WO2011088767A1 - 内容分发的方法、系统及调度服务器 - Google Patents

内容分发的方法、系统及调度服务器 Download PDF

Info

Publication number
WO2011088767A1
WO2011088767A1 PCT/CN2011/070207 CN2011070207W WO2011088767A1 WO 2011088767 A1 WO2011088767 A1 WO 2011088767A1 CN 2011070207 W CN2011070207 W CN 2011070207W WO 2011088767 A1 WO2011088767 A1 WO 2011088767A1
Authority
WO
WIPO (PCT)
Prior art keywords
service processing
service
content
service request
processing information
Prior art date
Application number
PCT/CN2011/070207
Other languages
English (en)
French (fr)
Inventor
韩润生
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP11734349.1A priority Critical patent/EP2515504B1/en
Publication of WO2011088767A1 publication Critical patent/WO2011088767A1/zh
Priority to US13/546,592 priority patent/US8443054B2/en

Links

Classifications

    • 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
    • H04L67/1008Server selection for load balancing based on parameters of servers, e.g. available memory or workload
    • 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
    • H04L67/50Network services
    • H04L67/60Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/61Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources taking into account QoS or priority requirements

Definitions

  • the CDNCF entity and the CCF entity are essentially scheduling servers, and the CDF entity stores the media content source required by the user, which is essentially a content distribution device, and multiple CDF entities form a cluster, which is responsible for directly transmitting media content to the user.
  • the prior art needs to consider the load balancing among multiple CDF entities when distributing the service request of the content to the CDF entity for the service request, according to load balancing. Principles for the distribution of business requests.
  • the CDF entity fails to provide services, the CDF entity that can provide the service is used to provide services to the user, and the service request will not be distributed to the faulty CDF entity.
  • the inventors have found that the prior art has at least the following disadvantages:
  • the prior art has at least the following disadvantages:
  • the actual processing capability is degraded, if still distributed according to the load balancing principle of the prior art.
  • Service requests may result in some services not being serviced in a timely manner, and at the same time, due to the availability of other CDF entities in the CDN network, resources will not be fully utilized.
  • SUMMARY OF THE INVENTION In order to reduce service loss, improve business success rate and user experience, and make full use of resources, The embodiment of the invention provides a method, a system and a scheduling server for content distribution. The technical solution is as follows:
  • a method for content distribution comprising:
  • the service processing information is one of a service processing success number, a service processing success rate, a service processing failure number, and a service processing failure rate.
  • the received service request is sent to the selected device.
  • a system for content distribution comprising: a scheduling server and a content distribution device;
  • the scheduling server is configured to receive a service request for accessing the content, and select a content distribution device to respond to the service request according to the service processing information of each content distribution device where the content is located, where the service processing information is the number of service processing success times. And one of a service processing success rate, a number of service processing failures, and a service processing failure rate; and transmitting the received service request to the selected content distribution device;
  • the content distribution device is configured to receive a service request sent by the scheduling server.
  • a scheduling server is also provided, the scheduling server comprising:
  • a first receiving module configured to receive a service request for accessing content
  • the selecting module is configured to: according to the service processing information of each device where the content is located, select a device to respond to the service request, where the service processing information is a service processing success number, a service processing success rate, a service processing failure number, and a service processing.
  • the service processing information is a service processing success number, a service processing success rate, a service processing failure number, and a service processing.
  • a sending module configured to send the service request received by the first receiving module to the device selected by the selecting module.
  • Embodiment 1 is a flowchart of a method for content distribution according to Embodiment 1 of the present invention
  • Embodiment 2 is a schematic structural diagram of a network provided by Embodiment 2 of the present invention.
  • FIG. 3 is a flowchart of a method for content distribution according to Embodiment 2 of the present invention.
  • FIG. 4 is a schematic structural diagram of a system for content distribution according to Embodiment 3 of the present invention.
  • FIG. 5 is a schematic structural diagram of a scheduling server according to Embodiment 4 of the present invention.
  • FIG. 6 is a schematic structural diagram of another scheduling server according to Embodiment 4 of the present invention.
  • FIG. 7 is a schematic structural diagram of an update module according to Embodiment 4 of the present invention.
  • FIG. 8 is a schematic structural diagram of a first selection module according to Embodiment 4 of the present invention.
  • FIG. 9 is a schematic structural diagram of a second selection module according to Embodiment 4 of the present invention.
  • FIG. 10 is a schematic structural diagram of a third selection module according to Embodiment 4 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS In order to make the objects, technical solutions and advantages of the present invention more comprehensible, the embodiments of the present invention will be further described in detail with reference to the accompanying drawings.
  • the embodiment provides a method for content distribution, and the method is specifically as follows: 101: Receive a service request for accessing content;
  • the service processing information is one of a service processing success number, a service processing success rate, a service processing failure number, and a service processing failure rate;
  • the method provided in this embodiment can not only improve the success rate of service processing, but also avoid resources by processing the service success rate according to the device, or the success rate of the service processing, or the number of service failures, or the service processing failure rate. Waste, reduce unnecessary business losses, and thus provide better service to users.
  • Embodiment 2
  • This embodiment provides a method for content distribution, which adjusts a distribution policy according to service processing information of the device, thereby achieving an effect of improving service success rate and avoiding resource waste.
  • the service processing information is one of the service processing success number, the service processing success rate, the service processing failure number, and the service processing failure rate.
  • the service processing information is taken as an example of the service processing success rate.
  • the service processing success rate or the failure rate can be implemented by the existing algorithm. This embodiment does not specifically limit this, and ensures that for each device where the same content is located, when calculating the respective service processing success rate or service processing failure rate, The algorithm can be.
  • only the ratio of the number of service processing successes to the number of service requests is taken as the service processing success rate.
  • the ratio of the number of service processing failures to the number of service requests is used as the service processing failure rate.
  • the content that the user wants to obtain is A, that is, the service request for distributing the content A is taken as an example, combined with the network structure shown in FIG. 2, and the distribution policy is adjusted according to the processing success rate of the device.
  • the method of content distribution provided in this embodiment will be described.
  • one CDNCF entity controls multiple CCF entities.
  • the CDNCF entity selects one of the plurality of CCF entities to be the most suitable CCF entity;
  • the entity controls a plurality of CDF entities, and the selected CCF entity will select a most suitable CDF entity among the plurality of CDF entities it controls, and the selected CDF entity is responsible for directly transmitting the media content to the user.
  • the C ⁇ CF entity and the CCF entity are essentially scheduling servers, and the CDF entity is a content distribution device that directly provides content to the user.
  • the service request distribution of the content A can be performed by the CCF entity in the CDCFF entity or the CCF entity.
  • the CCF entity distributes the service request of the content A to the CDF entity
  • the device storing the content A is the CDF entity 1 and the CDF entity 2 respectively.
  • the method of content distribution provided in this embodiment will be described in detail. Referring to Figure 3, the method flow is as follows:
  • the CCF entity receives the service request for accessing the content A, and selects a CDF entity to respond to the service request according to the service processing information of the CDF entity 1 and the CDF entity 1 where the content A is located; specifically, the CDF entity according to the content A 1 and the service processing information of the CDF entity 2, when selecting a CDF entity to respond to the service request, including but not limited to the following two selection methods:
  • the first selection method First, after a measurement period ends, determining whether there is an abnormal device according to the relationship between the service processing information of the CDF entity 1 and the CDF entity 1 where the content A is located and the preset threshold value;
  • the distribution ratio is set in each device where the content is located; the service request.
  • one device is selected to respond to the service request in the device with the first N bits, and N is a natural number smaller than the total number of all devices in which the content is located.
  • the measurement period in the first selection mode and the measurement period mentioned in the subsequent steps can be determined by the specific configuration.
  • the measurement period can be either a specific time or the number of times the service request is distributed.
