycle types under different classification standards.
  • the cycle type on January 25 can be a working day, the end of a month, and a business update day.
  • the bandwidth monitoring device can determine the cycle type of the current statistical cycle, and then determine multiple historical statistical cycles that have the same cycle type as the current statistical cycle and are closest to the current statistical cycle. It should be noted that when determining the historical statistical period, you can weigh the consistency of all period types and the distance from the current statistical period according to requirements. For example, you can set the period type with the highest similarity and greater than 90 within the last three months. %, The 5 most recent historical statistical periods, or you can select the 5 most recent historical statistical periods with the highest type similarity and greater than 60% within the last month.
  • Step 203 Determine a bandwidth dynamic interval corresponding to the monitoring reference period according to the average value of the bandwidth data and a preset confidence level.
  • the bandwidth monitoring device can determine the bandwidth dynamics corresponding to the monitoring reference period based on the average of the bandwidth data and a preset confidence level. Interval. Specifically, if the number of historical statistical periods is m-1, there are m averages of bandwidth data in total, and the values are Calculate the mean value of the average of m bandwidth data Standard deviation is The preset confidence level is 1- ⁇ , and the bandwidth dynamic interval can be:
  • the bandwidth data in the monitoring reference period of the current statistical period and 5 historical statistical periods are selected, and then each of them is calculated separately.
  • the average bandwidth data of 10 pieces of bandwidth data in the statistical period, and the bandwidth dynamic range is determined by the average bandwidth data of 6 statistical periods.
  • the Gaussian function can be used to remove outliers in the average value of all bandwidth data.
  • the processing in step 203 can be as follows: calculate the current statistical period and corresponding to multiple historical statistical periods, the target The mean and standard deviation of the average of multiple bandwidth data of the domain name; determine the mean fluctuation range of the target domain name according to the mean and standard deviation of the multiple bandwidth data mean and the preset mean fluctuation weight; multiple bandwidths in the monitoring reference period Among the data averages, the average value of bandwidth data that is outside the fluctuation range of the average value is excluded; the bandwidth dynamic range corresponding to the monitoring reference period is determined according to the average value of the remaining bandwidth data after removal and the preset confidence level.
  • the bandwidth monitoring device may first calculate the mean and standard deviation of the plurality of bandwidth data averages calculated in step 202, and then may calculate the mean and standard deviation of the plurality of bandwidth data averages.
  • the standard deviation and the preset mean fluctuation weight determine the mean fluctuation range of the target domain name.
  • the bandwidth monitoring device may remove the average value of the bandwidth data whose value is outside the range of the average value among all the average values of the bandwidth data of the target domain name during the monitoring reference period.
  • the bandwidth monitoring device can determine the bandwidth dynamic interval corresponding to the monitoring reference period according to the average value of the remaining bandwidth data after culling and a preset confidence level.
  • Standard deviation is
  • the preset mean fluctuation weight is ⁇ 2
  • the mean fluctuation range can be:
  • bandwidth data averages there are i bandwidth data averages whose values are outside the range of the mean fluctuation, and the average value of the remaining mi bandwidth data averages after excluding is calculated.
  • Standard deviation is The preset confidence level is 1- ⁇ , and the bandwidth dynamic interval can be:
  • the average value of the bandwidth data in different statistical periods has different degrees of influence on the bandwidth dynamic range. Accordingly, the process of determining the bandwidth dynamic range can be as follows: according to the average value of the remaining bandwidth data after the elimination and the corresponding prediction of each bandwidth data average. Set the periodic weight and the preset confidence level to determine the bandwidth dynamic interval corresponding to the monitoring reference period.
  • the bandwidth monitoring device can obtain a preset time weight corresponding to each bandwidth data mean after removing the abnormal bandwidth data mean by a Gaussian function.
  • the preset period weight can be set by a technician of the CDN system based on experience. It is used to determine the influence of the average of the bandwidth data of different statistical periods on monitoring the current bandwidth status. It can be understood that the preset period weight of the average of the bandwidth data average that the corresponding statistical period is closer to the current statistical period is higher. In this way, when determining the bandwidth dynamic interval, the bandwidth monitoring device can determine the bandwidth dynamic interval corresponding to the monitoring reference period according to the remaining bandwidth data average after culling, a preset cycle weight corresponding to each bandwidth data average, and a preset confidence level. .
  • the bandwidth monitoring device may copy multiple copies of the corresponding bandwidth data average according to the ratio of the preset period weight corresponding to each bandwidth data average, and then use the copy The average value of all subsequent bandwidth data is used to determine the corresponding bandwidth dynamic interval.
  • Step 204 Monitor the current bandwidth status of the target domain name according to the current bandwidth data and bandwidth dynamic range of the target domain name.
  • the bandwidth monitoring device can determine whether the current bandwidth data of the target domain name is in the bandwidth dynamic interval. If it is, the bandwidth monitoring device can consider that the current bandwidth status of the target domain name is normal; if it is not, the bandwidth monitoring device can determine that the current bandwidth status of the target domain name is abnormal, which can then trigger subsequent bandwidth exception alarms or preset exception response processing. .
  • the monitoring reference period of the bandwidth data of the target domain name is determined every preset monitoring interval duration; the average value of the bandwidth data of the target domain name in the monitoring reference period in the current statistical period and multiple historical statistical periods is obtained; according to the average of the bandwidth data And the preset confidence level, determine the bandwidth dynamic interval corresponding to the monitoring reference period; and monitor the bandwidth status of the current target domain name according to the current bandwidth data and bandwidth dynamic interval of the target domain name.
  • the historical bandwidth data is analyzed to determine a reasonable dynamic bandwidth interval, so that before the abnormal accumulation of bandwidth status affects the quality of service, abnormal bandwidth status can be found in time through the dynamic bandwidth interval and triggered.
  • the follow-up alarm or repair process helps to repair the abnormal state before the business party perceives the abnormal bandwidth, thereby ensuring the overall service quality of the customer.
  • an embodiment of the present invention further provides an apparatus for monitoring a bandwidth status.
  • the apparatus includes:
  • a first determining module 401 configured to determine a monitoring reference period of bandwidth data of a target domain name every preset monitoring interval duration
  • An obtaining module 402 configured to obtain an average bandwidth data value of the target domain name in the monitoring reference period in the current statistical period and multiple historical statistical periods;
  • a second determining module 403, configured to determine a bandwidth dynamic interval corresponding to the monitoring reference period according to the average value of the bandwidth data and a preset confidence level;
  • the monitoring module 404 is configured to monitor the current bandwidth status of the target domain name according to the current bandwidth data of the target domain name and the bandwidth dynamic interval.
  • the obtaining module 402 is specifically configured to:
  • the obtaining module 402 is specifically configured to:
  • the obtaining module 402 is further configured to:
  • the second determining module 403 is specifically configured to:
  • bandwidth data averages in the monitoring reference period excluding a bandwidth data average whose value is outside the range of the average fluctuation
  • a bandwidth dynamic interval corresponding to the monitoring reference period is determined according to the average value of the bandwidth data remaining after the culling and a preset confidence level.
  • the second determining module 403 is specifically configured to:
  • the bandwidth dynamic interval corresponding to the monitoring reference period is determined according to the average value of the bandwidth data remaining after culling, a preset period weight corresponding to each of the bandwidth data averages, and a preset confidence level.
  • the device further includes a third determining module 405, configured to:
  • the monitoring reference period of the bandwidth data of the target domain name is determined every preset monitoring interval duration; the average value of the bandwidth data of the target domain name in the monitoring reference period in the current statistical period and multiple historical statistical periods is obtained; according to the average of the bandwidth data And the preset confidence level, determine the bandwidth dynamic interval corresponding to the monitoring reference period; and monitor the bandwidth status of the current target domain name according to the current bandwidth data and bandwidth dynamic interval of the target domain name.
  • the historical bandwidth data is analyzed to determine a reasonable dynamic bandwidth interval, so that before the abnormal accumulation of bandwidth status affects the quality of service, abnormal bandwidth status can be found in time through the dynamic bandwidth interval and triggered.
  • the follow-up alarm or repair process helps to repair the abnormal state before the business party perceives the abnormal bandwidth, thereby ensuring the overall service quality of the customer.
  • the device for monitoring the bandwidth status is described by taking only the division of the above functional modules as an example. In practical applications, the above functions may be allocated by different functional modules as required Finished, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.
  • the apparatus for monitoring the bandwidth status provided by the foregoing embodiments and the method embodiment for monitoring the bandwidth status belong to the same concept. For specific implementation processes, refer to the method embodiment, and details are not described herein again.
  • FIG. 6 is a schematic structural diagram of a bandwidth monitoring device according to an embodiment of the present invention.
  • the bandwidth monitoring device 600 may have a large difference due to different configurations or performance, and may include one or more central processing units 622 (for example, one or more processors) and a memory 632, and one or more storage applications 642. Or the storage medium 630 of the data 644 (for example, one or one storage device in Shanghai).
  • the memory 632 and the storage medium 630 may be temporary storage or persistent storage.
  • the program stored in the storage medium 630 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the bandwidth monitoring device 600.
  • the central processing unit 622 may be configured to communicate with the storage medium 630, and execute a series of instruction operations in the storage medium 630 on the bandwidth monitoring device 600.
  • the bandwidth monitoring device 600 may also include one or more power sources 629, one or more wired or wireless network interfaces 650, one or more input / output interfaces 658, one or more keyboards 656, and / or, one or more operations System 641, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc.
  • the bandwidth monitoring device 600 may include a memory, and one or more programs, one or more programs stored in the memory, and configured to be executed by one or more processors.
  • the one or more programs include Perform the above instructions for monitoring the status of the bandwidth.
  • the program may be stored in a computer-readable storage medium.
  • the storage medium mentioned may be a read-only memory, a magnetic disk or an optical disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Environmental & Geological Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Algebra (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Physics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Pure & Applied Mathematics (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明公开了一种监控带宽状态的方法和装置，属于网络通信技术领域。所述方法包括：每隔预设监控间隔时长确定目标域名的带宽数据的监控参考时段；获取当前统计周期及多个历史统计周期中所述监控参考时段内所述目标域名的带宽数据均值；根据所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间；根据所述目标域名的当前带宽数据和所述带宽动态区间，监控当前所述目标域名的带宽状态。采用本发明可以及时发现异常的带宽状态，有助于在业务方感知到带宽异常之前修复异常状态，进而保证客户的整体业务质量。

