WO2010115325A1 - 分布式文件系统中业务内容处理、更新方法与装置 - Google Patents

分布式文件系统中业务内容处理、更新方法与装置 Download PDF

Info

Publication number
WO2010115325A1
WO2010115325A1 PCT/CN2009/073330 CN2009073330W WO2010115325A1 WO 2010115325 A1 WO2010115325 A1 WO 2010115325A1 CN 2009073330 W CN2009073330 W CN 2009073330W WO 2010115325 A1 WO2010115325 A1 WO 2010115325A1
Authority
WO
WIPO (PCT)
Prior art keywords
file block
virtual
service node
attribute list
service
Prior art date
Application number
PCT/CN2009/073330
Other languages
English (en)
French (fr)
Inventor
夏舰波
彭杰
张波
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to US13/257,209 priority Critical patent/US20120016916A1/en
Priority to EP09842886.5A priority patent/EP2418809A4/en
Publication of WO2010115325A1 publication Critical patent/WO2010115325A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/06—Protocols specially adapted for file transfer, e.g. file transfer protocol [FTP]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/10—File systems; File servers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/10—File systems; File servers
    • G06F16/18—File system types
    • G06F16/182—Distributed file systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1095—Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]

Definitions

  • the present invention relates to a service content processing and updating technology in a distributed file system, and in particular, to a service content processing and updating method and apparatus in a service node in a distributed file system.
  • the distributed file system is widely used in servers of video services such as IPTV (Internet Protocol Television) to manage the storage and reading of service data in a single service node (serving a region or a part of users).
  • IPTV Internet Protocol Television
  • a single service node serving a region or a part of users.
  • IPTV Internet Protocol Television
  • the distributed file system is generally solved by the method of service content scheduling, that is, all the service content is distributed and stored in each service node, and all the service contents are not stored on a single service node, but on each service node.
  • the provided service content directory is the same.
  • the scheduling method of the foregoing service content can save a large amount of storage space of each service node in the distributed file system, thereby reducing the formation cost of the distributed file system.
  • the scheduling method of the service content does not have corresponding service content on the service node, and when the user has more on-demand requests for the service content on the service node, although it can provide the user from other service nodes.
  • Business services but this consumes a large bandwidth of the distributed file system.
  • the quality of the service services provided is difficult to guarantee.
  • the main object of the present invention is to provide a service node in a distributed file system.
  • the business content processing, updating method and device can improve the service quality of the hot business content, and make the occupied storage space smaller.
  • a service content processing method for a service node in a distributed file system comprising:
  • the distributed file system After receiving the uploaded service content, the distributed file system generates a file block according to the service content, where the file block is stored in at least one of the service nodes, and the service node that does not store the file block stores a virtual file block corresponding to the file block, and setting an attribute list of the virtual file block, where the attribute list includes virtual identification information, access popularity value information, and at least one access address information of the file block.
  • the virtual file block has the same name and attribute value as the file block.
  • the step of the distributed file system generating a file block according to the service content comprises: the distributed file system directly using the service content as a file block, or dividing the service content into two or more services.
  • the content block is used as a file block.
  • a method for updating service content of a service node in a distributed file system comprising:
  • the service node After receiving the on-demand request of the file block, the service node determines whether the file block to be requested is a virtual file block, and is based on at least the file block corresponding to the virtual file block in the attribute list of the virtual file block.
  • An access address information reads the file block and responds to the on-demand request, and simultaneously updates an access heat value in the attribute list of the virtual file block in the service node, and determines whether the updated access heat value reaches the set value a threshold value, when the set threshold is reached, downloading the file block to the service node according to at least one access address information of the file block, and clearing or emptying the attribute list of the virtual file block, or deleting The virtual identification information in the attribute list of the virtual file block.
  • the step of the service node determining whether the file block to be requested is a virtual file block comprises:
  • Finding whether the file block to be requested includes the attribute list or determining whether the found attribute list is non-empty according to the identifier of the file block to be requested, or determining whether the virtual attribute information is included in the searched attribute list, and determining the location
  • the on-demand file block is a virtual file block.
  • updating the access heat in the attribute list of the virtual file block in the service node includes:
  • a service content processing device for a service node in a distributed file system comprising:
  • An uploading unit configured to upload business content to a distributed file system
  • a file block generating unit configured to generate a file block from the business content uploaded by the uploading unit
  • a publishing unit configured to store the file block in at least one of the service nodes, and store, in the service node that does not store the file block, a virtual file block corresponding to the file block;
  • a setting unit configured to be the virtual file block setting attribute list, the attribute list including virtual identification information, access popularity value information, and at least one access address information of the file block.
  • the virtual file block has the same name and attribute value as the file block.
  • a service content update device for a service node in a distributed file system comprising:
  • a receiving unit configured to receive an on-demand request of the file block
  • a first determining unit configured to determine whether the file block requested by the on-demand request is a virtual file block, and trigger a reading and response unit;
  • a reading and response unit configured to read the file block and respond to the on-demand request according to at least one access address information of the file block corresponding to the virtual file block in an attribute list of the virtual file block , triggering the update unit at the same time;
  • An update unit configured to update an access heat value in the attribute list of the virtual file block, and trigger a second determining unit
  • a second determining unit configured to determine whether the value of the access heat in the attribute list of the updated virtual file block reaches a set threshold, and the downloading and clearing unit is triggered when the set threshold is reached;
  • a downloading and clearing unit configured to download the file block to the service node where the virtual file block is located according to at least one access address information of the file block, and clear or clear the attribute list of the virtual file block , or delete the virtual identification letter in the attribute list of the virtual file block Interest.
  • the access heat value in the attribute list of the virtual file block is updated by the update unit by:
  • the access heat value in the attribute list of the virtual file block in the service node is incremented by one.
  • an attribute list is set for a virtual file block in a service node, and the attribute list includes virtual identification information, access popularity value information, and at least one access address information of the file block, where the user belongs to a certain service node.
  • the file block is hot on the service node, and the file block is downloaded to the service node according to at least one access address information of the file block in the virtual file block attribute list, and the file block is removed. Virtual identification information.
  • the service node can directly provide the service service to the user locally. Since the remote service node does not need to cache the requested service content, the service quality of the service is improved. Since only the service content whose access heat value exceeds the threshold value is downloaded to the service node with more user requests, and does not occupy the storage space of other service nodes, it is not necessary to equip each service node in the distributed file system. Large storage space, low implementation cost.