  • the configuration measurement period is 1.
  • Second, or 1 minute, etc., or as the number of distribution service requests is 10 as a measurement period, that is, every 10 service requests are distributed as one measurement period.
  • This embodiment does not limit the specific content of the measurement period.
  • the reason why the service request of the content fails may be caused by a plurality of reasons, which may be from the scheduling server CCF entity, or may be from an entity controlling the scheduling server, such as a CDNCF entity controlling the CCF entity, and the service request for the content fails in this embodiment. The reason is not specifically limited.
  • the embodiment determines that the CDF entity 1 is an abnormal device as an example. Specifically, when determining whether there is an abnormal device, it can be divided into the following two cases:
  • the content is stored on multiple devices:
  • the service processing success rate of each device where the content is located is sorted, and the abnormality device is determined according to the gap of the service processing success rate of each device. For example, if the difference between the service processing success rates of the two devices is higher than the preset threshold, then it is determined that there is an abnormal device, and the device with a low service success rate is determined to be an abnormal device.
  • the service processing success rate on the CDF entity 1 is 0.5
  • the service processing success rate on the CDF entity 2 is 1, due to the difference in the success rate of the service processing of the two CDF entities. The value is "50%".
  • Entity 1 is an abnormal device.
  • the service processing information is the number of successful service processing, if the content is stored on multiple devices, it is determined whether there is an abnormal device.
  • the processing mode is the same as the processing method of the service processing information for the service processing success rate, which is not described in this embodiment.
  • the service processing failure rate of each device where the content is located is sorted, and according to each The sorting result of the service processing failure rate of the device determines whether there is an abnormal device. For example, if the difference between the service processing failure rates of the two devices is higher than the preset threshold, it is determined that there is an abnormal device, and the device with a high service failure rate is determined to be an abnormal device. For example, for a service request of the same content A, the service processing failure rate on the CDF entity 1 is 0.5, and the service processing failure rate on the CDF entity 1 is 0, due to the difference in service processing success rates of the two CDF entities.
  • Entity 1 is an abnormal device.
  • the preset threshold is determined by the configuration. This embodiment does not limit the specific size of the preset threshold.
  • the service processing information is the number of service processing failures, if the content is stored on multiple devices, if the abnormal device is determined, the processing mode and the service processing information are the same as the processing failure rate of the service processing. This example will not be described in detail.
  • the content is stored on a device:
  • the device determines whether the device is an abnormal device based on the absolute value of the service processing success rate of the device.
  • whether the device is an abnormal device can be determined according to the relationship between the absolute value of the service processing success rate and the preset threshold. For example, if the content is only on the CDF entity 1, and the success rate of the CDF entity 1 for the content is less than the threshold, the device is determined to be an abnormal device.
  • the service processing success rate of the device is 0.4, for example, because the absolute value of the service processing success rate is 0.4, less than the threshold value of 0.5, the device is determined to be an abnormal device.
  • the specific value of the threshold may be determined by the configuration. This embodiment does not limit the specific size of the threshold.
  • the device determines whether the device is an abnormal device according to an absolute value of the service processing failure rate of the device.
  • whether the device is an abnormal device may be determined according to the relationship between the absolute value of the service processing failure rate and the threshold. For example, if the content is only on the CDF entity 1, and the service failure rate of the CDF entity 1 for the content is greater than the threshold, the device is determined to be an abnormal device.
  • the service processing failure rate of the device is 0.6.
  • the absolute value of the service processing failure rate is 0.6. If the value is less than the threshold value of 0.5, the device is determined to be an abnormal device.
  • the value of the value The volume value can be determined by the configuration. This embodiment does not limit the specific size of the threshold.
  • the abnormal devices referred to in this embodiment and the subsequent embodiments all refer to devices whose actual processing capability is lower than their own processing capabilities. There may be several reasons why the actual processing capacity of the device is lower than its own processing capacity, for example, due to internal vulnerabilities, partial or total leakage of memory, human error, interface performance damage, stored file corruption, etc. This embodiment does not describe this in detail. When the processing capacity of the abnormal device is zero, the abnormal device is considered to be in a fault state.
  • the embodiment extracts the content stored only on the abnormal device to other normal devices.
  • the method on the device of the same level does not limit the number of devices copied to other normal peer devices. It can be determined according to the actual situation. For example, the content on the abnormal device can be copied to 2 or 3 other normal devices. Level equipment.
  • the present embodiment when the distribution ratio is set in each device where the content is located according to the number of abnormal devices, the present embodiment does not specifically limit the set distribution ratio, and may be set according to actual conditions.
  • the CCF entity after receiving the service request, the CCF entity can distribute the service request of the content to the normal peer device and the abnormal device, thereby reducing the number of service requests that are distributed to the abnormal device.
  • the distribution ratio is not specifically limited, and may be determined by a specific configuration.
  • the content is stored not only on the abnormal device but also on other normal devices, it is necessary to reduce the proportion of the number of service requests distributed to the abnormal device. For example, when the CDF entity 1 is not determined to be an abnormal device, if the number of service requests for which the CDF entity 1 is distributed is 50% of the total number of service requests of the content source, the CDF entity 1 is determined to be an abnormal device, and the distribution is reduced. After the ratio of the number of service requests to the CDF entity 1 is reduced, the number of service requests that are distributed by the CDF entity 1 is reduced to 20% of the total number of service requests. This embodiment does not limit the reduction ratio, and can be set according to actual conditions.
  • the service processing success rate of the abnormal device or the number of service processing success times is zero in consecutive multiple measurement periods, or the service processing failure rate of the abnormal device is 1 in consecutive measurement periods, or abnormality
  • the number of service processing failures of the device is equal to the total number of service requests that the device is distributed in a plurality of consecutive measurement periods, the abnormal device is placed in a fault state, and the distribution of the service request to the abnormal device is stopped.
  • This embodiment does not limit the number of consecutive multiple measurement periods, and may be set according to configuration, for example, three consecutive measurement periods.
  • the current service processing information of each device needs to be sorted.
  • the service success rate or the number of successes of the first N-bit device is high, or the service failure rate or service failure of the first N-bit device is high.
  • the number of times is low. Therefore, if the service processing information is the service processing success rate or the number of service processing success times, the service processing success rate or the number of business processing success times is ranked from high to low.
  • the service processing failure rate or the number of service processing failures are arranged in the order of the service processing failure rate or the number of service processing failures from low to high, thereby ensuring that the first N-bit devices have higher service processing capabilities. This embodiment does not limit the specific numerical value of N.
  • the second selection mode needs to be sorted according to the current service processing information of each device, and since the service processing information of each device may be updated after each measurement period ends, it is required to be After the measurement period ends, the updated service processing information of each device in which the content is located is reordered, and a device response service request is selected according to the sorted result of the reordering of each device where the content is located.
  • the CCF entity sends the received service request to the selected CDF entity
  • the CCF entity selects a CDF entity to perform service request distribution according to the service processing information of each CDF entity, and can obtain, according to the foregoing step 301, the service processing success rate or the number of service processing success times of the selected CDF entity. High, or the business processing failure rate or the number of business processing failures is low, that is, the selected CDF entity has a higher service processing capability. Therefore, the received service request is sent to the selected CDF entity to ensure the service. Success rate.
  • the selected CDF entity processes the service request, and returns a service request response to the CCF entity after the processing ends;