Description

一种监控带宽状态的方法和装置 技术领域
本发明涉及网络通信技术领域，特别涉及一种监控带宽状态的方法和装置。
背景技术
为了提高对用户访问的反馈速度，同时保证业务后台系统的稳定性和高可用性，业务提供方大多选用CDN加速服务来实现相关的业务服务。从CDN提供方的角度来讲，CDN提供方往往会为每个开通CDN加速的业务提供方(可以称为客户)分配CDN带宽资源来实现CDN加速服务。
受用户需求更新、设备状态变化的影响，客户的带宽变化往往较难确定。一方面，在用户需求突发的情况下，客户的带宽会突增，提供相应加速服务的CDN节点的设备负载将会急剧升高，甚至进入过载状态，进而会影响到业务服务的质量；另一方面，如果客户的带宽突降，则相应的CDN节点有很大概率发生了故障，如果无法及时修复故障，业务服务质量将会大幅下滑，甚至会造成用户的大量流失。故而，目前亟需一种能够有效地监控CDN加速服务的带宽突变的方法，使得当CDN节点处于过载或故障状态时，CDN提供方可以及时发现并进行相应处理，以保证客户的整体业务质量。
发明内容
为了解决现有技术的问题，本发明实施例提供了一种监控带宽状态的方法和装置。所述技术方案如下：
第一方面，提供了一种监控带宽状态的方法，所述方法包括：
每隔预设监控间隔时长确定目标域名的带宽数据的监控参考时段；
获取当前统计周期及多个历史统计周期中所述监控参考时段内所述目标域名的带宽数据均值；
根据所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间；
根据所述目标域名的当前带宽数据和所述带宽动态区间，监控当前所述目标域名的带宽状态。
可选的，所述获取当前统计周期及多个历史统计周期中所述监控参考时段内所述目标域名的带宽数据均值，包括：
获取当前统计周期中所述监控参考时段内各指定采集时刻的所述目标域名的实时带宽数据；
根据所述实时带宽数据，计算并存储所述监控参考时段内的所述目标域名的带宽数据均值；
获取预先存储的多个历史统计周期中，所述监控参考时段内所述目标域名的带宽数据均值。
可选的，所述根据所述实时带宽数据，计算并存储所述监控参考时段内所述目标域名的带宽数据均值，包括：
根据所述实时带宽数据，计算所述监控参考时段内所述目标域名的带宽数据均值和带宽数据标准差；
根据所述带宽数据均值、带宽数据标准差和预设的带宽波动权值，确定所述目标域名的带宽波动范围；
剔除所述监控参考时段内所述目标域名的实时带宽数据中，数值在所述带宽波动范围外的实时带宽数据；
根据剔除后剩余的所述实时带宽数据，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值。
可选的，所述根据剔除后剩余的所述实时带宽数据，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值，还包括：
根据剔除后剩余的所述实时带宽数据和每个所述实时带宽数据对应的预设时间权重，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值。
可选的，所述根据所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间，包括：
计算所述当前统计周期以及多个历史统计周期对应的，所述目标域名的多个带宽数据均值的均值和标准差；
根据所述多个带宽数据均值的均值和标准差以及预设的均值波动权值，确定所述目标域名的均值波动范围；
在所述监控参考时段内的所述多个带宽数据均值中，剔除数值处于所述均值波动范围之外的带宽数据均值；
根据剔除后剩余的所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间。
可选的，所述根据剔除后剩余的所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间，包括：
根据剔除后剩余的所述带宽数据均值和每个所述带宽数据均值对应的预设周期权重，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间。
可选的，所述获取当前统计周期及多个历史统计周期中所述监控参考时段内所述目标域名的带宽数据均值之前，还包括：
确定周期类型与当前统计周期的周期类型相同的，且距离当前统计周期最近的多个历史统计周期。
第二方面，提供了一种监控带宽状态的装置，所述装置包括：
第一确定模块，用于每隔预设监控间隔时长确定目标域名的带宽数据的监控参考时段；
获取模块，用于获取当前统计周期及多个历史统计周期中所述监控参考时段内所述目标域名的带宽数据均值；
第二确定模块，用于根据所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间；
监控模块，用于根据所述目标域名的当前带宽数据和所述带宽动态区间，监控当前所述目标域名的带宽状态。
可选的，所述获取模块，具体用于：
获取当前统计周期中所述监控参考时段内各指定采集时刻的所述目标域名的实时带宽数据；
根据所述实时带宽数据，计算并存储所述监控参考时段内的所述目标域名的带宽数据均值；
获取预先存储的多个历史统计周期中，所述监控参考时段内所述目标域名的带宽数据均值。
可选的，所述获取模块，具体用于：
根据所述实时带宽数据，计算所述监控参考时段内所述目标域名的带宽数据均值和带宽数据标准差；
根据所述带宽数据均值、带宽数据标准差和预设的带宽波动权值，确定所述目标域名的带宽波动范围；
剔除所述监控参考时段内所述目标域名的实时带宽数据中，数值在所述带宽波动范围外的实时带宽数据；
根据剔除后剩余的所述实时带宽数据，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值。
可选的，所述获取模块，还用于：根据剔除后剩余的所述实时带宽数据和每个所述实时带宽数据对应的预设时间权重，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值。
可选的，所述第二确定模块，具体用于：
计算所述当前统计周期以及多个历史统计周期对应的，所述目标域名的多个带宽数据均值的均值和标准差；
根据所述多个带宽数据均值的均值和标准差以及预设的均值波动权值，确定所述目标域名的均值波动范围；
在所述监控参考时段内的所述多个带宽数据均值中，剔除数值处于所述均值波动范围之外的带宽数据均值；
根据剔除后剩余的所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间。
可选的，所述第二确定模块，具体用于：
根据剔除后剩余的所述带宽数据均值和每个所述带宽数据均值对应的预设周期权重，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间。
可选的，所述装置还包括第三确定模块，用于：