  • the invention is simple and practical to implement. BRIEF abstract
  • FIG. 1 is a flow chart of a method for processing a service content of a service node in a distributed file system according to the present invention
  • FIG. 2 is a flow chart of a method for updating a service content of a service node in a distributed file system according to the present invention
  • FIG. 3 is a schematic diagram showing the structure of a service content processing apparatus of a service node in a distributed file system according to the present invention
  • FIG. 4 is a schematic structural diagram of a service content updating apparatus of a service node in a distributed file system according to the present invention. Preferred embodiment of the invention
  • the basic idea of the present invention is: setting a property list for a virtual file block in a service node, wherein the attribute list includes virtual identification information, access popularity value information, and at least one access address information of the file block, where the user is on a service node.
  • the access heat value in the virtual file block attribute list is updated, even if the access heat value is increased by one, and compared with the set threshold, when the set threshold is reached, the virtual file is considered as the virtual file. If the file block corresponding to the block has a high degree of heat on the service node, downloading the file block to the service node according to at least one access address information of the file block in the virtual file block attribute list, and removing the file block The virtual identification information of the file block.
  • the service node can directly provide the service service to the user locally. Since the remote service node does not need to cache the requested service content, the service quality of the service is improved. Since only the service content whose access heat value exceeds the threshold value is downloaded to the service node with more user requests, and the storage space of other service nodes is not occupied, it is not necessary to equip each service node in the distributed file system. Large storage space, low implementation cost.
  • the invention is simple and practical to implement.
  • FIG. 1 is a flow chart of a service content processing method of a service node in a distributed file system according to the present invention. As shown in FIG. 1, the service content processing method of a service node in the distributed file system of the present invention includes the following steps:
  • Step 101 After receiving the uploaded business content, the distributed file system generates a file block according to the business content.
  • the distributed file system provides maintenance personnel with an upload interface for service content.
  • the maintenance personnel can upload various service contents through the upload interface, and the service content includes various audio and video files, such as movies or songs.
  • the implementation of the upload interface is easy.
  • the authenticated user can upload the business content, and the upload interface is maintained by the operator.
  • the distributed file system After receiving the service content uploaded by the maintenance personnel, the distributed file system generates a file block according to the user's needs and the operating characteristics of the distributed file system. Since the distributed file system of the present invention mainly faces the on-demand service, the user usually provides a classic movie segment or a song and a segment thereof for on-demand broadcasting by various users.
  • the business content needs to be segmented according to the user's needs or operational requirements, split into more than one file block, and then the divided file blocks are distributed to each service node in the distributed file system. On, for users to order.
  • Business content segmentation can be achieved using existing segmentation software. If the business content uploaded by the maintenance personnel is itself a file block that meets the requirements or the file content has been uploaded before the business content is uploaded, it will not be segmented.
  • Step 102 Store the file block divided in step 101 in at least one service node, store the virtual file block corresponding to the file block in the service node that does not store the file block, and set the virtual file A list of attributes of the file block, the attribute list including virtual identification information, access popularity value information, and at least one access address information of the file block.
  • step 102 the file block is released, that is, all the file blocks in the distributed file system are posted on each service node for the user to order.
  • all the file blocks cannot be stored on each service node, which will result in a relatively large storage space of each service node in the distributed file system, which will make the cost of the distributed file system higher. Therefore, the present invention All the file blocks in the distributed file system are distributed to each service node in the distributed file system, and all the service nodes share the storage task of the file block.
  • the directory of the file block is provided, it is distributed. All file block directories in the file system, that is, virtual file blocks exist on the service node.
  • the meaning of the virtual file block is that the file block is not stored on the service node displaying the virtual file block directory, but is stored on other service nodes.
  • the file block folder may be set on the corresponding service node, and the file block is stored in the corresponding folder, such as the file type, size, and creation time.
  • the user-oriented information of the virtual file block is exactly the same as the actual file block. The user does not know which file blocks are not stored on the service node that receives the on-demand request, and the information is shielded from the user. For each file block in a distributed file system, it is stored on at least one service node.
  • a property list is set for the virtual file, and the attribute list includes at least virtual identification information, access popularity value information, and at least one access address information of the file block, where the virtual identifier is used to identify the file block as The virtual file block, the access heat value is used to quantify the number of times the user clicks on the file block within a set time period, for example, the number of times a file block is in the service node in the service node within one hour, and the on-demand time in the set time period. The more times, the access heat of the file block The higher the value.
  • the storage address of the actual file block corresponding to the virtual file block is that after the service node receives the user's request for on-demand, according to the storage address of the actual file block recorded in the virtual file block attribute list, the actual file block is correspondingly stored.
  • the file node reads the file block, caches it on the service node where the virtual file block is called, and sends the cached file block data to the requested user, and plays the data of the file block to the user. In this way, when the user requests the on-demand service from the service node, even if the virtual file block is requested, the required service data can be obtained, and the corresponding service node will update the access heat value information in the virtual file block attribute list. Even if its access heat value is increased by one.
  • the service content update method of a service node in the distributed file system of the present invention includes the following steps:
  • Step 201 After receiving the on-demand request of the file block, the service node determines whether the file block to be requested is a virtual file block, and is based on the file corresponding to the virtual file block in the attribute list of the virtual file block. At least one access address information of the block reads the file block and responds to the on-demand request, and simultaneously updates an access heat value in the attribute list of the virtual file block in the service node.
  • the service node After receiving the on-demand request of the file block, the service node determines whether the file block to be requested is a virtual file block, which can be determined by searching whether the file block to be requested is set with an attribute list, if the virtual file block is in the distributed file system.
  • setting the property list is a virtual file block. If the virtual file block is converted into the actual file block, the service node only clears the attribute list or only deletes the corresponding information in the attribute list, such as the virtual identifier information, and determines whether the virtual identifier information is included in the attribute list of the virtual file block. Whether it is a virtual file block, including virtual identification information, is a virtual file block.