  • the step is a preferred step, and the service request response carries a service processing success identifier or a service processing failure identifier; since the selected CDF entity ends processing the service request, the CCF entity The body returns a service request response. Therefore, the CCF entity can learn whether the selected CDF entity processes the service successfully according to the service processing success identifier or the service processing failure identifier in the received service request response, so that the selected CDF entity can be obtained. Business processing information.
  • the service processing information of the selected device is updated according to the number of times the service request is received by the selected device, and the number of service processing successes or service processing failures.
  • the service processing information in this embodiment is a service processing success rate. Therefore, after a measurement period ends, the service processing success rate of the selected device is calculated according to the number of times the service request is received and the number of service processing success times. Then, the calculation obtains the service processing success rate as the business processing information in the next measurement period, that is, the service processing information of the selected device is updated.
  • the CCF entity in the above step 301 needs to process the service after the CDF entity is updated according to the service processing information of the CDF entity 1 and the CDF entity 1 where the content is located, and when a CDF entity responds to the service request. Information is selected.
  • the method provided in this embodiment can not only improve the success rate of service processing, but also distributes the service request according to the number of service success times, or the service processing success rate, or the number of service failures, or the service processing failure rate.
  • the method provided in this embodiment also supports copying the content to other normal peer devices, thereby ensuring that the content is only stored on an abnormal device. The success rate of the business.
  • this embodiment provides a content distribution system, where the system includes: a scheduling server 401 and a content distribution device 402;
  • a scheduling server 401 configured to receive a service request for accessing content;
  • the service processing information of the distribution device 402 is selected, and the content distribution device 402 is selected to respond to the service request, and the service processing information includes one of a service processing success number, a service processing success rate, a service processing failure number, and a service processing failure rate;
  • the service request is sent to the selected content distribution device 402;
  • the content distribution device 402 is configured to receive a service request sent by the scheduling server 401.
  • the content distribution device 402 is further configured to: after processing the received service request, return a service request response to the scheduling server 401;
  • the scheduling server 401 is further configured to receive the service request response returned by the content distribution device 402, and update the service processing information of the selected content distribution device 502 according to the service request response returned by the selected content distribution device 402;
  • the scheduling server 401 selects a content distribution device 402 to respond to the service request according to the service processing information of the content distribution device 402 where the content is located
  • the scheduling server 401 is specifically configured to use the updated service processing information according to each content distribution device 402 where the content is located.
  • a content distribution device 502 is selected to respond to the service request.
  • the content distribution system provided in this embodiment can not only improve the service processing by the scheduling server according to the number of times the service distribution device processes the service success, or the service processing success rate, or the number of service failures, or the service processing failure rate. Success rate can also avoid waste of resources and reduce unnecessary business losses.
  • the embodiment provides a scheduling server, where the scheduling server includes: a first receiving module 501, configured to receive a service request for accessing content;
  • the selecting module 502 is configured to select a device to respond to the service request according to the service processing information of each device where the content is located;
  • the service processing information is one of a service processing success number, a service processing success rate, a service processing failure number, and a service processing failure rate;
  • the sending module 503 is configured to send the service request received by the first receiving module 501 to the device selected by the selecting module 502.
  • the scheduling server further includes:
  • the second receiving module 504 is configured to receive a service request returned by the device selected by the selecting module 502;
  • An update module 505, configured to update service processing information of the selected device according to the service request response returned by the selected device;
  • the selecting module 502 is specifically configured to select a device to respond to the service request according to the updated service processing information of each device where the content is located.
  • the update module 505 includes:
  • the receiving unit 5051 is configured to receive, in a measurement period, a service request response returned by the selected device, where the service request response carries a service processing success identifier or a service processing failure identifier; and a recording unit 5052 is configured to record the selected device to receive The number of times the service request is received, and the number of business processing successes or business processing failures;
  • the updating unit 5053 is configured to update the service processing information of the selected device according to the number of times the selected device receives the service request, and the number of times of the service processing success or the number of service processing failures after the end of one measurement period.
  • the selection module 502 includes:
  • the determining unit 5021 is configured to determine, according to the relationship between the service processing information of each device where the content is located and the preset threshold value, after an end of a measurement period, whether an abnormal device exists;
  • the setting unit 5022 is configured to set a distribution ratio according to the number of abnormal devices in each device where the content is located; and select a device to respond to the service request.
  • the determining unit 5021 is configured to: when the difference of the service processing information between the devices where the content is located is higher than a preset threshold, determine that the device with low service processing information is an abnormal device; or, prepare; or The device that has failed the service processing failure or the service processing failure rate is determined to be an abnormal device.
  • This embodiment does not specifically limit the size of the preset threshold.
  • the selection module 502 further includes:
  • the copying unit 5024 is configured to copy the content to other normal peer devices when it is determined that the abnormal device exists, and the content is only stored on the abnormal device.
  • the setting unit 5022 is configured to: when the service processing success rate of the abnormal device or the number of service processing success times is zero in consecutive multiple measurement periods, or the service processing failure rate of the abnormal device is continuous in multiple measurement periods. 1 , or the number of service processing failures is equal to the total number of service requests that the device is distributed in a plurality of consecutive measurement cycles, then the abnormal device is placed in a fault state, and the stop is stopped. The abnormal device distributes the business request.
  • the selecting module 502 includes:
  • the sorting unit 5025 is configured to sort the service processing information of each device where the content is located, and the second selecting unit 5026 is configured to: in the order of the current service processing information of each device where the content is located, in the order of the first N bits. Select one device to respond to the service request, and N is a natural number less than the total number of all devices in the content.
  • the scheduling server provided in this embodiment can not only improve the success rate of the service processing but also improve the success rate of the service processing by distributing the service request according to the number of service success times, or the service processing success rate, or the number of service failures, or the service processing failure rate.
  • the resource is wasted, and the unnecessary service loss is reduced.
  • the scheduling server provided in this embodiment also supports copying the content to other normal peer devices, thereby ensuring the service. Success rate.
  • each functional module is only an example.
  • the foregoing functions may be allocated according to requirements, such as configuration requirements of corresponding hardware or convenience of implementation of software.
  • Different functional modules are completed, that is, the internal structure of the scheduling server is divided into different functional modules to complete all or part of the functions described above.
  • the corresponding functional modules in this embodiment may be implemented by corresponding hardware, or may be executed by corresponding hardware.
  • the foregoing sending module 503 may be hardware having the foregoing functions.
  • a transmitter having the transmission capability, and other general-purpose transmitting devices capable of performing the foregoing functions may also be a general transmitting device capable of executing a corresponding computer program to perform the aforementioned functions, or other hardware devices.
  • the serial numbers of the embodiments of the present invention are merely for the description, and do not represent the advantages and disadvantages of the embodiments.
  • All or part of the steps in the embodiments of the present invention may be implemented by software, and the corresponding software program may be stored in a readable storage medium such as an optical disk or a hard disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Hardware Design (AREA)
  • General Engineering & Computer Science (AREA)
  • Information Transfer Between Computers (AREA)