确定周期类型与当前统计周期的周期类型相同的，且距离当前统计周期最近的多个历史统计周期。
第三方面，提供了一种带宽监控设备，所述带宽监控设备包括处理器和存储器，所述存储器中存储有至少一条指令、至少一段程序、代码集或指令集，所述至少一条指令、所述至少一段程序、所述代码集或指令集由所述处理器加 载并执行以实现如第一方面所述的监控带宽状态的方法。
第四方面，提供了一种计算机可读存储介质，所述存储介质中存储有至少一条指令、至少一段程序、代码集或指令集，所述至少一条指令、所述至少一段程序、所述代码集或指令集由处理器加载并执行以实现如第一方面所述的监控带宽状态的方法。
本发明实施例提供的技术方案带来的有益效果是：
本发明实施例中，每隔预设监控间隔时长确定目标域名的带宽数据的监控参考时段；获取当前统计周期及多个历史统计周期中监控参考时段内目标域名的带宽数据均值；根据带宽数据均值，以及预设的置信水平，确定监控参考时段对应的带宽动态区间；根据目标域名的当前带宽数据和带宽动态区间，监控当前目标域名的带宽状态。这样，在监控带宽状态时，通过对历史的带宽数据进行分析，确定合理的带宽动态区间，从而可以在带宽状态异常积累到影响业务质量之前，通过带宽动态区间及时发现异常的带宽状态，并触发后续告警或修复处理，有助于在业务方感知到带宽异常之前修复异常状态，进而保证客户的整体业务质量。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例提供的一种CDN系统的场景架构示意图；
图2是本发明实施例提供的一种监控带宽状态的方法流程图；
图3是本发明实施例提供的一种确定带宽动态区间的逻辑原理图；
图4是本发明实施例提供的一种监控带宽状态的装置结构示意图；
图5是本发明实施例提供的一种监控带宽状态的装置结构示意图；
图6是本发明实施例提供的一种带宽监控设备的结构示意图。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚，下面将结合附图对本发明实施方式作进一步地详细描述。
本发明实施例提供了一种监控带宽状态的方法，该方法可以运用在CDN系统中，并具体可以由CDN系统中的带宽监控设备来实现。其中，带宽监控设备可以是CDN系统中用于监控享有CDN加速服务的客户的带宽状态的设备，其可以通过获取到的客户的带宽数据来监控客户的带宽状态，并基于监控结果进行带宽异常报警或者触发预设的异常应对处理。上述客户的带宽数据采集和存储可以由带宽监控设备来实现，也可以由CDN系统中的数据采集设备和数据存储设备来分别完成。CDN系统的具体场景架构可以如图1所示。上述带宽监控设备可以包括处理器、存储器、收发器，处理器可以用于进行下述流程中的监控带宽状态的处理，存储器可以用于存储下述处理过程中需要的数据以及产生的数据，收发器可以用于接收和发送下述处理过程中的相关数据。上述监控带宽状态、采集和存储带宽数据的处理均可以由部署在CDN系统内部的分布式系统来实现。本实施例中以“监控带宽状态、采集和存储带宽数据的处理全部由带宽监控设备来执行”为例进行说明，其它情况与之类似，不再赘述。
下面将结合具体实施方式，对图2所示的处理流程进行详细的说明，内容可以如下：
步骤201，每隔预设监控间隔时长确定目标域名的带宽数据的监控参考时段。
在实施中，以目标域名为例，在监控目标域名的带宽状态的过程中，带宽监控设备可以每隔预设监控间隔时长监控一次目标域名的带宽状态。具体来说，每次执行监控时带宽监控设备均可以先确定目标域名的带宽数据的监控参考时段。此处，监控参考时段可以是在监控带宽状态时，需要参考的所有带宽数据所覆盖的时间段，上述监控间隔时长和该监控参考时段均可以是CDN系统的技术人员根据带宽监控需求预先设定的，监控间隔时长可以是1分钟，即每隔一分钟监控一次目标域名的带宽状态，监控参考时段可以是当前时刻之前的30分钟，即每次监控时需要参考前30分钟的带宽数据。
步骤202，获取当前统计周期及多个历史统计周期中监控参考时段内目标域名的带宽数据均值。
在实施中，在开始监控带宽状态时，带宽监控设备可以确定当前时刻所属 的统计周期(即当前统计周期)，以及当前统计周期之前的多个历史统计周期，然后可以在确定了监控参考时段后，获取当前统计周期及多个历史统计周期中监控参考时段内目标域名的带宽数据均值。此处的统计周期可以是CDN系统的技术人员预先设定的，一个统计周期可以是一天或者一周等。值得一提的是，对于监控参考时段跨统计周期的情况，可以认为该监控参考时段属于后一统计周期，例如监控参考时段为30分钟，统计周期为1天，那么在1月2日的00:30之前确定的监控参考时段跨了1月1日和1月2日两个统计周期，而这些监控参考时段仍可以认为是1月2日的监控参考时段，同样适用本实施例公开的监控带宽状态的方法。
可选的，带宽监控设备可以通过缓存带宽数据均值的处理降低每次监控带宽状态时的计算量，相应的，步骤202的处理可以如下：获取当前统计周期中监控参考时段内各指定采集时刻的目标域名的实时带宽数据；根据实时带宽数据，计算并存储监控参考时段内的目标域名的带宽数据均值；获取预先存储的多个历史统计周期中，监控参考时段内目标域名的带宽数据均值。
在实施中，带宽监控设备可以在目标域名对应的业务服务器与CDN系统的接口处，在指定采集时刻采集目标域名的实时带宽数据，并将采集到的实时带宽数据存储在本地。这样，带宽监控设备在确定了监控参考时段后，可以仅获取当前统计周期中监控参考时段内各指定采集时刻的目标域名的实时带宽数据，然后根据这些实时带宽数据，计算监控参考时段内的目标域名的带宽数据均值，并将带宽数据均值存储在本地。可以理解，对于各个历史统计周期中监控参考时段内目标域名的带宽数据均值，带宽监控设备已在历史统计周期内的监控过程中完成了计算并存储在本地，故而，此时带宽监控设备无需再次计算，可以直接从本地获取预先存储的多个历史统计周期中，监控参考时段内目标域名的带宽数据均值。
可选的，在计算带宽数据均值时，可以先通过高斯函数剔除实时带宽数据中的异常值，相应的处理可以如下：根据实时带宽数据，计算监控参考时段内目标域名的带宽数据均值和带宽数据标准差；根据带宽数据均值、带宽数据标准差和预设的带宽波动权值，确定目标域名的带宽波动范围；剔除监控参考时段内目标域名的实时带宽数据中，数值在带宽波动范围外的实时带宽数据；根据剔除后剩余的实时带宽数据，重新计算并存储监控参考时段内目标域名的带 宽数据均值。
在实施中，带宽监控设备可以根据当前统计周期中监控参考时段内的目标域名的实时带宽数据，计算监控参考时段内目标域名的带宽数据均值和带宽数据标准差，然后可以根据带宽数据均值、带宽数据标准差和预设的带宽波动权值，确定目标域名的带宽波动范围。之后，带宽监控设备可以在监控参考时段内目标域名的所有实时带宽数据中，剔除掉数值处于上述带宽波动范围之外的实时带宽数据。进而，带宽监控设备可以根据剔除后剩余的实时带宽数据，重新计算监控参考时段内目标域名的带宽数据均值，并将计算得到的带宽数据均值存储在本地。具体的，对于n个实时带宽数据，其取值分别为x 1-1、x 1-2、…x 1-n，计算得到n个实时带宽数据的带宽数据均值为