  • the file block is read according to at least one access address information of the file block corresponding to the virtual file block in the attribute list of the virtual file block, and the read file block is read.
  • the data of the file block is cached on the service node, and the cached data of the file block is used to respond to the user's request for on-demand, and the file block that is requested by the user is played, and the service node
  • the count value of the access heat value in the attribute list of the file block is incremented by one.
  • the access heat value is a dynamic value, and the access heat value is automatically cleared and re-counted when each set count cycle arrives.
  • the service node sends the count value of the access heat value to one.
  • Step 202 Determine whether the updated access popularity value reaches a set threshold, and when yes, download the file block to the service node according to at least one access address information of the file block, and clear or empty the virtual file. A list of attributes of the block, or delete the virtual identification information in the attribute list of the virtual file block.
  • the service node determines whether the updated access heat value reaches a set threshold, and if the set threshold is reached, the virtual file block is in the If the on-demand frequency of the service node is still relatively high, the actual file block corresponding to the virtual file block is downloaded to the service node according to at least one access address information in the attribute list of the virtual file block, so as to be When the user clicks on the data again, the local file block data is directly responded to the user to be clicked, so as to improve the response speed of the service that the user subscribes to, and improve the service quality of the service. After the file block is downloaded, the virtual node is virtualized.
  • the attribute list of the file block is emptied or deleted directly, or only the virtual identification information in the attribute list of the virtual file block is deleted to declare that the file block is no longer a virtual file block.
  • the service node sets the virtual file block as a non-virtual file block, and is related to a manner of defining a virtual file block and a non-virtual file block in the distributed file system, and determining whether the file block is a virtual file block, and This is related.
  • the threshold value is an empirical value, which is set according to the actual system.
  • the service content update method of the service node in the distributed file system shown in FIG. 2 can be combined with the service content processing of the service node in the distributed file system shown in FIG. 1 in addition to the technical solution itself.
  • the method constitutes a technical solution.
  • steps 201 and 202 shown in FIG. 2 are included after steps 101 and 102 shown in FIG.
  • a service content processing apparatus of a service node in a distributed file system of the present invention includes an upload unit 30 and a file block generation.
  • the file block generating unit 31 is configured to generate a file block by using the business content uploaded by the uploading unit 30; dividing the uploaded business content into one or more file blocks according to business requirements or user requirements, etc.
  • the publishing unit 32 is configured to store the file block in at least one of the service nodes, and the service node that does not store the file block stores a virtual file block corresponding to the file block;
  • the file block generated by the block generating unit 31 is distributed to each service node in the distributed file system, and the directory of the file block on each service node is the same, where the file block is not stored on the service node of the actual file block. Is set as a virtual file block.
  • the published virtual file block on the service node has the same name as the actual file block, various attribute values such as file type, size, and creation time.
  • the setting unit 33 is configured to set an attribute list for the virtual file block, where the attribute list includes virtual identification information, access popularity value information, and at least one access address information of the file block.
  • the service content processing apparatus of the service node in the distributed file system shown in FIG. 3 is designed to implement the service content processing method of the service node in the distributed file system shown in FIG.
  • the implementation functions of the processing units in the apparatus shown in FIG. 3 can be understood by referring to the related description in the method shown in FIG. 1.
  • the functions of each unit can be implemented by running a program on the processor, or by corresponding logic.
  • the circuit is implemented.
  • the service content update apparatus of a service node in the distributed file system of the present invention includes a receiving unit 40, and a first judgment.
  • the first determining unit 41 is configured to determine whether the file block requested by the on-demand request is a virtual file block, and triggers the reading and responding unit 42; specifically, it may be determined according to whether the requested file block includes an attribute list, or It is determined according to whether the attribute list of the file block includes virtual identification information, and is a virtual file block.
  • the reading and responding unit 42 is configured to read the file block according to at least one access address information of the file block corresponding to the virtual file block in the attribute list of the virtual file block, and respond to the request for on-demand,
  • the update unit 43 is triggered at the same time.
  • the updating unit 43 is configured to update the access heat value in the attribute list of the virtual file block, and trigger the second determining unit 44, where the access heat value in the attribute list of the virtual file block is updated, and the count of the access heat value is added.
  • the second determining unit 44 is configured to determine whether the access popularity value in the attribute list of the updated virtual file block reaches a set width The value, when it is reached, triggers the download and clear unit 45; the value of the access heat value is the empirical value, and the different system experience values are different.
  • the downloading and clearing unit 45 is configured to download the file block to the service node where the virtual file block is located according to at least one access address information of the file block, and clear or clear the attribute list of the virtual file block. Or delete the virtual identification information in the attribute list of the virtual file block.
  • the service content update apparatus of the service node in the distributed file system shown in FIG. 4 is designed to implement the service content update method of the service node in the distributed file system shown in FIG. 2,
  • the implementation functions of the processing units in the apparatus shown in FIG. 4 can be understood by referring to the related description in the method shown in FIG. 2.
  • the functions of each unit can be implemented by running a program on the processor, or by corresponding logic.
  • the circuit is implemented.
  • the service content updating apparatus of the service node in the distributed file system shown in FIG. 4 may further include a service content processing apparatus of the service node in the distributed file system shown in FIG. 3.
  • an attribute list is set for a virtual file block in a service node, and the attribute list includes virtual identification information, access popularity value information, and at least one access address information of the file block, where the user belongs to a certain service node.
  • the access heat value in the virtual file block attribute list is updated, even if the access heat value is increased by one, and compared with the set threshold, when the set threshold is reached, the virtual file block is considered to be corresponding.
  • the file block is hot on the service node, and the file block is downloaded to the service node according to at least one access address information of the file block in the virtual file block attribute list, and the file block is removed.
  • the service node can directly provide the service service to the user locally. Since the remote service node does not need to cache the requested service content, the service quality of the service is improved. Since only the service content whose access heat value exceeds the threshold value is downloaded to the service node with more user requests, and does not occupy the storage space of other service nodes, it is not necessary to equip each service node in the distributed file system. Large storage space, low implementation cost.
  • the invention is simple and practical to implement.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Information Transfer Between Computers (AREA)