Description

内容分发的方法、 系统及调度服务器
本申请要求于 2010 年 1 月 22 日提交中国专利局、 申请号为 201010104411. 7 , 发明名称为 "内容分发的方法、 系统及调度服务器" 的中 国专利申请的优先权, 其全部内容通过引用结合在本申请中。 技术领域 本发明涉及多媒体技术领域, 特别涉及一种内容分发的方法、 系统及调 度服务器。 背景技术 在 CDN ( Content De l ivery Network, 内容分发网络) 网络架构中, 一个 CDNCF ( Content De l ivery Network Control Funct ion, CDN控制功能) 实体 控制多个 CCF ( Clus ter Control ler Funct ion , 集群控制功能) 实体, 一个 CCF实体又控制多个 CDF ( Content De l ivery Funct ion, 内容分发功能) 实体。 其中, CDNCF实体和 CCF实体本质上是调度服务器, 而 CDF实体存储用户所需的 媒体内容源, 本质上是内容分发设备, 多个 CDF实体组成一个集群, 负责直接 给用户传送媒体内容。
当用户请求一个内容时, 为了最大限度地为用户提供服务, 现有技术在 针对该业务请求向 CDF实体分发内容的业务请求时, 需要考虑多个 CDF实体之 间的负荷均衡, 按照负荷均衡的原则进行业务请求的分发。 同时, 在 CDF实体 出现故障不能提供服务时, 尽量使用可以提供服务的 CDF实体来为用户提供服 务, 而后续将不再向故障的 CDF实体分发业务请求。
在实现本发明的过程中, 发明人发现现有技术至少存在以下缺点: 在实际应用中, 有些 CDF实体虽然还能提供服务, 但实际处理能力下降, 如果仍按照现有技术的负荷均衡原则分发业务请求, 将可能导致部分业务得 不到及时服务, 而同时由于 CDN网络中还有其他 CDF实体可用, 也将导致资源 没有得到充分的利用。 发明内容 为了减少业务损失, 提高业务成功率及用户体验, 同时充分利用资源, 本发明实施例提供了一种内容分发的方法、 系统及调度服务器。 所述技术方 案如下:
一方面, 提供了一种内容分发的方法, 所述方法包括:
接收访问内容的业务请求;
根据所述内容所在的各个设备的业务处理信息, 选择一个设备响应所述 业务请求, 所述业务处理信息为业务处理成功次数、 业务处理成功率、 业务 处理失败次数、 业务处理失败率中的一个;
将接收到的所述业务请求发送给被选择的设备。
另一方面, 提供了一种内容分发的系统, 所述系统包括: 调度服务器和 内容分发设备;
所述调度服务器, 用于接收访问内容的业务请求; 根据所述内容所在的 各个内容分发设备的业务处理信息, 选择一个内容分发设备响应所述业务请 求, 所述业务处理信息为业务处理成功次数、 业务处理成功率、 业务处理失 败次数、 业务处理失败率中的一个; 并将接收到的所述业务请求发送给被选 择的内容分发设备;
所述内容分发设备, 用于接收所述调度服务器发送的业务请求。
还提供了一种调度服务器, 所述调度服务器包括:
第一接收模块, 用于接收访问内容的业务请求;
选择模块, 用于根据所述内容所在的各个设备的业务处理信息, 选择一 个设备响应所述业务请求, 所述业务处理信息为业务处理成功次数、 业务处 理成功率、 业务处理失败次数、 业务处理失败率中的一个;
发送模块, 用于将所述第一接收模块接收到的业务请求发送给所述选择 模块选择的设备。
本发明实施例提供的技术方案的有益效果是:
通过根据设备处理业务成功次数, 或业务处理成功率, 或处理业务失败 次数, 或者业务处理失败率来分发业务请求, 不仅可以提高业务处理的成功 率, 还能避免资源浪费, 减少不必要的业务损失, 进而可为用户提供更好的 服务。 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是本发明实施例一提供的内容分发的方法流程图;
图 2是本发明实施例二提供的网络结构示意图;
图 3是本发明实施例二提供的内容分发的方法流程图;
图 4是本发明实施例三提供的内容分发的系统结构示意图;
图 5是本发明实施例四提供的调度服务器结构示意图;
图 6是本发明实施例四提供的另一种调度服务器结构示意图;
图 7是本发明实施例四提供的更新模块结构示意图;
图 8是本发明实施例四提供的第一种选择模块结构示意图;
图 9是本发明实施例四提供的第二种选择模块结构示意图;
图 10是本发明实施例四提供的第三种选择模块结构示意图。 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。
实施例一
参见图 1 , 本实施例提供了一种内容分发的方法, 该方法流程具体如下: 101: 接收访问内容的业务请求;
102: 根据该内容所在的各个设备的业务处理信息, 选择一个设备响应业 务请求;
其中, 业务处理信息为业务处理成功次数、 业务处理成功率、 业务处理 失败次数、 业务处理失败率中的一个;
103: 将接收到的业务请求发送给被选择的设备。
本实施例提供的方法, 通过根据设备处理业务成功次数, 或业务处理成 功率, 或处理业务失败次数, 或者业务处理失败率来分发业务请求, 不仅可 以提高业务处理的成功率, 还能避免资源浪费, 减少不必要的业务损失, 进 而可为用户提供更好的服务。 实施例二
本实施例提供了一种内容分发的方法, 该方法通过根据设备的业务处理 信息来调整分发策略, 从而达到提高业务成功率, 避免资源浪费的效果。 其 中, 业务处理信息为业务处理成功次数、 业务处理成功率、 业务处理失败次 数、 业务处理失败率中的一个, 本实施例以业务处理信息为业务处理成功率 为例进行说明。 业务处理成功率或失败率可以通过现有算法实现, 本实施例 对此不作具体限定, 保证对于同一内容所在的各个设备, 在计算各自的业务 处理成功率或业务处理失败率时, 釆用统一算法即可。 此处仅以业务处理成 功次数与业务请求次数的比值作为业务处理成功率为例, 例如, 对于 CDF 实 体 1 , 接收到内容 A的业务请求次数为 5 , 业务处理成功次数为 2 , 则该 CDF 实体 1的业务处理成功率为 2/ 5=0. 4。 又例如, 以业务处理失败次数与业务请 求次数的比值作为业务处理失败率, 例如, 对于 CDF实体 1 , 接收到内容 A的 业务请求次数为 5 , 业务处理失败次数为 3 , 则该 CDF实体 1的业务处理失败 率为 3/ 5=0. 6。
为了便于说明, 本实施例以用户希望获取的内容为 A , 即针对分发内容 A 的业务请求为例, 结合图 2所示的网络结构, 并以根据设备处理业务成功率来 调整分发策略为例, 对本实施例提供的内容分发的方法进行说明。
如图 2所示, 一个 CDNCF实体控制多个 CCF实体, 当 CDNCF实体接收到访问 某个内容的业务请求后, CDNCF实体将在多个 CCF实体中选择一个最适合的 CCF 实体; 又由于一个 CCF实体控制多个 CDF实体, 则被选择的 CCF实体将在其控制 的多个 CDF实体中选择一个最适合的 CDF实体, 由该被选择的 CDF实体负责直接 给用户传送媒体内容。 由此可见, C匪 CF实体、 CCF实体本质上是调度服务器, CDF实体是直接给用户提供内容的内容分发设备。 无论是 CDNCF实体, 还是 CCF 实体, 均可以进行内容 A的业务请求分发, 本实施例以 CCF实体向 CDF实体分发 内容 A的业务请求, 存储内容 A的设备分别为 CDF实体 1和 CDF实体 2为例, 对本 实施例提供的内容分发的方法进行详细说明。 参见图 3 , 方法流程具体如下:
301: CCF实体接收访问内容 A的业务请求, 并根据内容 A所在的 CDF实 体 1和 CDF实体 1的业务处理信息, 选择一个 CDF实体响应该业务请求; 具体地, 在根据内容 A所在的 CDF实体 1和 CDF实体 2的业务处理信息, 选择一个 CDF实体响应该业务请求时, 包括但不限于如下两种选择方式: 第一种选择方式: 首先, 在一个测量周期结束之后, 根据内容 A所在的 CDF实体 1和 CDF 实体 1的业务处理信息与预设阔值的关系, 确定是否存在异常设备;
其次, 根据异常设备个数, 在内容所在的各个设备中设定分发比例; 业务请求。
第二种选择方式:
根据内容所在的各个设备当前的业务处理信息对应的排列顺序, 在排列 顺序为前 N位的设备中选择一个设备响应业务请求, N为小于内容所在的所有 设备总数的自然数。
下面, 分别对两种选择方式进行详细解释说明:
对于第一种选择方式中的测量周期, 以及后续步骤提到的测量周期, 均 可由具体配置决定, 测量周期既可以是具体时间, 也可以是分发业务请求的 次数, 例如, 配置测量周期为 1秒钟, 或 1分钟等等, 或以分发业务请求次 数为 10作为一个测量周期, 即以每分发 10次业务请求作为一个测量周期, 本实施例不对测量周期的具体内容进行限定。 关于内容的业务请求为何会失 败, 原因有多种, 可能来自于调度服务器 CCF 实体, 也可能来自于控制调度 服务器的实体, 如控制 CCF实体的 CDNCF实体, 本实施例对内容的业务请求 失败的原因不做具体限定。
关于第一种选择方式中如何确定是否存在异常设备, 为了便于说明, 本 实施例以确定 CDF实体 1为异常设备为例进行说明。 具体地, 确定是否存在 异常设备时, 可以分为以下两种情况:
一、 内容存储在多个设备上:
将内容所在的各个设备的业务处理成功率进行排序, 并按照各个设备的 业务处理成功率的差距, 确定是否存在异常设备。 如, 设两个设备的业务处 理成功率的差值高于预设阔值时, 则判断存在异常设备, 并确定业务处理成 功率低的设备为异常设备。 例如, 对同一个内容 A的业务请求, CDF实体 1上 的业务处理成功率为 0. 5 , CDF实体 2上的业务处理成功率为 1 , 由于这两个 CDF实体的业务处理成功率的差值为 "50%" , 如果预设阔值为 40% , 由于这两 个 CDF实体的业务处理成功率的差值 50%大于预设阔值 40% , 则将确定业务处 理成功率低的 CDF实体 1为异常设备。 同理, 如果业务处理信息为业务处理 成功次数, 则针对内容存储在多个设备上的情况, 确定是否存在异常设备时, 处理方式与业务处理信息为业务处理成功率的处理方式相同, 本实施例对此 不作赘述。
可选地, 如果该种情况是根据设备的业务处理失败率来确定是否存在异 常设备的, 则在每个测量周期结束之后, 将内容所在的各个设备的业务处理 失败率进行排序, 并根据各个设备的业务处理失败率的排序结果, 确定是否 存在异常设备。 如, 设两个设备的业务处理失败率的差值高于预设阔值时, 则判断存在异常设备, 并确定业务处理失败率高的设备为异常设备。 例如, 对同一个内容 A的业务请求, CDF实体 1上的业务处理失败率为 0. 5 , CDF实 体 1上的业务处理失败率为 0,由于这两个 CDF实体的业务处理成功率的差值 为 "50%" , 如果预设阔值为 40%, 由于这两个 CDF实体的业务处理成功率的差 值 50%大于预设阔值 40%, 则将确定业务处理失败率高的 CDF实体 1为异常设 备。 其中, 预设阔值可由配置决定, 本实施例不对预设阔值的具体大小进行 限定。 同理, 如果业务处理信息为业务处理失败次数, 则针对内容存储在多 个设备上的情况, 确定是否存在异常设备时, 处理方式与业务处理信息为业 务处理失败率的处理方式相同, 本实施例对此不作赘述。
二、 内容存储在一个设备上:
针对该种情况, 确定是否存在异常设备时, 则以该设备的业务处理成功 率的绝对值为依据, 确定该设备是否为异常设备。
具体地, 可按照业务处理成功率的绝对值与预设阔值的关系, 来确定设 备是否为异常设备。 例如, 设阔值为 0. 5 , 如果一个内容仅在 CDF实体 1上, 而 CDF实体 1对于该内容的业务处理成功率如果小于阔值, 则确定该设备为异常 设备。 以该设备的业务处理成功率为 0. 4为例, 由于其业务处理成功率的绝对 值为 0. 4 , 小于阔值 0. 5 , 则确定该设备为异常设备。 其中, 阔值的具体数值 可由配置决定, 本实施例不对阔值的具体大小进行限定。
可选地, 如果该种情况是根据设备的业务处理失败率来确定是否存在异 常设备的, 则以设备的业务处理失败率的绝对值为依据, 确定该设备是否为 异常设备。 具体地, 可按照业务处理失败率的绝对值与阔值的关系, 来确定 设备是否为异常设备。例如,设阔值为 0. 5 ,如果一个内容仅在 CDF实体 1上, 而 CDF实体 1对于该内容的业务处理失败率如果大于阔值, 则确定该设备为 异常设备。 以该设备的业务处理失败率为 0. 6 为例, 由于其业务处理失败率 的绝对值为 0. 6 , 小于阔值 0. 5 , 则确定该设备为异常设备。 其中, 阔值的具 体数值可由配置决定, 本实施例不对阔值的具体大小进行限定。
需要说明的是, 本实施例以及后续实施例所称的异常设备均是指实际处 理能力低于其自身应有处理能力的设备。 导致设备的实际处理能力低于其自 身应有处理能力的原因可能有多种, 例如, 由于设备内部出现漏洞、 内存部 分泄漏或全部泄漏、 人为误操作、 接口性能损坏、 存储的文件损坏等等, 本 实施例对此不做详细描述。 当异常设备的处理能力为零时, 则认为该异常设 备为故障状态。
优选地, 在确定存在异常设备后, 如果确定内容仅存储在一个异常设备 上, 则为了保证业务的成功率, 本实施例釆取了将仅存储在该异常设备上的 内容复制到其他正常的同级设备上的方式, 本实施例不对复制到其他正常的 同级设备的数量进行限定, 可根据实际情况确定, 例如, 可将异常设备上的 内容复制到 2个或 3个其他正常的同级设备。
对于第一种选择方式, 根据异常设备个数, 在内容所在的各个设备中设 定分发比例时, 本实施例不对设定的分发比例进行具体限定, 可根据实际情 况进行设定。针对内容仅存储在一个异常设备上的情况, CCF实体接收到业务 请求后, 可将内容的业务请求分发到正常的同级设备和异常设备上, 由此达 到缩减分发到异常设备的业务请求次数的目的, 本实施例不对分发比例进行 具体限定, 可由具体配置决定。
例如, 设仅有 CDF实体 1存储了内容 A , 则针对内容 A的业务请求, 在未 确定 CDF实体 1为异常设备时, 所有业务请求都分发到 CDF实体 1上; 当 CDF 实体 1被确定为异常设备时, 则需要将内容 A复制到其他正常的同级设备上, 并在下一测量周期分发对该内容 A 的业务请求时, 不再将所有业务请求分发 到 CDF实体 1上, 还需要将部分业务请求分发到被复制内容 A的其他正常的 同级设备上, 由此可见, 异常设备 CDF实体 1被分发的业务请求次数被缩减 了。
如果内容不仅存储在异常设备上, 还存储在其他正常设备上, 则需要缩 减分发到异常设备上的业务请求次数的比例。 例如, CDF实体 1未被确定为异 常设备时, 如果该 CDF实体 1被分发的业务请求次数为该内容源的总业务请 求次数的 50% , 则在确定 CDF实体 1为异常设备后, 缩减分发到 CDF实体 1的 业务请求次数的比例之后, CDF实体 1被分发的业务请求次数缩减为总业务请 求次数的 20% , 本实施例不对缩减比例进行限定, 可根据实际情况进行设定。 进一步地, 如果异常设备的业务处理成功率或业务处理成功次数在连续 多个测量周期内均为零, 或, 异常设备的业务处理失败率在连续多个测量周 期内均为 1 , 或, 异常设备的业务处理失败次数在连续多个测量周期内均与该 设备被分发的业务请求总次数相等时, 则将该异常设备置为故障状态, 并停 止向该异常设备分发业务请求。 本实施例不对连续多个测量周期的数量进行 限定, 可根据配置设定, 例如, 连续 3个测量周期。
对于第二种选择方式, 在排列顺序为前 N位的设备中选择一个设备响应 业务请求时, 需要对各个设备当前的业务处理信息进行排序, 排序的方式可 以有多种, 由于本实施例釆取的是从排列顺序为前 N位的设备中选择一个设 备响应业务请求, 则该前 N位设备的业务成功率或成功次数要高, 或者, 该 前 N位设备的业务失败率或业务失败次数要低, 因此, 在具体排序时, 如果 业务处理信息为业务处理成功率或业务处理成功次数, 则按照业务处理成功 率或业务处理成功次数由高到低的顺序排列, 如果业务处理信息是业务处理 失败率或业务处理失败次数, 则按照业务处理失败率或业务处理失败次数由 低到高的顺序排列, 由此保证前 N位设备具有较高的业务处理能力。 本实施 例不对 N的具体数值进行限定。
需要说明的是, 由于第二种选择方式中需要根据各个设备当前的业务处 理信息进行排序, 又由于各个设备的业务处理信息在每个测量周期结束之后 将可能被更新, 因此, 需要在每个测量周期结束之后, 根据内容所在的各个 设备更新后的业务处理信息重新进行排序, 并根据内容所在的各个设备重新 排序后的排序结果选择一个设备响应业务请求。
302: CCF实体将接收到的业务请求发送给被选择的 CDF实体;