Figure PCTCN2018113734-appb-000001
标准差为σ x1(n)，预设的带宽波动权值为ε 1，则带宽波动范围可以为：
Figure PCTCN2018113734-appb-000002
而n个实时带宽数据中存在j个取值在带宽波动范围之外的实时带宽数据，再计算剔除后剩余的n-j个实时带宽数据的带宽数据均值为
Figure PCTCN2018113734-appb-000003
可选的，不同时刻的实时带宽数据对当前带宽数据具备不同的影响程度，相应的，计算带宽数据均值的处理可以如下：根据剔除后剩余的实时带宽数据和每个实时带宽数据对应的预设时间权重，重新计算并存储监控参考时段内目标域名的带宽数据均值。
在实施中，带宽监控设备在通过高斯函数剔除了异常的实时带宽数据之后，可以获取每个实时带宽数据对应的预设时间权重，该预设时间权重可以是CDN系统的技术人员根据经验设定的，用于决定不同时刻的实时带宽数据对于监控当前带宽状态的影响，可以理解，采集时刻距离当前时刻越近的实时带宽数据的预设时间权重越高。这样，在重新计算带宽数据均值时，带宽监控设备可以将剔除后剩余的每个实时带宽数据和其对应的预设时间权重相乘，并将所有乘积累加后取均值，从而可以得到监控参考时段内目标域名的带宽数据均值，并对带宽数据均值进行存储。
可选的，考虑到不同周期类型的统计周期的带宽使用情况可能差异较大，故而在监控带宽状态时可以仅参考相同周期类型的统计周期，故而步骤202之前可以存在以下处理：确定周期类型与当前统计周期的周期类型相同的，且距离当前统计周期最近的多个历史统计周期。
在实施中，CDN系统的技术人员可以对按照域名的带宽使用情况将统计周 期划分为多种周期类型，每种周期类型下的所有统计周期内的带宽使用情况应大致符合同一规律，不同周期类型下的统计周期间的带宽使用情况应具备较差差异。例如，单位统计周期为1天时，则可以将统计周期分为节假日和工作日，也可以将统计周期分为月初、月中和月末，还可以将统计周期分为业务更新日、业务维护日和业务正常日等等。单个统计周期可以具备不同分类标准下的多个周期类型，例如1月25日的周期类型可以同时是工作日、月末和业务更新日。这样，带宽监控设备在确定了当前统计周期后，可以再确定当前统计周期的周期类型，然后确定周期类型与当前统计周期的周期类型相同的，且距离当前统计周期最近的多个历史统计周期。需要说明的是，在确定历史统计周期时，可以按需求权衡所有周期类型的一致度和距离当前统计周期的远近，例如，可以设定在最近三个月内选择周期类型相似度最高且大于90％的，5个最近的历史统计周期，也可以在最近一个月内选择周期类型相似度最高且大于60％的，5个最近的历史统计周期。
步骤203，根据带宽数据均值，以及预设的置信水平，确定监控参考时段对应的带宽动态区间。
在实施中，带宽监控设备在获取了多个统计周期中监控参考时段内目标域名的带宽数据均值后，可以根据这些带宽数据均值，以及预设的置信水平，来确定监控参考时段对应的带宽动态区间。具体的，历史统计周期个数为m-1，则总共存在m个带宽数据均值，其取值分别为
Figure PCTCN2018113734-appb-000004
计算得到m个带宽数据均值的均值为
Figure PCTCN2018113734-appb-000005
标准差为
Figure PCTCN2018113734-appb-000006
预设的置信水平为1-α，则带宽动态区间可以为：
Figure PCTCN2018113734-appb-000007
上述步骤201至步骤203确定带宽动态区间的逻辑原理可以参考图3所示，其中，在监控时刻，选择当前统计周期和5个历史统计周期的监控参考时段内的带宽数据，然后分别计算每个统计周期中10份带宽数据的带宽数据均值，再由6个统计周期的带宽数据均值确定带宽动态区间。
可选的，在确定带宽动态区间时，可以先通过高斯函数剔除所有带宽数据均值中的异常值，相应的，步骤203的处理可以如下：计算当前统计周期以及多个历史统计周期对应的，目标域名的多个带宽数据均值的均值和标准差；根据多个带宽数据均值的均值和标准差以及预设的均值波动权值，确定目标域名 的均值波动范围；在监控参考时段内的多个带宽数据均值中，剔除数值处于均值波动范围之外的带宽数据均值；根据剔除后剩余的带宽数据均值，以及预设的置信水平，确定监控参考时段对应的带宽动态区间。
在实施中，带宽监控设备在确定当前时刻的带宽动态区间的过程中，可以先计算步骤202中计算得到的多个带宽数据均值的均值和标准差，然后可以根据多个带宽数据均值的均值和标准差以及预设的均值波动权值，确定目标域名的均值波动范围。之后，带宽监控设备可以在监控参考时段内目标域名的所有带宽数据均值中，剔除掉数值处于上述均值波动范围之外的带宽数据均值。进而，带宽监控设备可以根据剔除后剩余的带宽数据均值，以及预设的置信水平，确定监控参考时段对应的带宽动态区间。具体的，对于m个带宽数据均值：
Figure PCTCN2018113734-appb-000008
计算得到m个带宽数据均值的均值为
Figure PCTCN2018113734-appb-000009
标准差为
Figure PCTCN2018113734-appb-000010
预设的均值波动权值为ε 2，则均值波动范围可以为：
Figure PCTCN2018113734-appb-000011
而m个带宽数据均值中存在i个取值在均值波动范围之外的带宽数据均值，再计算剔除后剩余的m-i个带宽数据均值的均值为
Figure PCTCN2018113734-appb-000012
标准差为
Figure PCTCN2018113734-appb-000013
预设的置信水平为1-α，则带宽动态区间可以为：
Figure PCTCN2018113734-appb-000014
可选的，不同统计周期的带宽数据均值对带宽动态区间具备不同的影响程度，相应的，确定带宽动态区间的处理可以如下：根据剔除后剩余的带宽数据均值和每个带宽数据均值对应的预设周期权重，以及预设的置信水平，确定监控参考时段对应的带宽动态区间。
在实施中，带宽监控设备在通过高斯函数剔除了异常的带宽数据均值之后，可以获取每个带宽数据均值对应的预设时间权重，该预设周期权重可以是CDN系统的技术人员根据经验设定的，用于决定不同统计周期的带宽数据均值对于监控当前带宽状态的影响，可以理解，对应的统计周期距离当前统计周期越近的带宽数据均值的预设周期权重越高。这样，在确定带宽动态区间时，带宽监控设备可以根据剔除后剩余的带宽数据均值和每个带宽数据均值对应的预设周期权重，以及预设的置信水平，确定监控参考时段对应的带宽动态区间。此处示例性地给出一种使用预设周期权重的方式：带宽监控设备可以按照各个带宽数据均值对应的预设周期权重之比，在将相应的带宽数据均值复制多份，然后 可以利用复制后的所有带宽数据均值来确定相应的带宽动态区间。
步骤204，根据目标域名的当前带宽数据和带宽动态区间，监控当前目标域名的带宽状态。
在实施中，带宽监控设备在确定了监控参考时段对应的带宽动态区间后，可以判断目标域名的当前带宽数据是否处于该带宽动态区间中。如果处于，带宽监控设备则可以认为目标域名当前的带宽状态正常；如果不处于，带宽监控设备则可以判定目标域名当前的带宽状态异常，进而可以触发后续的带宽异常报警或者预设的异常应对处理。
本发明实施例中，每隔预设监控间隔时长确定目标域名的带宽数据的监控参考时段；获取当前统计周期及多个历史统计周期中监控参考时段内目标域名的带宽数据均值；根据带宽数据均值，以及预设的置信水平，确定监控参考时段对应的带宽动态区间；根据目标域名的当前带宽数据和带宽动态区间，监控当前目标域名的带宽状态。这样，在监控带宽状态时，通过对历史的带宽数据进行分析，确定合理的带宽动态区间，从而可以在带宽状态异常积累到影响业务质量之前，通过带宽动态区间及时发现异常的带宽状态，并触发后续告警或修复处理，有助于在业务方感知到带宽异常之前修复异常状态，进而保证客户的整体业务质量。
基于相同的技术构思，本发明实施例还提供了一种监控带宽状态的装置，如图4所示，所述装置包括：
第一确定模块401，用于每隔预设监控间隔时长确定目标域名的带宽数据的监控参考时段；
获取模块402，用于获取当前统计周期及多个历史统计周期中所述监控参考时段内所述目标域名的带宽数据均值；
第二确定模块403，用于根据所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间；
监控模块404，用于根据所述目标域名的当前带宽数据和所述带宽动态区间，监控当前所述目标域名的带宽状态。
可选的，所述获取模块402，具体用于：
获取当前统计周期中所述监控参考时段内各指定采集时刻的所述目标域名 的实时带宽数据；
根据所述实时带宽数据，计算并存储所述监控参考时段内的所述目标域名的带宽数据均值；
获取预先存储的多个历史统计周期中，所述监控参考时段内所述目标域名的带宽数据均值。
可选的，所述获取模块402，具体用于：
根据所述实时带宽数据，计算所述监控参考时段内所述目标域名的带宽数据均值和带宽数据标准差；
根据所述带宽数据均值、带宽数据标准差和预设的带宽波动权值，确定所述目标域名的带宽波动范围；
剔除所述监控参考时段内所述目标域名的实时带宽数据中，数值在所述带宽波动范围外的实时带宽数据；
根据剔除后剩余的所述实时带宽数据，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值。
可选的，所述获取模块402，还用于：
根据剔除后剩余的所述实时带宽数据和每个所述实时带宽数据对应的预设时间权重，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值。
可选的，所述第二确定模块403，具体用于：
计算所述当前统计周期以及多个历史统计周期对应的，所述目标域名的多个带宽数据均值的均值和标准差；
根据所述多个带宽数据均值的均值和标准差以及预设的均值波动权值，确定所述目标域名的均值波动范围；
在所述监控参考时段内的所述多个带宽数据均值中，剔除数值处于所述均值波动范围之外的带宽数据均值；
根据剔除后剩余的所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间。
可选的，所述第二确定模块403，具体用于：
根据剔除后剩余的所述带宽数据均值和每个所述带宽数据均值对应的预设周期权重，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间。
可选的，如图5所示，所述装置还包括第三确定模块405，用于：
确定周期类型与当前统计周期的周期类型相同的，且距离当前统计周期最近的多个历史统计周期。
本发明实施例中，每隔预设监控间隔时长确定目标域名的带宽数据的监控参考时段；获取当前统计周期及多个历史统计周期中监控参考时段内目标域名的带宽数据均值；根据带宽数据均值，以及预设的置信水平，确定监控参考时段对应的带宽动态区间；根据目标域名的当前带宽数据和带宽动态区间，监控当前目标域名的带宽状态。这样，在监控带宽状态时，通过对历史的带宽数据进行分析，确定合理的带宽动态区间，从而可以在带宽状态异常积累到影响业务质量之前，通过带宽动态区间及时发现异常的带宽状态，并触发后续告警或修复处理，有助于在业务方感知到带宽异常之前修复异常状态，进而保证客户的整体业务质量。
需要说明的是：上述实施例提供的监控带宽状态的装置在监控带宽状态时，仅以上述各功能模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即将装置的内部结构划分成不同的功能模块，以完成以上描述的全部或者部分功能。另外，上述实施例提供的监控带宽状态的装置与监控带宽状态的方法实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
图6是本发明实施例提供的带宽监控设备的结构示意图。该带宽监控设备600可因配置或性能不同而产生比较大的差异，可以包括一个或一个以上中央处理器622(例如，一个或一个以上处理器)和存储器632，一个或一个以上存储应用程序642或数据644的存储介质630(例如一个或一个以上海量存储设备)。其中，存储器632和存储介质630可以是短暂存储或持久存储。存储在存储介质630的程序可以包括一个或一个以上模块(图示没标出)，每个模块可以包括对带宽监控设备600中的一系列指令操作。更进一步地，中央处理器622可以设置为与存储介质630通信，在带宽监控设备600上执行存储介质630中的一系列指令操作。
带宽监控设备600还可以包括一个或一个以上电源629，一个或一个以上有线或无线网络接口650，一个或一个以上输入输出接口658，一个或一个以上键 盘656，和/或，一个或一个以上操作系统641，例如Windows Server，Mac OS X，Unix，Linux，FreeBSD等等。
带宽监控设备600可以包括有存储器，以及一个或者一个以上的程序，其中一个或者一个以上程序存储于存储器中，且经配置以由一个或者一个以上处理器执行所述一个或者一个以上程序包含用于进行上述监控带宽状态的指令。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成，也可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，上述提到的存储介质可以是只读存储器，磁盘或光盘等。
以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (16)

  1. 一种监控带宽状态的方法，其特征在于，所述方法包括：
    每隔预设监控间隔时长确定目标域名的带宽数据的监控参考时段；
    获取当前统计周期及多个历史统计周期中所述监控参考时段内所述目标域名的带宽数据均值；
    根据所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间；
    根据所述目标域名的当前带宽数据和所述带宽动态区间，监控当前所述目标域名的带宽状态。
  2. 根据权利要求1所述的方法，其特征在于，所述获取当前统计周期及多个历史统计周期中所述监控参考时段内所述目标域名的带宽数据均值，包括：
    获取当前统计周期中所述监控参考时段内各指定采集时刻的所述目标域名的实时带宽数据；
    根据所述实时带宽数据，计算并存储所述监控参考时段内的所述目标域名的带宽数据均值；
    获取预先存储的多个历史统计周期中，所述监控参考时段内所述目标域名的带宽数据均值。
  3. 根据权利要求2所述的方法，其特征在于，所述根据所述实时带宽数据，计算并存储所述监控参考时段内所述目标域名的带宽数据均值，包括：
    根据所述实时带宽数据，计算所述监控参考时段内所述目标域名的带宽数据均值和带宽数据标准差；
    根据所述带宽数据均值、带宽数据标准差和预设的带宽波动权值，确定所述目标域名的带宽波动范围；
    剔除所述监控参考时段内所述目标域名的实时带宽数据中，数值在所述带宽波动范围外的实时带宽数据；
    根据剔除后剩余的所述实时带宽数据，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值。
  4. 根据权利要求3所述的方法，其特征在于，所述根据剔除后剩余的所述实时带宽数据，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值，还包括：
    根据剔除后剩余的所述实时带宽数据和每个所述实时带宽数据对应的预设时间权重，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值。
  5. 根据权利要求1所述的方法，其特征在于，所述根据所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间，包括：
    计算所述当前统计周期以及多个历史统计周期对应的，所述目标域名的多个带宽数据均值的均值和标准差；
    根据所述多个带宽数据均值的均值和标准差以及预设的均值波动权值，确定所述目标域名的均值波动范围；
    在所述监控参考时段内的所述多个带宽数据均值中，剔除数值处于所述均值波动范围之外的带宽数据均值；
    根据剔除后剩余的所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间。
  6. 根据权利要求5所述的方法，其特征在于，所述根据剔除后剩余的所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间，包括：
    根据剔除后剩余的所述带宽数据均值和每个所述带宽数据均值对应的预设周期权重，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间。
  7. 根据权利要求1所述的方法，其特征在于，所述获取当前统计周期及多个历史统计周期中所述监控参考时段内所述目标域名的带宽数据均值之前，还包括：
    确定周期类型与当前统计周期的周期类型相同的，且距离当前统计周期最近的多个历史统计周期。
  8. 一种监控带宽状态的装置，其特征在于，所述装置包括：
    第一确定模块，用于每隔预设监控间隔时长确定目标域名的带宽数据的监控参考时段；
    获取模块，用于获取当前统计周期及多个历史统计周期中所述监控参考时段内所述目标域名的带宽数据均值；
    第二确定模块，用于根据所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间；
    监控模块，用于根据所述目标域名的当前带宽数据和所述带宽动态区间，监控当前所述目标域名的带宽状态。
  9. 根据权利要求8所述的装置，其特征在于，所述获取模块，具体用于：
    获取当前统计周期中所述监控参考时段内各指定采集时刻的所述目标域名的实时带宽数据；
    根据所述实时带宽数据，计算并存储所述监控参考时段内的所述目标域名的带宽数据均值；
    获取预先存储的多个历史统计周期中，所述监控参考时段内所述目标域名的带宽数据均值。
  10. 根据权利要求9所述的装置，其特征在于，所述获取模块，具体用于：
    根据所述实时带宽数据，计算所述监控参考时段内所述目标域名的带宽数据均值和带宽数据标准差；
    根据所述带宽数据均值、带宽数据标准差和预设的带宽波动权值，确定所述目标域名的带宽波动范围；
    剔除所述监控参考时段内所述目标域名的实时带宽数据中，数值在所述带宽波动范围外的实时带宽数据；
    根据剔除后剩余的所述实时带宽数据，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值。
  11. 根据权利要求10所述的装置，其特征在于，所述获取模块，还用于：根据剔除后剩余的所述实时带宽数据和每个所述实时带宽数据对应的预设时间 权重，重新计算并存储所述监控参考时段内所述目标域名的带宽数据均值。
  12. 根据权利要求8所述的装置，其特征在于，所述第二确定模块，具体用于：
    计算所述当前统计周期以及多个历史统计周期对应的，所述目标域名的多个带宽数据均值的均值和标准差；
    根据所述多个带宽数据均值的均值和标准差以及预设的均值波动权值，确定所述目标域名的均值波动范围；
    在所述监控参考时段内的所述多个带宽数据均值中，剔除数值处于所述均值波动范围之外的带宽数据均值；
    根据剔除后剩余的所述带宽数据均值，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间。
  13. 根据权利要求12所述的装置，其特征在于，所述第二确定模块，具体用于：
    根据剔除后剩余的所述带宽数据均值和每个所述带宽数据均值对应的预设周期权重，以及预设的置信水平，确定所述监控参考时段对应的带宽动态区间。
  14. 根据权利要求8所述的装置，其特征在于，所述装置还包括第三确定模块，用于：
    确定周期类型与当前统计周期的周期类型相同的，且距离当前统计周期最近的多个历史统计周期。
  15. 一种带宽监控设备，其特征在于，所述带宽监控设备包括处理器和存储器，所述存储器中存储有至少一条指令、至少一段程序、代码集或指令集，所述至少一条指令、所述至少一段程序、所述代码集或指令集由所述处理器加载并执行以实现如权利要求1至7任一所述的监控带宽状态的方法。
  16. 一种计算机可读存储介质，其特征在于，所述存储介质中存储有至少一条指令、至少一段程序、代码集或指令集，所述至少一条指令、所述至少一 段程序、所述代码集或指令集由处理器加载并执行以实现如权利要求1至7任一所述的监控带宽状态的方法。
PCT/CN2018/113734 2018-09-30 2018-11-02 一种监控带宽状态的方法和装置 WO2020062439A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US16/959,128 US10965565B2 (en) 2018-09-30 2018-11-02 Method and apparatus for monitoring bandwidth condition
EP18935279.2A EP3860052A4 (en) 2018-09-30 2018-11-02 METHOD AND DEVICE FOR MONITORING THE CONDITION OF THE BANDWIDTH

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811160633.3 2018-09-30
CN201811160633.3A CN109039833B (zh) 2018-09-30 2018-09-30 一种监控带宽状态的方法和装置

Publications (1)

Publication Number Publication Date
WO2020062439A1 true WO2020062439A1 (zh) 2020-04-02

Family

ID=64615542

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/113734 WO2020062439A1 (zh) 2018-09-30 2018-11-02 一种监控带宽状态的方法和装置

Country Status (4)

Country Link
US (1) US10965565B2 (zh)
EP (1) EP3860052A4 (zh)
CN (1) CN109039833B (zh)
WO (1) WO2020062439A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114205329A (zh) * 2021-11-16 2022-03-18 杭州华橙软件技术有限公司 一种设备编码方法及装置

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109412977B (zh) * 2018-10-18 2022-05-17 网宿科技股份有限公司 一种域名带宽调节方法及相关设备
CN110505115B (zh) * 2019-07-30 2021-07-13 网宿科技股份有限公司 一种监控交换机跑高风险的方法和装置
CN110602569B (zh) * 2019-09-18 2021-11-09 深圳市梦网视讯有限公司 一种基于带宽趋势的带宽复用方法和系统
CN110830604B (zh) * 2019-10-28 2021-09-10 腾讯科技（深圳）有限公司 Dns调度方法、装置
CN111241477B (zh) * 2020-01-07 2023-10-20 支付宝(杭州)信息技术有限公司 构建监控基准线的方法、监控数据对象状态的方法及装置
CN111210225B (zh) * 2020-01-14 2024-10-11 中国建设银行股份有限公司 交易系统的状态监控方法、装置、设备和存储介质
CN114143223B (zh) * 2020-08-13 2023-11-21 贵州白山云科技股份有限公司 一种带宽异常检测方法、装置、介质及设备
CN112187870B (zh) * 2020-09-04 2022-10-11 网宿科技股份有限公司 一种带宽平滑方法及装置
CN112347423A (zh) * 2020-10-27 2021-02-09 上汽通用五菱汽车股份有限公司 半径分析原始数据的数据筛选方法及装置、存储介质
CN112527544B (zh) * 2020-11-23 2022-04-29 聚好看科技股份有限公司 一种服务器、触发熔断的方法及装置
CN114584565B (zh) * 2020-12-01 2024-01-30 中移(苏州)软件技术有限公司 一种应用防护方法及系统、电子设备及存储介质
CN114915559B (zh) * 2021-02-08 2024-01-23 北京金山云网络技术有限公司 Cdn节点带宽规划方法、装置及存储介质
CN113641667B (zh) * 2021-08-12 2022-05-20 深圳市润迅通投资有限公司 一种分布式大数据采集平台的数据异常监控系统及方法
CN113904992B (zh) * 2021-09-28 2023-10-17 咪咕文化科技有限公司 带宽资源调度方法、装置、计算设备及存储介质
CN114363219B (zh) * 2022-01-07 2024-03-19 上海哔哩哔哩科技有限公司 数据处理方法及装置
CN114844762B (zh) * 2022-04-19 2023-11-07 中南大学 告警真实性检测方法和装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102271068A (zh) * 2011-09-06 2011-12-07 电子科技大学 一种dos/ddos攻击检测方法
CN104732762A (zh) * 2015-04-21 2015-06-24 银江股份有限公司 一种交通异常路段概率识别方法
CN104954192A (zh) * 2014-03-27 2015-09-30 东华软件股份公司 一种网络流量监测方法和设备
US20160277264A1 (en) * 2015-03-20 2016-09-22 Sony Corporation System and method for remote monitoring of api performance and user behavior associated with user interface

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101651568B (zh) * 2009-07-01 2011-12-07 青岛农业大学 一种网络流量预测和异常检测方法
US9350616B1 (en) * 2010-05-11 2016-05-24 Trend Micro Inc. Bandwidth prediction using a past available bandwidth value and a slope calculated from past available bandwidth values
US9954788B2 (en) * 2011-06-03 2018-04-24 Apple Inc. Bandwidth estimation based on statistical measures
US8812662B2 (en) * 2011-06-29 2014-08-19 Sonic Ip, Inc. Systems and methods for estimating available bandwidth and performing initial stream selection when streaming content
US9467361B2 (en) * 2011-12-20 2016-10-11 Shoretel, Inc. Bandwidth utilization monitoring for a communication system
CN102571493A (zh) * 2012-01-09 2012-07-11 华为技术有限公司 网络检测的方法及设备
US20130263167A1 (en) * 2012-03-30 2013-10-03 Bytemobile, Inc. Adaptive Traffic Management in Cellular Wireless Networks
EP2728827A1 (en) * 2012-10-31 2014-05-07 British Telecommunications public limited company Communications network using a tunnel to connect two network nodes
CN103678085B (zh) * 2013-12-16 2016-05-04 上海证券交易所 一种流量动态可控的系统监控数据采集方法
US9806973B2 (en) * 2014-03-18 2017-10-31 Ciena Corporation Bandwidth analytics in a software defined network (SDN) controlled multi-layer network for dynamic estimation of power consumption
US9450879B2 (en) * 2014-05-09 2016-09-20 Nexgen Storage, Inc. Adaptive bandwidth throttling
CN105530219B (zh) * 2014-09-28 2019-12-10 腾讯科技（深圳）有限公司 一种连接检测方法及装置
US20160357935A1 (en) * 2015-06-04 2016-12-08 True Health Diagnostics Llc Novel set of fasting blood biomarkers to detect patients with impaired glucose tolerance
CN106027288A (zh) * 2016-05-10 2016-10-12 华北电力大学 一种配电线路信息监测业务通信流量预测方法
CN107705149A (zh) * 2017-09-22 2018-02-16 平安科技（深圳）有限公司 数据实时监控方法、装置、终端设备及存储介质
US20190372857A1 (en) * 2018-05-29 2019-12-05 Arris Enterprises Llc Capacity planning and recommendation system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102271068A (zh) * 2011-09-06 2011-12-07 电子科技大学 一种dos/ddos攻击检测方法
CN104954192A (zh) * 2014-03-27 2015-09-30 东华软件股份公司 一种网络流量监测方法和设备
US20160277264A1 (en) * 2015-03-20 2016-09-22 Sony Corporation System and method for remote monitoring of api performance and user behavior associated with user interface
CN104732762A (zh) * 2015-04-21 2015-06-24 银江股份有限公司 一种交通异常路段概率识别方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3860052A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114205329A (zh) * 2021-11-16 2022-03-18 杭州华橙软件技术有限公司 一种设备编码方法及装置

Also Published As

Publication number Publication date
EP3860052A1 (en) 2021-08-04
US20200366583A1 (en) 2020-11-19
US10965565B2 (en) 2021-03-30
CN109039833A (zh) 2018-12-18
EP3860052A4 (en) 2021-10-27
CN109039833B (zh) 2022-11-22

Similar Documents

Publication Publication Date Title
WO2020062439A1 (zh) 一种监控带宽状态的方法和装置
CN108776934B (zh) 分布式数据计算方法、装置、计算机设备及可读存储介质
US8099379B2 (en) Performance evaluating apparatus, performance evaluating method, and program
US8880679B2 (en) Techniques for web server management
WO2021004063A1 (zh) 一种缓存服务器的带宽调度方法及装置
US11316792B2 (en) Method and system of limiting traffic
CN112231075A (zh) 一种基于云服务的服务器集群负载均衡控制方法及系统
CN110223146B (zh) 客户购电服务全过程监控系统及方法
US11388064B2 (en) Prediction based on time-series data
CN110474852B (zh) 一种带宽调度方法及装置
US9501326B2 (en) Processing control system, processing control method, and processing control program
CN111124819A (zh) 全链路监控的方法和装置
CN112749056A (zh) 应用服务指标监控方法、装置、计算机设备和存储介质
US9292336B1 (en) Systems and methods providing optimization data
CN108509313A (zh) 一种业务监控方法、平台及存储介质
CN105049509A (zh) 一种集群调度方法、负载均衡器以及集群系统
CN112506619A (zh) 作业处理方法、装置、电子设备、存储介质和程序产品
CN115033477B (zh) 一种面向大规模微服务的性能异常主动检测和处理方法及系统
US11295224B1 (en) Metrics prediction using dynamic confidence coefficients
CN113407363A (zh) 一种基于远程字典服务的滑窗计数方法及装置
CN117492944A (zh) 任务调度方法、装置、电子设备及可读存储介质
CN110519317B (zh) 一种数据传输方法以及设备
CN107426012B (zh) 一种基于超融合架构的故障恢复方法及其装置
CN113656252A (zh) 故障定位方法、装置、电子设备以及存储介质
CN116821738A (zh) 基于Kmeans的形变监测算法、系统、设备及可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18935279

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2018935279

Country of ref document: EP

Effective date: 20210427