Description

分布式文件系统中业务内容处理、 更新方法与装置
技术领域
本发明涉及分布式文件系统中业务内容处理、 更新技术， 尤其涉及一种 分布式文件系统中业务节点中的业务内容处理、 更新方法与装置。
背景技术
分布式文件系统广泛应用于网络电视（ IPTV, Internet Protocol Television ) 等视频业务的服务器中， 用于管理单个业务节点 （为一个区域或一部分用户 提供服务） 内业务数据的存储和读取。 随着用户对业务多样性需求的不断提 高， 业务应用中需存储的业务数据量也越来越多， 分布式文件系统中各个业 务节点要提供完整的业务服务， 就需要海量的存储空间， 这将造成整个分布 式文件系统的成本大幅提高， 给运营商带来很大压力。
为了解决上述问题， 分布式文件系统普遍釆用业务内容调度的方法来解 决， 即所有的业务内容分散存储到各个业务节点， 单个业务节点上并不存储 所有的业务内容， 但每个业务节点上所提供的业务内容目录却是相同的， 当 用户在某个节点上点播一个并未存储在本业务节点的业务内容时， 该业务节 点查找存储所点播业务内容的其它节点， 将用户所点播的业务内容緩存到本 业务节点， 再提供给请求用户。
使用上述业务内容的调度方法， 可大量节省分布式文件系统中各个业务 节点的存储空间， 从而降低了分布式文件系统的组建成本。 但仍存在不足之 处， 业务内容的调度方法， 在业务节点上没有相应的业务内容， 而用户对该 业务节点上的该业务内容的点播请求又较多时， 虽然能从其他业务节点向用 户提供业务服务， 但这需耗费分布式文件系统较大的带宽， 在点播用户数较 多时， 所提供的业务服务的质量很难得到保证。
发明内容
有鉴于此， 本发明的主要目的在于提供一种分布式文件系统中业务节点 的业务内容处理、 更新方法与装置， 能提升了热点业务内容的服务质量， 且 使所占用的存储空间较小。
为达到上述目的， 本发明的技术方案是这样实现的：
一种分布式文件系统中业务节点的业务内容处理方法， 包括：
接收到上载的业务内容后， 分布式文件系统根据所述业务内容生成文件 块， 所述文件块存储于至少一个所述业务节点中， 未存储所述文件块的所述 业务节点中存储有与所述文件块对应的虚拟文件块， 并设置所述虚拟文件块 的属性列表， 所述属性列表包括虚拟标识信息、 访问热度值信息以及所述文 件块的至少一个访问地址信息。
优选地， 所述虚拟文件块与所述文件块的名称、 属性值相同。
优选地，分布式文件系统根据所述业务内容生成文件块的所述步骤包括： 所述分布式文件系统将所述业务内容直接作为文件块， 或将所述业务内 容分割成两个以上的业务内容块而作为文件块。
一种分布式文件系统中业务节点的业务内容更新方法， 包括：
接收到文件块的点播请求后， 业务节点判断所点播的文件块是否为虚拟 文件块， 是则根据所述虚拟文件块的属性列表中的与所述虚拟文件块对应的 所述文件块的至少一个访问地址信息读取所述文件块并响应所述点播请求， 同时， 更新所述业务节点中所述虚拟文件块的属性列表中的访问热度值， 并 判断更新后的访问热度值是否达到设定阔值， 达到所述设定阔值则根据所述 文件块的至少一个访问地址信息将所述文件块下载到所述业务节点， 并清除 或清空所述虚拟文件块的属性列表， 或删除所述虚拟文件块的属性列表中的 虚拟标识信息。
优选地， 业务节点判断所点播的文件块是否为虚拟文件块的所述步骤包 括：
根据所点播的文件块的标识查找所点播的文件块是否包括属性列表或判 断所查找到的属性列表是否为非空， 或判断所查找到的属性列表中是否包括 虚拟标识信息， 是则判断所点播的文件块为虚拟文件块。
优选地， 更新所述业务节点中所述虚拟文件块的属性列表中的访问热度 值的所述步骤包括：
使所述业务节点中所述虚拟文件块的属性列表中的访问热度值加一。 一种分布式文件系统中业务节点的业务内容处理装置， 包括：
上载单元， 其设置成向分布式文件系统上载业务内容；
文件块生成单元， 其设置成将通过所述上载单元上载的所述业务内容生 成文件块；
发布单元， 其设置成将所述文件块存储于至少一个所述业务节点中， 未 存储所述文件块的所述业务节点中存储有与所述文件块对应的虚拟文件块； 以及
设置单元， 其设置成为所述虚拟文件块设置属性列表， 所述属性列表包 括虚拟标识信息、访问热度值信息以及所述文件块的至少一个访问地址信息。
优选地， 所述虚拟文件块与所述文件块的名称、 属性值相同。
一种分布式文件系统中业务节点的业务内容更新装置， 包括：
接收单元， 其设置成接收文件块的点播请求；
第一判断单元， 其设置成判断所述点播请求所点播的文件块是否为虚拟 文件块， 是时触发读取及响应单元；
读取及响应单元， 其设置成根据所述虚拟文件块的属性列表中的与所述 虚拟文件块对应的所述文件块的至少一个访问地址信息读取所述文件块并响 应所述点播请求， 同时触发更新单元；
更新单元， 其设置成更新所述虚拟文件块的属性列表中的访问热度值， 并触发第二判断单元；
第二判断单元， 其设置成判断更新后的所述虚拟文件块的属性列表中的 访问热度值是否达到设定阔值， 达到所述设定阔值则触发下载及清除单元； 以及
下载及清除单元， 其设置成根据所述文件块的至少一个访问地址信息将 所述文件块下载到所述虚拟文件块所在的所述业务节点， 并清除或清空所述 虚拟文件块的属性列表， 或删除所述虚拟文件块的属性列表中的虚拟标识信 息。
优选地， 所述虚拟文件块的属性列表中的访问热度值由所述更新单元通 过以下方式更新：
使所述业务节点中所述虚拟文件块的属性列表中的访问热度值加一。 本发明中， 为业务节点中的虚拟文件块设置属性列表， 属性列表中记载 有虚拟标识信息、访问热度值信息以及所述文件块的至少一个访问地址信息， 在用户对某业务节点中的某虚拟文件块点播后， 更新虚拟文件块属性列表中 的访问热度值， 即使其访问热度值加一， 并与设定阔值进行比较， 达到设定 阔值时， 即认为该虚拟文件块对应的文件块在该业务节点上的热度较高， 则 根据虚拟文件块属性列表中的所述文件块的至少一个访问地址信息， 将所述 文件块下载到该业务节点上， 并去除所述文件块的虚拟标识信息。 这样， 在 其他用户再向该业务节点点播所述文件块时， 该业务节点能直接在本地向用 户提供业务服务， 由于不必再到远程业务节点緩存所点播业务内容， 提升了 业务的服务质量， 由于仅是将访问热度值超出阔值的业务内容下载到用户请 求较多的业务节点上， 并不会占用其他业务节点的存储空间， 不必为分布式 文件系统中的每个业务节点均配备很大的存储空间， 实现成本较低。 本发明 实现简单且实用。 附图概述
图 1 为本发明分布式文件系统中业务节点的业务内容处理方法的流程 图；
图 2 为本发明分布式文件系统中业务节点的业务内容更新方法的流程 图；