其中,由于 CCF实体是根据各个 CDF实体的业务处理信息来选择一个 CDF 实体进行业务请求分发的, 从上述步骤 301 中可以得出, 被选择的 CDF实体 的业务处理成功率或业务处理成功次数较高, 或业务处理失败率或业务处理 失败次数较低, 也就是说, 被选择的 CDF 实体的业务处理能力较高, 因此, 将接收到的业务请求发送给被选择的 CDF实体, 可以保证业务成功率。
303: 被选择的 CDF实体处理该业务请求, 并在处理结束之后, 向 CCF实 体返回业务请求响应;
该步骤为优选步骤, 业务请求响应中携带了业务处理成功标识或业务处 理失败标识; 由于被选择的 CDF实体在结束处理该业务请求之后, 向 CCF实 体返回业务请求响应, 因此, CCF实体可根据接收到的业务请求响应中的业务 处理成功标识或业务处理失败标识, 获知被选择的 CDF 实体处理业务是否成 功, 从而可以得到该被选择的 CDF实体的业务处理信息。
304:接收被选择的 CDF实体返回的业务请求响应,并根据该被选择的 CDF 实体返回的业务请求响应, 更新该被选择的 CDF实体的业务处理信息。
具体地, 在根据被选择设备返回的业务请求响应, 更新该被选择设备的 业务处理信息时, 包括如下步骤:
首先, 在一个测量周期内, 接收被选择设备返回的业务请求响应, 业务 请求响应中携带了业务处理成功标识或业务处理失败标识;
其次, 记录被选择设备接收到业务请求的次数, 以及业务处理成功次数 或业务处理失败次数;
最后, 在一个测量周期结束之后, 根据被选择设备接收到业务请求的次 数, 以及业务处理成功次数或业务处理失败次数, 更新被选择设备的业务处 理信息。 例如: 本实施例的业务处理信息为业务处理成功率, 因此, 当一个 测量周期结束之后, 根据被选择设备接收到业务请求的次数和业务处理成功 次数, 来计算被选择设备的业务处理成功率, 然后将计算获得业务处理成功 率作为下一个测量周期内的业务处理信息, 即更新了被选择设备的业务处理 信息。
在该步骤的基础上, 上述步骤 301 中的 CCF 实体在根据内容所在的 CDF 实体 1和 CDF实体 1的业务处理信息, 选择一个 CDF实体响应业务请求时, 需要根据 CDF实体被更新后的业务处理信息进行选择。
综上所述, 本实施例提供的方法, 通过根据设备处理业务成功次数, 或 业务处理成功率, 或处理业务失败次数, 或者业务处理失败率来分发业务请 求, 不仅可以提高业务处理的成功率, 还能避免资源浪费, 减少不必要的业 务损失; 另外, 当内容仅存储在一个异常设备上时, 本实施例提供的方法还 支持将该内容复制到其他正常的同级设备上, 从而保证了业务的成功率。 实施例三
参见图 4 , 本实施例提供了一种内容分发系统, 该系统包括: 调度服务器 401和内容分发设备 402 ;
调度服务器 401 , 用于接收访问内容的业务请求; 根据内容所在的各个内 容分发设备 402的业务处理信息, 选择一个内容分发设备 402响应业务请求, 业务处理信息包括业务处理成功次数、 业务处理成功率、 业务处理失败次数、 业务处理失败率中的一个; 并将接收到的业务请求发送给被选择的内容分发 设备 402;
内容分发设备 402 , 用于接收调度服务器 401发送的业务请求。
进一步地, 内容分发设备 402 , 还用于在处理接收到的业务请求之后, 向 调度服务器 401返回业务请求响应;
调度服务器 401 ,还用于接收内容分发设备 402返回的业务请求响应, 并 根据被选择的内容分发设备 402返回的业务请求响应, 更新被选择的内容分 发设备 502的业务处理信息;
相应地, 调度服务器 401在根据内容所在的各个内容分发设备 402的业 务处理信息, 选择一个内容分发设备 402 响应业务请求时, 具体用于根据内 容所在的各个内容分发设备 402 更新后的业务处理信息, 选择一个内容分发 设备 502响应业务请求。
本实施例提供的内容分发的系统, 通过调度服务器根据内容分发设备处 理业务成功次数, 或业务处理成功率, 或处理业务失败次数, 或者业务处理 失败率来分发业务请求, 不仅可以提高业务处理的成功率, 还能避免资源浪 费, 减少不必要的业务损失。 实施例四
参见图 5 , 本实施例提供了一种调度服务器, 该调度服务器包括: 第一接收模块 501 , 用于接收访问内容的业务请求;
选择模块 502 , 用于根据内容所在的各个设备的业务处理信息, 选择一个 设备响应业务请求;
其中, 业务处理信息为业务处理成功次数、 业务处理成功率、 业务处理 失败次数、 业务处理失败率中的一个;
发送模块 503 ,用于将第一接收模块 501接收到的业务请求发送给选择模 块 502选择的设备。
进一步地, 参见图 6 , 该调度服务器还包括:
第二接收模块 504 ,用于接收被选择模块 502选择的设备返回的业务请求 向应; 更新模块 505 , 用于根据被选择设备返回的业务请求响应, 更新被选择设 备的业务处理信息;
相应地, 选择模块 502 , 具体用于根据内容所在的各个设备更新后的业务 处理信息, 选择一个设备响应业务请求。
具体地, 参见图 7 , 更新模块 505包括:
接收单元 5051 , 用于在一个测量周期内, 接收被选择设备返回的业务请 求响应, 该业务请求响应中携带了业务处理成功标识或业务处理失败标识; 记录单元 5052 , 用于记录被选择设备接收到业务请求的次数, 以及业务 处理成功次数或业务处理失败次数;
更新单元 5053 , 用于在一个测量周期结束之后, 根据被选择设备接收到 业务请求的次数, 以及业务处理成功次数或业务处理失败次数, 更新被选择 设备的业务处理信息。
参见图 8 , 选择模块 502包括:
确定单元 5021 , 用于在一个测量周期结束之后, 根据内容所在的各个设 备的业务处理信息与预设阔值的关系, 确定是否存在异常设备;
设定单元 5022 , 用于根据异常设备个数, 在内容所在的各个设备中设定 分发比例; 择一个设备响应业务请求。
其中, 确定单元 5021 , 用于在内容所在的各个设备之间的业务处理信息 的差值高于预设阔值时, 则确定业务处理信息低的设备为异常设备; 或者, 备; 或者, 将业务处理失败次数或业务处理失败率高于预设阔值的设备确定 为异常设备, 本实施例不对预设阔值的大小进行具体限定。
参见图 9 , 选择模块 502还包括:
复制单元 5024 , 用于在确定存在异常设备, 且内容仅存储在异常设备上 时, 则将内容复制到其他正常的同级设备上。
其中, 设定单元 5022 , 用于在异常设备的业务处理成功率或业务处理成 功次数在连续多个测量周期内均为零, 或异常设备的业务处理失败率在连续 多个测量周期内均为 1 ,或业务处理失败次数在连续多个测量周期内均与该设 备被分发的业务请求总次数相等时, 则将异常设备置为故障状态, 并停止向 异常设备分发业务请求。
可选地, 参见图 10, 选择模块 502 , 包括:
排序单元 5025 , 用于对内容所在的各个设备的业务处理信息进行排序; 第二选择单元 5026 , 用于根据内容所在的各个设备当前的业务处理信息 对应的排列顺序, 在排列顺序为前 N位的设备中选择一个设备响应业务请求, N为小于内容所在的所有设备总数的自然数。
本实施例提供的调度服务器, 通过根据设备处理业务成功次数, 或业务 处理成功率, 或处理业务失败次数, 或者业务处理失败率来分发业务请求, 不仅可以提高业务处理的成功率, 还能避免资源浪费, 减少不必要的业务损 失; 另外, 当内容仅存储在一个异常设备上时, 本实施例提供的调度服务器 还支持将该内容复制到其他正常的同级设备上, 从而保证了业务的成功率。
需要说明的是, 以上调度服务器的实施方式中, 各功能模块的划分仅是 举例说明, 实际应用中可以根据需要, 比如相应硬件的配置要求或者软件的 实现的便利考虑, 而将上述功能分配由不同的功能模块完成, 即将所述调度 服务器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。 而且实际应用中, 本实施例中的相应的功能模块可以是由相应的硬件 实现,也可以由相应的硬件执行相应的软件完成,例如,前述的发送模块 503 , 可以为具有执行前述功能的硬件, 比如, 具有该发送能力的发送器, 以及其 他通用的能够执行前述功能的发送装置, 也可以是能够执行相应计算机程序 从而完成前述功能的一般发送设备, 或者其他硬件设备。 上述本发明实施例序号仅仅为了描述, 不代表实施例的优劣。
本发明实施例中的全部或部分步骤, 可以利用软件实现, 相应的软件程 序可以存储在可读取的存储介质中, 如光盘或硬盘等。
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

权利要求
1、 一种内容分发的方法, 其特征在于, 所述方法包括:
接收访问内容的业务请求;
根据所述内容所在的各个设备的业务处理信息, 选择一个设备响应所述 业务请求, 所述业务处理信息为业务处理成功次数、 业务处理成功率、 业务 处理失败次数、 业务处理失败率中的一个;
将接收到的所述业务请求发送给被选择的设备。
2、 根据权利要求 1所述的方法, 其特征在于, 所述根据所述内容所在的 各个设备的业务处理信息, 选择一个设备响应所述业务请求之后, 还包括: 接收被选择设备返回的业务请求响应, 并根据所述被选择设备返回的业 务请求响应, 更新所述被选择设备的业务处理信息;
相应地, 根据所述内容所在的各个设备的业务处理信息, 选择一个设备 响应所述业务请求, 具体包括:
根据所述内容所在的各个设备更新后的业务处理信息, 选择一个设备响 应所述业务请求。
3、 根据权利要求 2所述的方法, 其特征在于, 所述根据所述被选择设备 返回的业务请求响应, 更新所述被选择设备的业务处理信息, 包括:
在一个测量周期内, 接收所述被选择设备返回的业务请求响应, 所述业 务请求响应中携带了业务处理成功标识或业务处理失败标识;
记录所述被选择设备接收到所述业务请求的次数, 以及业务处理成功次 数或业务处理失败次数;
在一个测量周期结束之后, 根据所述被选择设备接收到所述业务请求的 次数, 以及业务处理成功次数或业务处理失败次数, 更新所述被选择设备的 业务处理信息。
4、 根据权利要求 1所述的方法, 其特征在于, 所述根据所述内容所在的 各个设备的业务处理信息, 选择一个设备响应所述业务请求, 包括:
在一个测量周期结束之后, 根据所述内容所在的各个设备的业务处理信 息与预设阔值的关系, 确定是否存在异常设备; 根据异常设备个数, 在所述内容所在的各个设备中设定分发比例; 业务请求。
5、 根据权利要求 4所述的方法, 其特征在于, 所述根据所述内容所在的 各个设备的业务处理信息与预设阔值的关系, 确定是否存在异常设备, 包括: 如果所述内容所在的各个设备之间的业务处理信息的差值高于预设阔 值, 则确定业务处理信息低的设备为异常设备;
或者, 将业务处理成功次数或业务处理成功率低于预设阔值的设备确定 为异常设备;
或者, 将业务处理失败次数或业务处理失败率高于预设阔值的设备确定 为异常设备。
6、 根据权利要求 4所述的方法, 其特征在于, 所述确定是否存在异常设 备之后, 还包括:
如果确定存在异常设备, 且所述内容仅存储在所述异常设备上, 则将所 述内容复制到其他正常的同级设备上。
7、 根据权利要求 4所述的方法, 其特征在于, 所述根据异常设备个数, 在所述内容所在的各个设备中设定分发比例, 包括:
如果异常设备的业务处理成功率或业务处理成功次数在连续多个测量周 期内均为零, 或所述异常设备的业务处理失败率在连续多个测量周期内均为 1 , 或业务处理失败次数在连续多个测量周期内均与该设备被分发的业务请求 总次数相等时, 则将所述异常设备置为故障状态, 并停止向所述异常设备分 发业务请求。
8、 根据权利要求 1所述的方法, 其特征在于, 所述根据所述内容所在的 各个设备的业务处理信息, 选择一个设备响应所述业务请求, 包括:
根据所述内容所在的各个设备当前的业务处理信息对应的排列顺序, 在 排列顺序为前 N位的设备中选择一个设备响应所述业务请求, N为小于所述内 容所在的所有设备总数的自然数。
9、 根据权利要求 8所述的方法, 其特征在于, 所述在排列顺序为前 N位 的设备中选择一个设备响应所述业务请求之后, 还包括:
在每个测量周期结束之后, 根据所述内容所在的各个设备更新后的业务 处理信息重新进行排序, 并根据所述内容所在的各个设备重新排序后的排序 结果选择一个设备响应业务请求。
10、 一种内容分发系统, 其特征在于, 所述系统包括: 调度服务器和内 容分发设备;
所述调度服务器, 用于接收访问内容的业务请求; 根据所述内容所在的 各个内容分发设备的业务处理信息, 选择一个内容分发设备响应所述业务请 求, 所述业务处理信息为业务处理成功次数、 业务处理成功率、 业务处理失 败次数、 业务处理失败率中的一个; 并将接收到的所述业务请求发送给被选 择的内容分发设备;
所述内容分发设备, 用于接收所述调度服务器发送的业务请求。
11、 根据权利要求 10所述的系统, 其特征在于,
所述内容分发设备, 还用于在处理接收到的所述业务请求之后, 向所述 调度服务器返回业务请求响应;
所述调度服务器, 还用于接收所述内容分发设备返回的业务请求响应, 并根据被选择的内容分发设备返回的业务请求响应, 更新所述被选择的内容 分发设备的业务处理信息;
相应地, 所述调度服务器在根据所述内容所在的各个内容分发设备的业 务处理信息, 选择一个内容分发设备响应所述业务请求时, 具体用于根据所 述内容所在的各个内容分发设备更新后的业务处理信息, 选择一个内容分发 设备响应所述业务请求。
12、 一种调度服务器, 其特征在于, 所述调度服务器包括:
第一接收模块, 用于接收访问内容的业务请求;
选择模块, 用于根据所述内容所在的各个设备的业务处理信息, 选择一 个设备响应所述业务请求, 所述业务处理信息为业务处理成功次数、 业务处 理成功率、 业务处理失败次数、 业务处理失败率中的一个; 发送模块, 用于将所述第一接收模块接收到的业务请求发送给所述选择 模块选择的设备。
1 3、 根据权利要求 12所述的调度服务器, 其特征在于, 所述调度服务器 还包括:
第二接收模块, 用于接收所述被选择模块选择的设备返回的业务请求响 应;
更新模块, 用于根据被选择设备返回的业务请求响应, 更新所述被选择 设备的业务处理信息;
相应地, 所述选择模块, 具体用于根据所述内容所在的各个设备更新后 的业务处理信息, 选择一个设备响应所述业务请求。
14、 根据权利要求 1 3所述的调度服务器, 其特征在于, 所述更新模块, 包括:
接收单元, 用于在一个测量周期内, 接收所述被选择设备返回的业务请 求响应, 所述业务请求响应中携带了业务处理成功标识或业务处理失败标识; 记录单元, 用于记录所述被选择设备接收到所述业务请求的次数, 以及 业务处理成功次数或业务处理失败次数;
更新单元, 用于在一个测量周期结束之后, 根据所述被选择设备接收到 所述业务请求的次数, 以及业务处理成功次数或业务处理失败次数, 更新所 述被选择设备的业务处理信息。
15、 根据权利要求 12所述的调度服务器, 其特征在于, 所述选择模块, 包括:
确定单元, 用于在一个测量周期结束之后, 根据所述内容所在的各个设 备的业务处理信息与预设阔值的关系, 确定是否存在异常设备;
设定单元, 用于根据异常设备个数, 在所述内容所在的各个设备中设定 分发比例; 择一个设备响应所述业务请求。
16、 根据权利要求 15所述的调度服务器, 其特征在于, 所述选择模块, 还包括:
复制单元, 用于在确定存在异常设备, 且所述内容仅存储在所述异常设 备上时, 则将所述内容复制到其他正常的同级设备上。
17、 根据权利要求 15所述的调度服务器, 其特征在于, 所述设定单元, 用于在异常设备的业务处理成功率或业务处理成功次数在连续多个测量周期 内均为零, 或所述异常设备的业务处理失败率在连续多个测量周期内均为 1 , 或业务处理失败次数在连续多个测量周期内均与该设备被分发的业务请求总 次数相等时, 则将所述异常设备置为故障状态, 并停止向所述异常设备分发 业务请求。
18、 根据权利要求 12所述的调度服务器, 其特征在于, 所述选择模块, 包括:
排序单元, 用于对所述内容所在的各个设备的业务处理信息进行排序; 第二选择单元, 用于根据所述内容所在的各个设备当前的业务处理信息 对应的排列顺序, 在排列顺序为前 N位的设备中选择一个设备响应所述业务 请求, 所述 N为小于所述内容所在的所有设备总数的自然数。
PCT/CN2011/070207 2010-01-22 2011-01-12 内容分发的方法、系统及调度服务器 WO2011088767A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP11734349.1A EP2515504B1 (en) 2010-01-22 2011-01-12 Content delivery method, system and schedule server
US13/546,592 US8443054B2 (en) 2010-01-22 2012-07-11 Method, system, and scheduling server for content delivery