图 3为本发明分布式文件系统中业务节点的业务内容处理装置的组成结 构示意图； 以及
图 4为本发明分布式文件系统中业务节点的业务内容更新装置的组成结 构示意图。 本发明的较佳实施方式
本发明的基本思想是： 为业务节点中的虚拟文件块设置属性列表， 属性 列表中记载有虚拟标识信息、 访问热度值信息以及所述文件块的至少一个访 问地址信息， 在用户对某业务节点中的某虚拟文件块点播后， 更新虚拟文件 块属性列表中的访问热度值， 即使其访问热度值加一， 并与设定阔值进行比 较， 达到设定阔值时， 即认为该虚拟文件块对应的文件块在该业务节点上的 热度较高， 则根据虚拟文件块属性列表中的所述文件块的至少一个访问地址 信息， 将所述文件块下载到该业务节点上， 并去除所述文件块的虚拟标识信 息。 这样， 在其他用户再向该业务节点点播所述文件块时， 该业务节点能直 接在本地向用户提供业务服务， 由于不必再到远程业务节点緩存所点播业务 内容， 提升了业务的服务质量， 由于仅是将访问热度值超出阔值的业务内容 下载到用户请求较多的业务节点上， 并不会占用其他业务节点的存储空间， 不必为分布式文件系统中的每个业务节点均配备^大的存储空间， 实现成本 较低。 本发明实现简单且实用。
为使本发明的目的、 技术方案和优点更加清楚明白， 以下举实施例并参 照附图， 对本发明进一步详细说明。
图 1 为本发明分布式文件系统中业务节点的业务内容处理方法的流程 图， 如图 1所示， 本发明分布式文件系统中业务节点的业务内容处理方法包 括以下步骤：
步骤 101 : 接收到上载的业务内容后， 分布式文件系统根据所述业务内 容生成文件块。
分布式文件系统为维护人员提供业务内容的上载接口， 维护人员通过上 载接口能上载各种业务内容， 所述业务内容包括各种音、 视频文件， 如电影 或歌曲等文件。 上载接口的实现是容易的， 通过向维护人员提供业务内容的 上载界面， 设置相应的身份验证码， 通过身份验证的用户即可实现业务内容 的上载， 上载接口由运营商维护。 接收到维护人员上载的业务内容后， 分布 式文件系统根据用户的需求及分布式文件系统的运营特点， 将上载的业务内 容生成文件块。 由于本发明分布式文件系统所面对的主要是点播业务， 一般 为用户提供经典的电影片段或歌曲及其片段， 以供各种用户进行点播， 因此， 对于所上载的完整的电影或歌曲， 需按用户的需求或运营需求对业务内容进 行分割， 分割成一个以上的文件块， 再将所分割的文件块发布到分布式文件 系统中的各业务节点上， 供用户点播。 业务内容分割可使用现有的分割软件 来实现。 如果维护人员所上载的业务内容本身即是符合要求的文件块或上载 业务内容之前已进行过文件分割， 则不再对其进行分割处理。
步骤 102: 将步骤 101 中所分割的文件块存储于至少一个业务节点中， 未存储所述文件块的所述业务节点中存储有与所述文件块对应的虚拟文件 块， 并设置所述虚拟文件块的属性列表， 所述属性列表包括虚拟标识信息、 访问热度值信息以及所述文件块的至少一个访问地址信息。
步骤 102中即实现文件块的发布， 即将分布式文件系统中所有的文件块 发布于各业务节点上， 以供用户点播。 但是， 所有的文件块并不能全部存储 于各业务节点上， 这样将会导致分布式文件系统中各业务节点的存储空间要 相当大， 将使分布式文件系统的成本较高， 因此， 本发明是将分布式文件系 统中所有的文件块分散存储到分布式文件系统中各业务节点中， 以所有的业 务节点分担文件块的存储任务， 但是， 在提供文件块的目录时， 却是分布式 文件系统中所有的文件块目录， 即业务节点上存在虚拟文件块。 虚拟文件块 的含义是，所述文件块并不没有存储在显示该虚拟文件块目录的业务节点上， 而是存储在其他业务节点上。 在将文件块存储于业务节点上时， 可以在对应 的业务节点上设置所述文件块的文件夹， 将所述文件块存储于对应的文件夹 如文件类型、 大小及创建时间等均相同， 虚拟文件块面向用户的信息与实际 的文件块是完全相同的， 用户并不知道哪些文件块未存储在接收了点播请求 的业务节点上， 这些信息是对用户屏蔽的。 对于分布式文件系统中的每一个 文件块， 至少存储于一个业务节点上。
对于业务节点上的虚拟文件块， 为其设置属性列表， 所述属性列表中至 少包括虚拟标识信息、 访问热度值信息以及所述文件块的至少一个访问地址 信息， 虚拟标识用于标识文件块为虚拟文件块， 访问热度值用于量化在设定 的时间段内用户点播所述文件块的次数， 例如 1小时内某文件块在其所在业 务节点中的点播次数， 设定时间段内的点播次数越多， 所述文件块的访问热 度值就越高。 而虚拟文件块所对应实际文件块的存储地址则是在业务节点接 收到用户的点播请求后， 根据该虚拟文件块属性列表中记载的实际文件块的 存储地址， 到对应存储所述实际文件块的业务节点上读取该文件块， 緩存到 所点播虚拟文件块所在的业务节点上， 再将所緩存的文件块数据发送给所请 求用户， 向用户播放该文件块的数据。 这样， 用户在向业务节点请求点播业 务时， 即使所请求的是虚拟文件块， 仍能得到所需的业务数据， 并且， 相应 的业务节点将会更新虚拟文件块属性列表中的访问热度值信息， 即使其访问 热度值加一。
图 2 为本发明分布式文件系统中业务节点的业务内容更新方法的流程 图， 如图 2所示， 本发明分布式文件系统中业务节点的业务内容更新方法包 括以下步骤：
步骤 201 : 接收到文件块的点播请求后， 业务节点判断所点播的文件块 是否为虚拟文件块， 是时根据所述虚拟文件块的属性列表中的与所述虚拟文 件块对应的所述文件块的至少一个访问地址信息读取所述文件块并响应所述 点播请求， 同时， 更新所述业务节点中所述虚拟文件块的属性列表中的访问 热度值。
业务节点接收到文件块的点播请求后，判断所点播的文件块是否为虚拟 文件块， 具体可通过查找所点播的文件块是否设置有属性列表来确定， 若分 布式文件系统中对于虚拟文件块设置属性列表而非虚拟文件块不设置属性列 表时， 设置属性列表的即为虚拟文件块。 而如果虚拟文件块转换为实际文件 块时， 业务节点仅清空属性列表或仅将属性列表中的对应信息如虚拟标识信 息删除时， 通过判断虚拟文件块的属性列表中是否包括虚拟标识信息来确定 是否为虚拟文件块， 包括虚拟标识信息的， 即为虚拟文件块。 如果所点播文 件块是虚拟文件块， 则根据该虚拟文件块的属性列表中与所述虚拟文件块对 应的所述文件块的至少一个访问地址信息读取所述文件块， 将所读取的所述 文件块的数据緩存于所述业务节点上， 并用所读取的所述文件块的緩存数据 响应用户的点播请求， 向用户播放其所点播的文件块， 所述业务节点将所述 虚拟文件块的属性列表中的访问热度值的计数值加一。 访问热度值是一个动 态值， 在每个设定计数周期到来时， 该访问热度值会被自动清零并重新计数， 在用户请求点播所述虚拟文件块后， 所述业务节点将其访问热度值的计数值 力口一。
步骤 202: 判断更新后的访问热度值是否达到设定阔值， 达到时根据所 述文件块的至少一个访问地址信息将所述文件块下载到所述业务节点， 并清 除或清空所述虚拟文件块的属性列表， 或删除所述虚拟文件块的属性列表中 的虚拟标识信息。
在所述虚拟文件块的访问热度值更新后， 所述业务节点判断该更新后的 访问热度值是否达到了设定阔值， 如果达到了设定阔值， 说明所述虚拟文件 块在所述业务节点上的点播频率还是比较高的， 则根据所述虚拟文件块的属 性列表中的至少一个访问地址信息， 将所述虚拟文件块对应的实际文件块下 载到所述业务节点上， 以在用户再次点播时， 直接将本地的文件块数据响应 给所点播的用户， 以提高用户所点播业务的响应速度， 提升业务的服务质量， 文件块下载完毕后， 将所述业务节点上所述虚拟文件块的属性列表清空或直 接删除， 或者仅删除所述虚拟文件块的属性列表中的虚拟标识信息， 以宣示 所述文件块不再是虚拟文件块。 所述业务节点设置所述虚拟文件块为非虚拟 文件块的方式， 与分布式文件系统中定义虚拟文件块、 非虚拟文件块的方式 有关， 判断所述文件块是否为虚拟文件块， 也与此有关。 访问热度值的设定 阔值是一个经验值， 根据实际系统而设定。
本领域技术人员应当理解， 图 2所示的分布式文件系统中业务节点的业 务内容更新方法除自身构成技术方案外， 还可结合图 1所示的分布式文件系 统中业务节点的业务内容处理方法而构成技术方案， 此时， 在图 1所示的步 骤 101、 102之后包括图 2所示的步骤 201、 202。
图 3为本发明分布式文件系统中业务节点的业务内容处理装置的组成结 构示意图， 如图 3所示， 本发明分布式文件系统中业务节点的业务内容处理 装置包括上载单元 30、 文件块生成单元 31、 发布单元 32和设置单元 33 , 其 中， 上载单元 30用于向分布式文件系统上载业务内容； 上载单元 30可通过 上载接口而实现， 维护人员通过上载单元 30可上载各种业务内容。 文件块生 成单元 31用于将通过上载单元 30上载的所述业务内容生成文件块； 即将所 上载的业务内容按业务需求或用户需求等分割为一个以上的文件块， 当所上 载的业务内容符合发布要求或上载之前已分割为文件块时， 直接将所述业务 内容作为文件块。发布单元 32用于将所述文件块存储于至少一个所述业务节 点中， 未存储所述文件块的所述业务节点中存储有与所述文件块对应的虚拟 文件块； 发布单元 32将文件块生成单元 31所生成的文件块发布到分布式文 件系统中各业务节点上， 各业务节点上的文件块的目录都是相同的， 其中， 未存储实际文件块的业务节点上， 该文件块被设置成虚拟文件块。 业务节点 上的发布的虚拟文件块与实际的文件块的名称、 各种属性值如文件类型、 大 小及创建时间等均相同。 设置单元 33用于为所述虚拟文件块设置属性列表， 所述属性列表包括虚拟标识信息、 访问热度值信息以及所述文件块的至少一 个访问地址信息。
本领域技术人员应当理解， 本发明图 3所示的分布式文件系统中业务节 点的业务内容处理装置是为实现图 1所示的分布式文件系统中业务节点的业 务内容处理方法而设计的， 图 3所示装置中的各处理单元的实现功能可参照 图 1所示的方法中的相关描述而理解， 各单元的功能可通过运行于处理器上 的程序而实现， 也可通过相应的逻辑电路而实现。
图 4为本发明分布式文件系统中业务节点的业务内容更新装置的组成结 构示意图， 如图 4所示， 本发明分布式文件系统中业务节点的业务内容更新 装置包括接收单元 40、 第一判断单元 41、 读取及响应单元 42、 更新单元 43、 第二判断单元 44和下载及清除单元 45 , 其中， 接收单元 40用于接收文件块 的点播请求； 所述点播请求来自于各种用户。 第一判断单元 41用于判断所述 点播请求所点播的文件块是否为虚拟文件块， 是时触发读取及响应单元 42; 具体的， 可以根据所请求文件块是否包括属性列表来判断， 或者依据所述文 件块的属性列表是否包括虚拟标识信息来判断， 是时即为虚拟文件块。 读取 及响应单元 42 用于根据所述虚拟文件块的属性列表中的与所述虚拟文件块 对应的所述文件块的至少一个访问地址信息读取所述文件块并响应所述点播 请求， 同时触发更新单元 43。 更新单元 43用于更新所述虚拟文件块的属性 列表中的访问热度值， 并触发第二判断单元 44, 这里更新所述虚拟文件块的 属性列表中的访问热度值即将访问热度值的计数加一。第二判断单元 44用于 判断更新后的所述虚拟文件块的属性列表中的访问热度值是否达到设定阔 值， 达到时触发下载及清除单元 45; 访问热度值的阔值为经验值， 不同的系 统经验值不同。下载及清除单元 45用于根据所述文件块的至少一个访问地址 信息将所述文件块下载到所述虚拟文件块所在的所述业务节点， 并清除或清 空所述虚拟文件块的属性列表， 或删除所述虚拟文件块的属性列表中的虚拟 标识信息。
本领域技术人员应当理解， 本发明图 4所示的分布式文件系统中业务节 点的业务内容更新装置是为实现图 2所示的分布式文件系统中业务节点的业 务内容更新方法而设计的， 图 4所示装置中的各处理单元的实现功能可参照 图 2所示的方法中的相关描述而理解， 各单元的功能可通过运行于处理器上 的程序而实现， 也可通过相应的逻辑电路而实现。
图 4所示的分布式文件系统中业务节点的业务内容更新装置中还可包括 图 3所示的分布式文件系统中业务节点的业务内容处理装置。
以上所述， 仅为本发明的较佳实施例而已， 并非用于限定本发明的保护 范围。 工业实用性
本发明中， 为业务节点中的虚拟文件块设置属性列表， 属性列表中记载 有虚拟标识信息、访问热度值信息以及所述文件块的至少一个访问地址信息， 在用户对某业务节点中的某虚拟文件块点播后， 更新虚拟文件块属性列表中 的访问热度值， 即使其访问热度值加一， 并与设定阔值进行比较， 达到设定 阔值时， 即认为该虚拟文件块对应的文件块在该业务节点上的热度较高， 则 根据虚拟文件块属性列表中的所述文件块的至少一个访问地址信息， 将所述 文件块下载到该业务节点上， 并去除所述文件块的虚拟标识信息。 这样， 在 其他用户再向该业务节点点播所述文件块时， 该业务节点能直接在本地向用 户提供业务服务， 由于不必再到远程业务节点緩存所点播业务内容， 提升了 业务的服务质量， 由于仅是将访问热度值超出阔值的业务内容下载到用户请 求较多的业务节点上， 并不会占用其他业务节点的存储空间， 不必为分布式 文件系统中的每个业务节点均配备很大的存储空间， 实现成本较低。 本发明 实现简单且实用。

Claims

权 利 要 求 书
1、 一种分布式文件系统中业务节点的业务内容处理方法， 包括： 接收到上载的业务内容后， 分布式文件系统根据所述业务内容生成文件 块， 所述文件块存储于至少一个所述业务节点中， 未存储所述文件块的所述 业务节点中存储有与所述文件块对应的虚拟文件块， 并设置所述虚拟文件块 的属性列表， 所述属性列表包括虚拟标识信息、 访问热度值信息以及所述文 件块的至少一个访问地址信息。
2、根据权利要求 1所述的方法， 其中所述虚拟文件块与所述文件块的名 称、 属性值相同。
3、根据权利要求 1所述的方法， 其中分布式文件系统根据所述业务内容 生成文件块的所述步骤包括：
所述分布式文件系统将所述业务内容直接作为文件块， 或将所述业务内 容分割成两个以上的业务内容块而作为文件块。
4、 一种分布式文件系统中业务节点的业务内容更新方法， 包括： 接收到文件块的点播请求后， 业务节点判断所点播的文件块是否为虚拟 文件块， 是则根据所述虚拟文件块的属性列表中的与所述虚拟文件块对应的 所述文件块的至少一个访问地址信息读取所述文件块并响应所述点播请求， 同时更新所述业务节点中所述虚拟文件块的属性列表中的访问热度值， 并判 断更新后的访问热度值是否达到设定阔值， 达到所述设定阔值则根据所述文 件块的所述至少一个访问地址信息将所述文件块下载到所述业务节点， 并清 除或清空所述虚拟文件块的属性列表， 或删除所述虚拟文件块的属性列表中 的虚拟标识信息。
5、根据权利要求 4所述的方法， 其中业务节点判断所点播的文件块是否 为虚拟文件块的所述步骤包括：
根据所点播的文件块的标识查找所点播的文件块是否包括属性列表或判 断所查找到的属性列表是否为非空， 或判断所查找到的属性列表中是否包括 虚拟标识信息， 是则判断所点播的文件块为虚拟文件块。
6、根据权利要求 4所述的方法， 其中更新所述业务节点中所述虚拟文件 块的属性列表中的访问热度值的所述步骤包括：
使所述业务节点中所述虚拟文件块的属性列表中的访问热度值加一。
7、 一种分布式文件系统中业务节点的业务内容处理装置， 包括： 上载单元， 其设置成向分布式文件系统上载业务内容；
文件块生成单元， 其设置成将通过所述上载单元上载的所述业务内容生 成文件块；
发布单元， 其设置成将所述文件块存储于至少一个所述业务节点中， 在 未存储所述文件块的所述业务节点中存储与所述文件块对应的虚拟文件块； 以及
设置单元， 其设置成为所述虚拟文件块设置属性列表， 所述属性列表包 括虚拟标识信息、访问热度值信息以及所述文件块的至少一个访问地址信息。
8、根据权利要求 7所述的装置， 其中所述虚拟文件块与所述文件块的名 称、 属性值相同。
9、 一种分布式文件系统中业务节点的业务内容更新装置， 包括： 接收单元， 其设置成接收文件块的点播请求；
第一判断单元， 其设置成判断所述点播请求所点播的文件块是否为虚拟 文件块， 是则触发读取及响应单元；
读取及响应单元， 其设置成根据所述虚拟文件块的属性列表中的与所述 虚拟文件块对应的所述文件块的至少一个访问地址信息读取所述文件块并响 应所述点播请求， 同时触发更新单元；
更新单元， 其设置成更新所述虚拟文件块的属性列表中的访问热度值， 并触发第二判断单元；
第二判断单元， 其设置成判断更新后的所述虚拟文件块的属性列表中的 访问热度值是否达到设定阔值， 达到所述设定阔值则触发下载及清除单元； 以及
下载及清除单元， 其设置成根据所述文件块的所述至少一个访问地址信 息将所述文件块下载到所述虚拟文件块所在的所述业务节点， 并清除或清空 所述虚拟文件块的属性列表， 或删除所述虚拟文件块的属性列表中的虚拟标 识信息。
10、 根据权利要求 9所述的装置， 其中所述虚拟文件块的属性列表中的 使所述业务节点中所述虚拟文件块的属性列表中的访问热度值加一。
PCT/CN2009/073330 2009-04-09 2009-08-18 分布式文件系统中业务内容处理、更新方法与装置 WO2010115325A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US13/257,209 US20120016916A1 (en) 2009-04-09 2009-08-18 Method and Apparatus for Processing and Updating Service Contents in a Distributed File System
EP09842886.5A EP2418809A4 (en) 2009-04-09 2009-08-18 METHOD AND DEVICE FOR PROCESSING AND UPDATING SERVICE CONTENT IN A DISTRIBUTED FILE SYSTEM

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200910081758.1 2009-04-09
CN200910081758A CN101527736A (zh) 2009-04-09 2009-04-09 分布式文件系统中业务内容处理、更新方法与装置

Publications (1)

Publication Number Publication Date
WO2010115325A1 true WO2010115325A1 (zh) 2010-10-14

Family

ID=41095429

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/073330 WO2010115325A1 (zh) 2009-04-09 2009-08-18 分布式文件系统中业务内容处理、更新方法与装置

Country Status (4)

Country Link
US (1) US20120016916A1 (zh)
EP (1) EP2418809A4 (zh)
CN (1) CN101527736A (zh)
WO (1) WO2010115325A1 (zh)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101883012B (zh) * 2010-07-09 2012-04-18 四川长虹电器股份有限公司 一种对网络边缘节点中存储资源的处理方法
CN102426552B (zh) * 2011-10-31 2014-06-04 华为数字技术（成都）有限公司 存储系统服务质量控制方法、装置和系统
TWI507891B (zh) * 2012-03-23 2015-11-11 Egis Technology Inc 具雲端儲存空間管理功能之電子裝置、雲端儲存系統、其方法及其電腦程式產品
JP5962269B2 (ja) * 2012-07-09 2016-08-03 富士通株式会社 評価装置，分散格納システム，評価方法及び評価プログラム
JP2014146246A (ja) * 2013-01-30 2014-08-14 Fujitsu Ltd アクセス数管理プログラム、アクセス数管理装置及びアクセス数管理方法
CN104023050B (zh) * 2014-05-16 2017-06-27 中国科学院计算技术研究所 一种设备云环境下设备协同多屏互动系统及其方法
CN104780228A (zh) * 2015-05-09 2015-07-15 北京银信长远科技股份有限公司 提升网络内容服务器效率的方法
CN106257461B (zh) * 2016-08-29 2023-09-08 董晓明 数据存储方法及装置
CN106385588A (zh) * 2016-09-23 2017-02-08 北京赢点科技有限公司 一种基于点播热度的资源调度方法及装置
CN108388613B (zh) * 2018-02-08 2020-09-11 竞技世界（北京）网络技术有限公司 一种缓存数据的更新方法
CN109325010B (zh) * 2018-09-26 2023-08-04 中国平安人寿保险股份有限公司 日志查看方法、装置、计算机设备和存储介质
CN110351290B (zh) * 2019-07-17 2021-08-06 中国联合网络通信集团有限公司 数据缓存方法及系统

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1379339A (zh) * 2001-03-26 2002-11-13 微软公司 无服务器的分布式文件系统
CN1972311A (zh) * 2006-12-08 2007-05-30 华中科技大学 一种基于集群均衡负载的流媒体服务器系统

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1692356B (zh) * 2002-11-14 2014-06-04 易斯龙系统公司 用于对现存文件重新条带化的方法
US20080263610A1 (en) * 2007-04-19 2008-10-23 Youbiquity, Llc System for distributing electronic content assets over communication media having differing characteristics
US8738621B2 (en) * 2009-01-27 2014-05-27 EchoStar Technologies, L.L.C. Systems and methods for managing files on a storage device

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1379339A (zh) * 2001-03-26 2002-11-13 微软公司 无服务器的分布式文件系统
CN1972311A (zh) * 2006-12-08 2007-05-30 华中科技大学 一种基于集群均衡负载的流媒体服务器系统

Also Published As

Publication number Publication date
CN101527736A (zh) 2009-09-09
EP2418809A4 (en) 2014-04-23
EP2418809A1 (en) 2012-02-15
US20120016916A1 (en) 2012-01-19

Similar Documents

Publication Publication Date Title
WO2010115325A1 (zh) 分布式文件系统中业务内容处理、更新方法与装置
EP3022708B1 (en) Content source discovery
JP5592489B2 (ja) キャッシング情報のシステム及び方法
EP2266043B1 (en) Cache optimzation
CN104243425B (zh) 一种在内容分发网络中进行内容管理的方法、装置及系统
CN108429777B (zh) 一种基于缓存的数据更新方法及服务器
US20130268499A1 (en) Information management method, and computer for providing information
CN107197359B (zh) 视频文件缓存方法及装置
WO2011140842A1 (zh) 多媒体文件发布方法和装置
CN104768079B (zh) 多媒体资源分发方法、装置及系统
WO2017101576A1 (zh) 一种数据资源存储方法及装置
CN102307234A (zh) 基于移动终端的资源检索方法
CN108540510B (zh) 一种云主机创建方法、装置及云服务系统
JP7392168B2 (ja) Ｃｄｎにおけるｕｒｌリフレッシュ方法、装置、機器及びｃｄｎノード
CN108566370A (zh) 一种数据回源的方法以及装置
CN109740073A (zh) 一种基于url的图片实时处理系统及方法
CN110908965A (zh) 一种对象存储管理方法、装置、设备及存储介质
CN110895515A (zh) 内存缓存管理方法、多媒体服务器及计算机存储介质
US7249219B1 (en) Method and apparatus to improve buffer cache hit rate
CN113127923A (zh) 管理权限的方法和装置
CN103118049B (zh) 一种通过网络缓存下载文件的方法和系统
CN102298527B (zh) 一种cpe并行升级版本的方法及实现系统
WO2021146896A1 (zh) 文件分享方法、装置及系统
WO2010031297A1 (zh) 一种wap网关提取业务的实现方法和系统
CN106022161B (zh) 一种数据处理方法及装置

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: 09842886

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 13257209

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2009842886

Country of ref document: EP