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201010104411.7A CN102137133B (zh) 2010-01-22 2010-01-22 内容分发的方法、系统及调度服务器
CN201010104411.7 2010-01-22

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US13/546,592 Continuation US8443054B2 (en) 2010-01-22 2012-07-11 Method, system, and scheduling server for content delivery

Publications (1)

Publication Number Publication Date
WO2011088767A1 true WO2011088767A1 (zh) 2011-07-28

Family

ID=44296788

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/070207 WO2011088767A1 (zh) 2010-01-22 2011-01-12 内容分发的方法、系统及调度服务器

Country Status (4)

Country Link
US (1) US8443054B2 (zh)
EP (2) EP2515504B1 (zh)
CN (1) CN102137133B (zh)
WO (1) WO2011088767A1 (zh)

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9100549B2 (en) * 2008-05-12 2015-08-04 Qualcomm Incorporated Methods and apparatus for referring media content
CN103944927B (zh) * 2013-01-17 2018-07-17 深圳市腾讯计算机系统有限公司 服务器伸缩处理方法和装置
EP3089435B1 (en) 2014-01-20 2023-08-30 Huawei Technologies Co., Ltd. Service processing method and network device
US9354963B2 (en) * 2014-02-26 2016-05-31 Microsoft Technology Licensing, Llc Service metric analysis from structured logging schema of usage data
CN105721394B (zh) * 2014-12-03 2019-02-05 中国移动通信集团公司 一种面向内容分发网络的业务接入设备、方法及系统
US10298713B2 (en) * 2015-03-30 2019-05-21 Huawei Technologies Co., Ltd. Distributed content discovery for in-network caching
CN104834722B (zh) * 2015-05-12 2018-03-02 网宿科技股份有限公司 基于cdn的内容管理系统
CN105847864B (zh) * 2016-05-13 2019-11-05 合一网络技术(北京)有限公司 一种推荐视频清晰度的方法及装置
CN107547236B (zh) * 2016-06-29 2020-07-31 南京中兴软件有限责任公司 告警处理能力模型建立、告警信息处理的方法和装置
CN106162233B (zh) * 2016-07-08 2019-12-31 合一网络技术(北京)有限公司 码率推荐方法及装置
CN107769943B (zh) * 2016-08-17 2021-01-08 阿里巴巴集团控股有限公司 一种主备集群切换的方法和设备
CN106878169B (zh) * 2016-08-18 2020-08-04 阿里巴巴集团控股有限公司 一种业务路由方法和装置
CN106357783B (zh) * 2016-09-29 2019-06-14 北京奇艺世纪科技有限公司 一种边缘节点分配方法及装置
CN107105037B (zh) * 2017-04-25 2020-06-05 上海幻电信息科技有限公司 一种基于文件校验的分布式视频cdn资源管理系统及方法
CN110266741B (zh) * 2018-03-12 2020-06-30 贵州白山云科技股份有限公司 一种内容分发网络中的客户业务自动调度方法及装置
CN114236240B (zh) * 2022-02-25 2023-10-20 荣耀终端有限公司 进液检测电路和电子设备

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6249800B1 (en) * 1995-06-07 2001-06-19 International Business Machines Corporartion Apparatus and accompanying method for assigning session requests in a multi-server sysplex environment
CN1512363A (zh) * 2002-12-31 2004-07-14 联想(北京)有限公司 提高商务机群可服务性的方法
CN1856137A (zh) * 2005-04-27 2006-11-01 华为技术有限公司 一种确定集中控制服务器的方法及系统

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6173322B1 (en) * 1997-06-05 2001-01-09 Silicon Graphics, Inc. Network request distribution based on static rules and dynamic performance data
US6006264A (en) * 1997-08-01 1999-12-21 Arrowpoint Communications, Inc. Method and system for directing a flow between a client and a server
US6141759A (en) * 1997-12-10 2000-10-31 Bmc Software, Inc. System and architecture for distributing, monitoring, and managing information requests on a computer network
US6938256B2 (en) * 2000-01-18 2005-08-30 Galactic Computing Corporation System for balance distribution of requests across multiple servers using dynamic metrics
US6747957B1 (en) * 2000-04-28 2004-06-08 Cisco Technology, Inc. Network availability monitor
US7562153B2 (en) * 2000-05-12 2009-07-14 AT&T Intellectual Property II, L. P. Method and apparatus for content distribution network brokering and peering
US6795858B1 (en) * 2000-12-29 2004-09-21 Cisco Technology, Inc. Method and apparatus for metric based server selection
US20020120743A1 (en) * 2001-02-26 2002-08-29 Lior Shabtay Splicing persistent connections
US20050005019A1 (en) * 2003-05-19 2005-01-06 Michael Harville Service management using multiple service location managers
EP1927921A1 (en) * 2003-08-08 2008-06-04 Teamon Systems, Inc. Communications system providing server load balancing based upon weighted health metrics and related method
JP4617847B2 (ja) * 2004-11-04 2011-01-26 株式会社日立製作所 情報処理システム及びアクセス方法
CN100466625C (zh) 2006-09-07 2009-03-04 华为技术有限公司 一种实现业务流量控制的方法及系统
US8065429B2 (en) * 2007-06-28 2011-11-22 Nokia Corporation System, apparatus and method for associating an anticipated success indication with data delivery
US7930393B1 (en) * 2008-09-29 2011-04-19 Amazon Technologies, Inc. Monitoring domain allocation performance
CN101521683B (zh) * 2009-03-19 2012-07-25 深圳市新飞扬数码技术有限公司 一种网络游戏系统及其接入方法、客户端通讯方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6249800B1 (en) * 1995-06-07 2001-06-19 International Business Machines Corporartion Apparatus and accompanying method for assigning session requests in a multi-server sysplex environment
CN1512363A (zh) * 2002-12-31 2004-07-14 联想(北京)有限公司 提高商务机群可服务性的方法
CN1856137A (zh) * 2005-04-27 2006-11-01 华为技术有限公司 一种确定集中控制服务器的方法及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2515504A4 *

Also Published As

Publication number Publication date
EP2804362A1 (en) 2014-11-19
EP2515504A4 (en) 2013-03-06
EP2515504B1 (en) 2015-10-28
US20120278451A1 (en) 2012-11-01
US8443054B2 (en) 2013-05-14
EP2515504A1 (en) 2012-10-24
CN102137133A (zh) 2011-07-27
CN102137133B (zh) 2015-01-21

Similar Documents

Publication Publication Date Title
WO2011088767A1 (zh) 内容分发的方法、系统及调度服务器
US9495392B2 (en) System and method for parallel multiplexing between servers in a cluster
US9733983B2 (en) System and method for surge protection and rate acceleration in a traffic director environment
US8886690B2 (en) Distributed data storage and access systems
US8838535B2 (en) Providing services across systems that manage distributed replicas
US10382380B1 (en) Workload management service for first-in first-out queues for network-accessible queuing and messaging services
US9703610B2 (en) Extensible centralized dynamic resource distribution in a clustered data grid
US20100010999A1 (en) Data Access in Distributed Systems
US8954976B2 (en) Data storage in distributed resources of a network based on provisioning attributes
AU2004266017B2 (en) Hierarchical management of the dynamic allocation of resources in a multi-node system
JP2013525931A (ja) コンテンツ配信に利用される動的バインド
US10785350B2 (en) Heartbeat in failover cluster
CN105025053A (zh) 基于云存储技术的分布式文件的上传方法及其系统
WO2021057108A1 (zh) 一种读数据方法、写数据方法及服务器
WO2010006134A2 (en) Distributed data storage and access systems
CA3030504A1 (en) Blockchain network and task scheduling method therefor
CN105242983A (zh) 一种数据存储方法以及一种数据存储管理服务器
KR20160097372A (ko) 데이터 센터들에서의 효율적인 자원 이용
CN115834587A (zh) 一种选择目标存储服务器的方法、装置及电子设备
CN115378962B (zh) 一种基于iSCSI协议的存储集群的高可用性连通方法和系统
CN104468674B (zh) 数据迁移方法及装置
WO2023029485A1 (zh) 数据处理方法、装置、计算机设备及计算机可读存储介质
CN114064362B (zh) 用于分布式存储中的数据恢复方法、系统及计算机可读存储介质
Li et al. High-Available Cloud Platform Based on OpenStack
JP5464746B2 (ja) データベースを分散共有する資源管理装置、プログラム及び方法

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: 11734349

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2011734349

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE