Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberCN102640472 B
Publication typeGrant
Application numberCN 200980162918
PCT numberPCT/IB2009/007753
Publication date14 Sep 2016
Filing date14 Dec 2009
Priority date14 Dec 2009
Also published asCN102640472A, EP2514171A1, US9049199, US9462051, US20120290677, US20150215396, WO2011073707A1
Publication number200980162918.3, CN 102640472 B, CN 102640472B, CN 200980162918, CN-B-102640472, CN102640472 B, CN102640472B, CN200980162918, CN200980162918.3, PCT/2009/7753, PCT/IB/2009/007753, PCT/IB/2009/07753, PCT/IB/9/007753, PCT/IB/9/07753, PCT/IB2009/007753, PCT/IB2009/07753, PCT/IB2009007753, PCT/IB200907753, PCT/IB9/007753, PCT/IB9/07753, PCT/IB9007753, PCT/IB907753
InventorsH.普塔拉特, A.达莫拉, S.黑尔奎斯特, A.约翰逊
Applicant瑞典爱立信有限公司
Export CitationBiBTeX, EndNote, RefMan
External Links: SIPO, Espacenet
动态缓存选择方法和系统 Dynamic cache selection methods and systems translated from Chinese
CN 102640472 B
Abstract  translated from Chinese
用于选择在通信网络[80]中连接的多个缓存[14]可用的资源[20]的节点、计算机软件和方法。 Cache node for selecting a plurality of communication networks [80] in connection [14] available resources [20], the computer software and methods. 该方法包括从用户[18]接收对所述资源[20]的请求;标识所述多个缓存[14]中存储所述资源[20]的一个或多个缓存[14a、14b、14c];确定所述用户[18]与存储所述资源[20]的所述一个或多个缓存[14]中的每个缓存[14a、14b、14c]之间的路径所关联的总成本,所述总成本包括反映所述通信网络[80]的拓扑的静态成本[Cf]和指示所述通信网络[80]的变化状况的动态成本[Δ];以及基于一个缓存[14a]的最低总成本从存储所述资源[20]的一个或多个缓存[14a、14b、14c]选择所述一个缓存[14a]。 The method includes receiving a request for the resource [20] from the user [18]; identifying the plurality of cache [14] storing the resources [20] One or more cache [14a, 14b, 14c]; determining the user [18] and the storage resource [20] of said one or more cache [14] in each cache [14a, 14b, 14c] between the total cost associated with the path, the the total cost includes the cost of dynamic reflects the communications network [80] the cost of a static topology [Cf] and indicates that the communications network [80] change the situation [Δ]; and based on a cache [14a] from the lowest total cost the storage resource [20] or more cache [14a, 14b, 14c] to select the cache [14a].
Claims(18)  translated from Chinese
1. 一种用于选择在通信网络[80]中连接的多个缓存[14]可用的资源[20]的方法,所述方法的特征在于: 从用户[18 ]接收对所述资源[20 ]的请求; 标识所述多个缓存[14]中存储所述资源[20]的一个或多个缓存[14a、14b、14c]; 确定所述用户[18]与存储所述资源[20]的一个或多个缓存[14]中的每个缓存[14a、 14b、14c]之间的路径所关联的总成本,其中所述总成本包括反映所述通信网络[80]的拓扑的静态成本Cf和指示所述通信网络[80]的变化状况的动态成本△;以及基于一个缓存[14a]的最低总成本从存储所述资源[20]的一个或多个缓存[14a、14b、 14c]选择所述一个缓存[14a], 其中所述总成本与所述静态成本和所述动态成本之间的差有关; 还包括: 计算用于相同路径的动态成本与静态成本之间的差; 对所计算的差应用可信度函数以使所计算的差为时间相关的,以及确定所计算的差是否在预定范围内;以及在成本函数中组合与所接收的参数关联的动态成本和与指派成本关联的静态成本以确定所述总成本; 其中所述指派成本是运营商定义的量且确定所述静态成本。 1. A method for selecting a communication network in a plurality of cache [80] in the connection [14] of the resources available [20], said method characterized by: receiving from a user [18] to the resource [20 ] request; identifying the plurality of cache [14] storing the resources [20] one or more cache [14a, 14b, 14c]; determining the user [18] and the storage resource [20] one or more cache [14] each cache [14a, 14b, 14c] path between the total cost associated, which reflect the total cost includes the cost of the static communications network [80] topology dynamic cost △ Cf and instruct the communication network [80] change in status; and based on a cache [14a] of the lowest total cost of storage resources from a [20] or more cache [14a, 14b, 14c] selecting the one cache [14a], wherein the difference between the total cost and the related costs of the static and the dynamic between the cost; further comprising: calculating a difference between the cost of the same dynamic and static path between the cost; for the calculated differential applications credibility function so that the calculated difference is time-dependent, and determining whether the calculated difference is within a predetermined range; and the combination is associated with the received parameters in the cost function and a dynamic cost assignment static cost associated costs to determine the total cost thereof; wherein the amount of the assignment cost is operator-defined to determine the static and cost.
2. 如权利要求1所述的方法,还包括: 将所述通信网络的节点之间的真实路径转换成在任何缓存对之间包括单个路径的虚拟模型,其中所述节点将所述多个缓存彼此连接。 2. The method of claim 1, further comprising: converting the real path between nodes of the communication network between any pair of cache comprising a virtual model of a single path, wherein said plurality of said node cache connected to each other.
3. 如权利要求2所述的方法,还包括: 对所述虚拟模型中的每个路径赋予指派成本。 The method as claimed in claim 2, further comprising: a virtual model of said each path in the assigned assignment cost.
4. 如权利要求3所述的方法,还包括: 接收表征所述真实路径的性能的参数; 生成表征任何缓存对的状态矩阵;以及基于所接收的参数来生成所述动态成本。 4. The method of claim 3, further comprising: receiving said real path characterizing the performance parameter; generating any characterization of the cache state matrix; and based on the received parameters to generate the dynamic cost.
5. 如权利要求4所述的方法,其中所接收的参数包括抖动、分组丢失、往返时间、IP层拓扑、路径可用容量等中的一个或多个。 5. The method of claim 4, wherein the received parameters including jitter, packet loss, round-trip time, IP layer topology, paths available capacity of one or more.
6. 如权利要求4所述的方法,还包括: 通过使用以下中的一个或多个来填充所述状态矩阵的元素:(i)自动测量系统,其中给定某个时间段两个节点以某个概率来在彼此之间执行测量;(ii)作为测量探测的应用数据;以及(iii)所述网络的节点的被动监视。 6. The method of claim 4, further comprising: by using the following elements in a filling or state of the plurality of matrices to: (i) automatic measurement system, wherein the two nodes in a given period of time to a probability of performing the measurements among each other; (ii) application as a measurement data detection; passive and (iii) the monitoring of the network node.
7. 如权利要求6所述的方法,其中所述状态矩阵的元素至少包括所述网络的测量参数、 所述测量参数的方差、与进行所述测量的时间关联的时间戳以及指示自此后不能够使用所述测量参数的时间的超时参数。 7. The method of claim 6, wherein said state matrix comprises elements of at least the measured parameters of the network, the variance of the measured parameters, the measurements performed with the associated time stamp indicating the time since then no It can be measured using the time parameter of the timeout parameter.
8. 如权利要求1所述的方法,还包括: 将所述成本函数计算为(i)所述静态成本与(ii)所计算的差与所述可信度函数的积之间的和。 8. The method of claim 1, further comprising: calculating the cost function as (i) the cost and volume and static (ii) and the calculated difference between the reliability function.
9. 如权利要求8所述的方法,还包括: 选择所述可信度函数以产生仅在零与一之间的数。 9. The method of claim 8, further comprising: said selected function to generate confidence number between zero and only one.
10. -种用于选择在通信网络[80]中连接的多个缓存[14]可用的资源[20]的节点[14、 1600],所述节点[14、1600]的特征在于: 接口[1608],配置成连接到所述通信网络[80]的其它节点[12、14]和/或多个缓存[14];以及处理器[1602 ],连接到所述接口[ 1608 ]并配置成: 从用户[18 ]接收对所述资源[20 ]的请求, 标识所述多个缓存[14]中存储所述资源[20]的一个或多个缓存[14、14b、14c], 确定所述用户[18]与存储所述资源[20]的一个或多个缓存[14]中的每个缓存[14、 14b、14c]之间的路径所关联的总成本,其中所述总成本包括反映所述通信网络[80]的拓扑的静态成本Cf和指示所述通信网络[80]的变化状况的动态成本△,以及基于一个缓存[14a]的最低成本从存储所述资源[20]的一个或多个缓存[14、14b、14c] 选择所述一个缓存[14a], 其中所述总成本与所述静态成本和所述动态成本之间的差有关; 其中处理器还被配置: 计算用于相同路径的动态成本与静态成本之间的差; 对所计算的差应用可信度函数以使所计算的差为时间相关的,以及确定所计算的差是否在预定范围内;以及在成本函数中组合与所接收的参数关联的动态成本和与指派成本关联的静态成本以确定所述总成本; 其中所述指派成本是运营商定义的量且确定所述静态成本。 10. - kind of used to select multiple caches communications networks [80] in connection [14] available resources [20] nodes [14, 1600], the node [14,1600] is characterized in that: the interface [ 1608], to configure the other nodes connected to a communication network [80] of [12, 14] and / or a plurality of buffer [14]; and a processor [1602], connected to said interface [1608] and arranged : receiving from the user [18] request for the resource [20], and identifying the plurality of cache memory of the [14] resources [20] one or more cache [14,14b, 14c], determine the said user [18] and the storage resources [20] one or more cache [14] each cache [14, 14b, 14c] between the total costs associated with the path, where the total cost includes dynamic cost △ reflect the communications network [80] static topology cost Cf and indicating said communication network [80] of the status change, based on a cache and [14a] of the lowest cost resource from storage [20] one or more cache [14,14b, 14c] to select the cache [14a], wherein the difference between the total cost and the related costs of the static and the dynamic between the cost; wherein the processor is further configured to: calculate for the difference between the cost of the same dynamic and static route between the cost; calculated on the difference between the application function so that the reliability of the calculated difference is time-dependent, and determining whether the calculated difference is within a predetermined range; and dynamic cost combined cost function parameters associated with the received and assigned a static cost associated costs to determine the total cost; wherein the amount of the costs are assigned to an operator-defined and determine the static costs.
11. 如权利要求10所述的节点,其中所述处理器还配置成: 将所述通信网络的节点之间的真实路径转换成在任何缓存对之间包括单个路径的虚拟模型,其中所述节点将所述多个缓存彼此连接。 11. A node as claimed in claim 10, wherein the processor is further configured to: convert the real path between nodes of the communication network a virtual buffer model between any pair comprising a single path, wherein the the plurality of nodes connected to each other caches.
12. 如权利要求11所述的节点,其中所述处理器还配置成: 对所述虚拟模型中的每个路径赋予指派成本。 12. The node according to claim 11, wherein the processor is further configured to: assign given the cost of each path in the virtual model.
13. 如权利要求12所述的节点,其中所述处理器还配置成: 接收表征所述真实路径的性能的参数; 生成表征任何缓存对的状态矩阵;以及基于所接收的参数来生成所述动态成本。 13. A node as claimed in claim 12, wherein the processor is further configured to: receive the real path performance characterization parameters; generating any characterization of the cache state matrix; and based on the received parameters to generate the dynamic cost.
14. 如权利要求13所述的节点,其中所接收的参数包括抖动、分组丢失、往返时间、IP层拓扑、路径可用容量等中的一个或多个。 14. The node according to claim 13, wherein the received parameters including jitter, packet loss, round-trip time, IP layer topology, paths available capacity of one or more.
15. 如权利要求13所述的节点,其中所述处理器还配置成: 通过使用以下中的一个或多个来填充所述状态矩阵的元素:(i)自动测量系统,其中给定某个时间段两个节点以某个概率来在彼此之间执行测量;(ii)作为测量探测的应用数据;以及(iii)所述网络的节点的被动监视。 15. The node according to claim 13, wherein the processor is further configured to: the element by using the following one or more of the state of filling of the matrices: (i) automatic measurement system, in which a given two nodes in a certain period of time to perform the measurement probability of each other; (ii) application as a measurement data detection; passive and (iii) the monitoring of the network node.
16. 如权利要求15所述的节点,其中所述状态矩阵的元素至少包括所述网络的测量参数、所述测量参数的方差、与进行所述测量的时间关联的时间戳以及指示自此后不能够使用所述测量参数的时间的超时参数。 16. The node according to claim 15, wherein said state matrix comprises elements of at least the measured parameters of the network, the variance of the measured parameters, the measurements performed with the associated time stamp indicating the time since then no It can be measured using the time parameter of the timeout parameter.
17. 如权利要求10所述的节点,其中所述处理器还配置成: 将所述成本函数计算为(i)所述静态成本与(ii)所计算的差与所述可信度函数的积之间的和。 17. The node according to claim 10, wherein the processor is further configured to: calculate the cost function of (i) the cost of the static and (ii) the difference between the calculated and the reliability of the function and the plot between.
18. -种用于选择在通信网络[80]中连接的多个缓存[14]可用的资源[20]的设备,所述设备包括: 用于从用户[18]接收对所述资源[20]的请求的部件; 用于标识所述多个缓存[14]中存储所述资源[20]的一个或多个缓存[14a、14b、14c]的部件; 用于确定所述用户[20]与存储所述资源[20]的一个或多个缓存[14]中的每个缓存[14a、14b、14c]之间的路径所关联的总成本的部件,其中所述总成本包括反映所述通信网络[80]的拓扑的静态成本Cf和指示所述通信网络[80]的变化状况的动态成本△;以及用于基于一个缓存[14a]的最低成本从存储所述资源[20]的一个或多个缓存[14a、 14b、14c]选择所述一个缓存[14a]的部件, 其中所述总成本与所述静态成本和所述动态成本之间的差有关; 还包括: 用于计算用于相同路径的动态成本与静态成本之间的差的部件; 用于对所计算的差应用可信度函数以使所计算的差为时间相关的,以及确定所计算的差是否在预定范围内的部件;以及用于在成本函数中组合与所接收的参数关联的动态成本和与指派成本关联的静态成本以确定所述总成本的部件; 其中所述指派成本是运营商定义的量且确定所述静态成本。 18. - kind of used to select multiple caches in a communication network [80] in connection [14] available resources [20], the apparatus comprising: means for receiving from the user [18] to the resource [20 ] of the member requests; means for identifying the plurality of cache [14] storing the resource [20] one or more cache [14a, 14b, 14c]; a unit for determining the user [20] a memory of the resources [20] or more cache [14] each cache [14a, 14b, 14c] of the total cost of the path between the associated components, which reflect the total cost includes the and based on a cache for a [14a] of the lowest cost resource from storage [20]; △ dynamic cost communications network [80] static topology cost Cf and indicating said communication network [80] of the status change or multiple cache [14a, 14b, 14c] to select the cache [14a] member, wherein the difference between the total cost and the related costs of the static and the dynamic between the cost; further comprising: means for computing with poor dynamic cost components in the same path between the static cost; for poor application reliability function calculated so that the calculated difference is time-dependent, and determining whether the calculated difference within a predetermined range and dynamic cost for the combined cost function associated with the received parameters and the costs associated with the static costs assigned to determine the total cost of the part;; member of an operator-defined wherein the amount of the assignment cost is determined and the static costs.
Description  translated from Chinese
动态缓存选择方法和系统 Dynamic cache selection methods and systems

技术领域 TECHNICAL FIELD

[0001] 本发明一般设及系统、软件和方法,W及更具体来说设及用于动态地选择具有期望内容的缓存的机制和技术。 [0001] The present invention is generally installed and systems, software and methods, W, and more specifically located and for dynamically selecting a desired content caching mechanisms and techniques.

背景技术 Background technique

[0002] 在过去数年间,社交网络和内容交付网络(content delivery network,CDN,其在本文中用于涵盖内容交付和内容分发网络二者)的发展是引人注目的。 [0002] In the developments of the past few years, social networking and content delivery network (content delivery network, CDN, which is used to cover the content delivery and content distribution network both in this article) is striking. 越来越多的人优选经由社交网络来联系同辈或朋友或经由CDN来下载内容。 More and more people preferred to contact their peers via social networks or via a CDN or friends to download content. 内容类型包括万维网对象、可下载对象(媒体文件、软件、文档)、应用、实时媒体流和因特网交付的其它组件(DNS、路由和数据库查询)。 Content types include Web objects, downloadable objects (media files, software, documentation), applications, real-time media streaming and other components of Internet delivery (DNS, routing and database queries).

[0003] 手持装置(大多数为移动电话)的发展进度使得从CDN下载内容的过程成为很多用户的常见任务。 [0003] handheld devices (mostly mobile phones) make progress in the development process of downloading content from the CDN become many common tasks the user. 但是,正如接下来论述的,存在CDN智能地将数据文件分配到缓存子集的容量的限制。 However, as discussed next, there is a CDN intelligently allocate data files to limit the capacity of the cache subset.

[0004] CDN是存储数据的副本的、放置在网络中的各种点W便将用于从遍布网络的客户端访问数据的带宽最大化的计算机的系统。 [0004] CDN is a copy of the data is placed in the network at various points throughout the bandwidth of W put from the network for the client to access the data to maximize the computer system. 客户端访问接近该客户端的数据副本,与所有客户端访问相同中央服务器相反,W便避免接近该服务器的瓶颈。 Client access to a copy of the data closer to the client, and all clients accessing the same central server In contrast, W will avoid going near the server bottleneck. 但是,客户端访问"接近" 该客户端的数据的概念并不非常有效率,因为CDN的当前实现似乎缺少智能机制来将数据文件分配到缓存的子集。 However, client access "close" to the concept of data the client is not very efficient, because the current implementation CDN seems to lack intelligence mechanism to allocate data files to the cached subset. 现有CDN所使用的一个原理是只能从客户端直接附接到的边缘节点中的缓存提取数据文件。 A principle used by existing CDN edge node can be attached directly from the client cache to extract data files. 如果客户端需求的数据文件未存储在该边缘节点中,则CDN首先需要将期望的数据文件缓存在那里。 If the client needs the data file is not stored in the edge node, the CDN will first need to desired data file cache there.

[000引为了更好地理解传统CDN网络,图1示出包括多个节点12和14的真实世界网络10的简化示例。 [000 lead to a better understanding of traditional CDN networks, Figure 1 shows a simplified example includes a plurality of nodes 12 and 14 real-world network 10. 最下一行的圆表示边缘节点14,而其余的圆表示中间节点12。 The bottom row circle 14 represents an edge node, and the remaining circles indicate the intermediate node 12. 假定每个最终用户计算机(客户端)可W连接到仅一个边缘节点14。 Assume that each end-user's computer (client) W can be connected to only one edge node 14. 由图1中的从客户端的计算机或移动装置18到对应边缘节点14的虚线来指示运种连接16。 Figure 1 from a computer or mobile device client edge node 18 to the corresponding dotted line 14 to indicate the types of transport connection 16. 还假定仅将内容或数据文件缓存在边缘节点14上。 It is also assumed only the content or data files are cached on the edge node 14. 在此特殊示例中,虚线22指示将一个特定数据文件20存储在=个缓存14a、14b和14c中。 In this particular example, the dashed line 22 indicating a particular data file stored in = 20 buffers 14a, 14b and 14c. 注意,虚线22不表示通信链路。 Note that the dotted line does not represent a communication link 22.

[0006] 网络10中的中间节点12将全世界的边缘节点14彼此连接。 [0006] The network 10 to the intermediate node 12 connected to each edge node 14 over the world. 节点12和14之间的线是运些节点之间的真实链路。 Line between nodes 12 and 14 is the link between the real transport those nodes. 在实践中,所有节点12和14可W是路由器。 In practice, all the nodes 12 and 14 may be W is a router. 真实节点12和14之间的每个链路24与由字母"C"指示的通信成本关联。 Real costs associated link 24 and the nodes 12 and 14 between each of the letter "C" indicates a communication. 两个边缘节点14之间的通信路径可W取网络10中的不同途径,如图1中所示。 A communication path between the two edges 14 at the W node network 10 to take different pathways, as shown in Fig.

[0007] 因此,当用户18向用户18连接到的边缘节点14d请求数据文件20时,假定节点14d 没有该数据文件,则边缘节点Hd需要标识哪个其它节点具有该数据文件,W及哪个节点最接近。 [0007] Thus, when the user 18 is connected to the user 18 to the edge node 14d request data file 20, assume that node 14d not the data file, then the edge node Hd need which other node identifier with the data file, W, and which node is the most close. 现有CDN网络可W将与链路24关联的静态成本纳入考虑。 W existing CDN network may be included in the link 24 associated with static cost considerations. 但是,在网络的状况中的任何变化未在现有静态成本中反映,因为此成本不会变化,并且因此,现有CDN网络缺乏此信息并且无法精确地描述实况网络。 However, any change in the status of the network is not reflected in the existing static costs, because this cost does not change, and therefore, this lack of information existing CDN network and can not accurately describe the live network.

[000引因此,需要找到一种方法将请求的数据本地化到存储该数据文件的副本的"最接近"缓存。 [000 cited therefore, need to find a way to localize the requested data to store a copy of the data file is "closest" to the cache. 在此情况中,最接近意味着该缓存节点至客户端连接到的边缘节点具有最小成本的路径。 In this case, it means that the cache closest to the client node connected to the edge node has a path of least cost. 现有CDN中所没有的另一个特征是在给定网络中有限存储量的情况下且为了优化网络资源,能够决定将哪个数据文件存储在哪些边缘节点上的智能机制。 Another feature of the existing CDN that are not in the case of a given network and the limited amount of memory in order to optimize network resources, can decide which data files are stored intelligent mechanism on which edge node.

[0009] 现有CDN的另一个问题在于,虽然有若干个主动端到端探测方法(用于表征网络链路或路径的能力的通用技术),并且新方法正在持续地开发和精化,但是从此类测量获益的应用经常要求比仅仅端到端估计更多的知识。 [0009] Another existing problem is that the CDN, although there are a number of active end detection method (used to characterize the ability of a network link or path common technique), and new methods are continuously developed and refined, but benefit from such measurement applications often require more than just knowledge-end estimate. 另一个问题是,对于如何维护和汇总来自若干源的测量数据,则没有明确的协议。 Another problem is how to maintain and aggregate measurement data from several sources, there is no clear agreement. 例如,CDN实现常常依赖于一些现有方法W推断用于服务内容请求的"最佳"源节点。 For example, CDN achieve some prior methods often rely on W inferred for the "best" source node content service requests. 为了CDN能够基于网络测量进行智能内容迁移,需要在所有节点之间保持相干测量统计的系统。 In order to be able to network-based measurement CDN intelligent content migration, it is necessary to maintain the coherence between all nodes in the statistical measurement system.

[0010] 因此,会期望提供避免前述问题和缺点的装置、系统和方法。 [0010] Accordingly, it is desirable to provide avoids the aforementioned problems and disadvantages of the devices, systems and methods. WO 98/31107公开一种称为复制品路由选择的新方法,其自动地将请求服务的客户端计算机定向到用于该服务的服务器复制品。 WO 98/31107 discloses a new method called copy routing, the client computer will be automatically directed to service requests for copies of the server service. 美国专利申请09/01 721 67针对一种用于管理客户的内容分发资源的方法,其包括提供用于管理内容分发的内容分发网络中屯、、提供多个CDN、提供要分发的客户的内容W及选择所述多个CDN的至少其中之一W用于由所述CDN中屯、分发内容。 U.S. Patent Application 09/01 72 167 is directed to a content management customer resource distribution method comprising providing content delivery network for managing content distribution Tuen ,, provide multiple CDN, providing customers to distribute content W and selecting the at least one of the plurality of CDN W used by the CDN Tuen, distributing content. WO 02/ 25463公开一种方法和设备,其提供从很多数据服务器中间进行最优数据服务器的动态路由选择,所述很多数据服务器形成通信网络上的分布式联系缓存。 WO 02/25463 discloses a method and apparatus that provides a lot of data from the server intermediate optimal dynamic routing data servers, the servers form a distributed contact a lot of data cache on the communication network.

发明内容 SUMMARY

[0011] 期望将资产(例如,内容)从对于用户和/或运营商最适合的位置提供到用户。 [0011] the desired asset (for example, content) from the user and / or operator of the most suitable location or provide to users. 适合的位置可W是对网络的运营商最便宜的位置。 W is a suitable location can be on the network operators cheapest location. W此方式,从对网络的运营商最便宜的位置接收内容释放了网络中的带宽和/或平衡网络的路径。 W in this way, from the reception of the network operators cheapest location content released network path bandwidth and / or the balance of the network. 还可W通过优化如流传输媒体的某些应用所要求的路径特征来确定适合的位置。 W may also be characterized by optimizing certain applications, such as flow paths of a transmission medium required to determine the suitable location.

[0012] 根据一示范实施例,有一种方法用于选择在通信网络中连接的多个缓存可用的资源。 [0012] According to an exemplary embodiment, there is a method for selecting a plurality of buffers in a communications network connection available resources. 该方法包括从用户接收对资源的请求;标识多个缓存中存储该资源的一个或多个缓存; 确定用户与存储该资源的一个或多个缓存中的每个缓存之间的路径所关联的总成本,该总成本包括反映通信网络的拓扑的静态成本和指示通信网络的变化状况的动态成本;W及基于一个缓存的最低总成本从存储该资源的一个或多个缓存选择该一个缓存。 The method includes receiving a request for a resource from a user; identifying a plurality of cache memory cache one or more of the resource; to determine the user path cache between each cache stores the resources associated with the one or more the total cost of the total cost includes the cost of the dynamic changes in the status of the static cost and instructs the communication network reflects the communications network topology; W and is based on a minimum total cost cache, select the cache from a cache to store one or more of the resource.

[0013] 根据另一个示范实施例,有一种节点用于选择在通信网络中连接的多个缓存可用的资源。 [0013] In accordance with another exemplary embodiment, there is a plurality of buffers for selecting a node in a communications network connection available resources. 该节点包括配置成连接到通信网络的其它节点和/或多个缓存的接口; W及连接到该接口的处理器。 The node is configured to include the other nodes connected to the communication network and / or a plurality of cache interface; the W and is connected to the processor interface. 该处理器配置成从用户接收对资源的请求;标识多个缓存中存储该资源的一个或多个缓存;确定用户与存储该资源的一个或多个缓存中的每个缓存之间的路径所关联的总成本,该总成本包括反映通信网络的拓扑的静态成本和指示通信网络的变化状况的动态成本;W及基于一个缓存的最低总成本从存储该资源的一个或多个缓存选择该一个缓存。 The processor is configured to receive a request for a resource from a user; identifying a plurality of cache memory cache one or more of the resource; to determine the path of one or more cache user and stores the resources in each cache between the the total cost associated with the total cost includes the dynamic changes in the cost and the cost of the static condition of instructing the communication network to reflect the communications network topology; and the W on the lowest total cost of a buffer selected from the one or more caches that store a resource cache.

[0014] 根据又一个示范实施例,有一种包括计算机可运行指令的计算机可读介质,所述指令在被运行时,实现一种用于选择在通信网络中连接的多个缓存可用的资源的方法。 [0014] According to still another exemplary embodiment, there is a computer comprising a computer readable medium operational instructions, the instructions when running, implement a cache for selecting a plurality of connections in a communications network resources available method. 该方法包括从用户接收对资源的请求;标识多个缓存中存储该资源的一个或多个缓存;确定用户与存储该资源的一个或多个缓存中的每个缓存之间的路径所关联的总成本,该总成本包括反映通信网络的拓扑的静态成本和指示通信网络的变化状况的动态成本;W及基于一个缓存的最低总成本从存储该资源的一个或多个缓存选择该一个缓存。 The method includes receiving a request for a resource from a user; identifying a plurality of cache memory cache one or more of the resource; to determine the user path cache between each cache stores the resources associated with the one or more the total cost of the total cost includes the cost of the dynamic changes in the status of the static cost and instructs the communication network reflects the communications network topology; W and is based on a minimum total cost cache, select the cache from a cache to store one or more of the resource.

[0015] 目的在于克服前文中论述的一些缺点,W及提供能够发现对于用户更适合位置的缓存的节点。 [0015] aimed at overcoming some of the disadvantages previously discussed, W, and provide the ability to find more suitable location for the user node's cache. 一个或多个独立权利要求有利地提供最佳缓存W响应用户的请求。 One or more independent claims advantageously provides optimal caching W response to user requests.

附图说明 BRIEF DESCRIPTION

[0016] 并入本说明书中并构成其一部分的附图示出一个或多个实施例,并且连同描述来说明运些实施例。 BRIEF [0016], which are incorporated in and constitute a part of illustrating one or more embodiments and together with the description to illustrate the operation of these examples. 在附图中: In the drawings:

[0017] 图1是传统CDN的示意图; [0017] FIG. 1 is a schematic view of a conventional CDN;

[0018] 图2是根据一示范实施例的状态矩阵的说明; [0018] FIG. 2 is an exemplary embodiment according to the state of the described matrix;

[0019] 图3是发送方和接收方之间的路径的示意图; [0019] FIG. 3 is a schematic view of the sender and receiver path between;

[0020] 图4是彼此连接且具有不同级别的可用带宽的S个不同链路的示意图; [0020] FIG. 4 is a schematic diagram of S different available bandwidth links connected to each other and having different levels;

[0021] 图5是示出网络的节点之间的主动探测方法的示意图; [0021] FIG. 5 is a diagram showing a method of detecting active between the nodes of the network;

[0022] 图6是示出由根据一示范实施例的状态矩阵收集到什么信息的示意图; [0022] FIG. 6 is a diagram showing the state in accordance with an exemplary embodiment of what information is collected from the matrix;

[0023] 图7是根据一示范实施例的状态矩阵的元素的说明; [0023] FIG. 7 is a diagram illustrating an exemplary embodiment according to the state of the elements of the matrix;

[0024] 图8是对应于真实CDN网络的虚拟网络的示意图; [0024] FIG. 8 is a schematic diagram corresponding to the real CDN network virtual network;

[0025] 图9示出根据一示范实施例的与可用带宽估计关联的动态成本; [0025] Figure 9 shows an embodiment of dynamic cost estimate available bandwidth according to an exemplary associated;

[0026] 图10示出根据一示范实施例的与链路容量关联的指派成本; [0026] FIG. 10 shows the link capacity associated with the cost of assignment according to an exemplary embodiment;

[0027] 图11示出根据一示范实施例的静态成本与动态成本之间的成本差; [0027] FIG. 11 illustrates an exemplary embodiment of the costs of static and dynamic cost difference between the cost;

[0028] 图12示出根据一示范实施例的与成本差关联的可信度函数; [0028] FIG. 12 shows the difference between the cost function associated with reliability according to an exemplary embodiment;

[0029] 图13示出根据一示范实施例的CDN网络执行的用于向用户提供期望资源的各种步骤; [0029] Figure 13 illustrates the various steps of the implementation of the embodiment CDN network for providing a desired resource to the user in accordance with an exemplary;

[0030] 图14示出根据一示范实施例的收集有关CDN网络的数据的决策引擎; [0030] FIG 14 illustrates an exemplary decision engine to collect data relating to the CDN network according to an embodiment;

[0031] 图15是示出根据一示范实施例的用于选择期望资源的方法的流程图;W及 [0031] FIG. 15 is a flowchart illustrating an exemplary embodiment of the method of selecting a desired resource; and the W

[0032] 图16是配置成实现图14和/或图15中所示的步骤的节点的示意图。 [0032] FIG. 16 is a schematic configuration diagram of the node in step 14 and / or the implementation shown in FIG. 15.

具体实施方式 detailed description

[0033] 示范实施例的下文描述设及附图。 [0033] Hereinafter described exemplary embodiments and drawings provided. 不同附图中的相同引用号标识相同或相似的单元。 Different figures the same reference numerals identify the same or similar elements. 下文详细描述并不限制本发明。 The following detailed description does not limit the present invention. 而是,本发明的范围由所附权利要求书限定。 Rather, the scope of the invention defined by the appended claims. 为了简明, 下文实施例是关于CDN网络的术语和结构来论述的。 For simplicity, the examples hereinafter is about the terminology and structure to CDN network discussed. 但是,接下来要论述的实施例不限于运些网络,而是可W应用于其它现有网络。 However, the embodiments to be discussed next are not limited to these transport networks, but may be applied to other existing web W.

[0034] 本说明书通篇对"一个实施例"或"一实施例"的引述意味着结合一实施例描述的特殊特性、结构或特征被包括在本发明的至少一个实施例中。 [0034] Reference throughout this specification to "one embodiment" means that a particular feature or refer to an embodiment described, structure, or characteristic "one embodiment" is included in at least one embodiment of the present invention. 因此,本说明书通篇各种位置中出现的短语"在一个实施例中"或"在一实施例中"不一定全部指相同的实施例。 Accordingly, a variety of locations throughout this specification appearances of the phrases "in one embodiment" or "in an embodiment" are not necessarily all referring to the same embodiment. 另外,可W任何适合的方式在一个或多个实施例中进行组合特殊特性、结构或特征。 In addition, W in any suitable manner embodiment combined special characteristics, structure, or characteristic in one or more embodiments.

[0035] 根据一示范实施例,一种方法和网络配置成使用捕获网络的一个或多个属性的成本函数。 [0035] According to an exemplary embodiment, a method and a network configured to use the network to capture one or more attributes of the cost function. 该成本函数包括从主动网络测量推导的成本与网络的真实和/或虚拟节点拓扑的组合。 The cost function comprises a combination of real and / or virtual node topology derived from the active network measurement of the cost of the network. 该主动网络测量方法(例如,参见2009年的爱思唯尔计算机网络(Elsevier Computer 化tworks,COM肥T)53卷15期2617-2645页的Bergfeldt等人的使用过滤和变化检测的实时可用带宽估计,,其全部内容通过引用并入本文)将测量数据存储在稍后定义的状态矩阵中。 The active network measurement method (for example, see 2009 Elsevier computer network (Elsevier Computer technology tworks, COM fat T) 53 15 2617-2645 Volume Bergfeldt et al page filtering and change detection using real-time available bandwidth ,, estimate the entire contents of which are incorporated herein by reference) measurement data is stored is defined in a later state matrix. 主动网络测量将网络的状态纳入考虑。 Active Network Measurement status of the network into account. 基于测量,将动态成本与边缘节点之间的路径关联,而真实和/或虚拟节点拓扑产生静态成本。 Based on the measurement, the cost associated with the dynamic path between edge nodes, while the real and / or virtual node topology generate static costs. 可W由运营商修改静态成本,正如接下来论述的。 W can modify the static costs by the operator, as discussed next. 由决策引擎将每个成本组合W形成总成本。 By the decision engine will cost each combination forming a W total cost. 因此,在CDN中基于包括动态成本和静态成本的总成本来确定由客户端提取的内容或缓存之间的内容迁移。 Thus, in the CDN based on static and dynamic cost includes the cost of the assembly have been determined by the client to extract the contents of the cache contents or between migration. 该决策引擎可W基于状态矩阵、拓扑和可信度函数来执行成本估计,正如接下来论述的。 The decision engine may matrix W based on the state, topology and reliability functions to perform cost estimates, as discussed next.

[0036] 根据另一个示范实施例,引入状态矩阵,并且状态矩阵表示分布式系统的网络测量数据并将其汇总。 [0036] According to another exemplary embodiment, the introduction of the state of the matrix, and the state matrix network measurement data distributed systems and summary. 此信息能够用于选择要在CDN中的两个通信节点之间(例如客户端与缓存节点之间)使用的适合的路径或一组路径。 This information can be used to choose the path that you want (for example, between the client and the caching node) between CDN two communicating nodes or a set of paths.

[0037] 在论述如何计算CDN中的边缘节点之间的各种路径的总成本并使其为时间相关的之前,本发明示范实施例论述用于生成状态矩阵的各种方式。 [0037] In the discussion of how to calculate the total cost of the various paths between the CDN edge node and before it is time dependent, the present exemplary embodiment discussed various ways for generating a state matrix. 状态矩阵可W定义成包括从一组节点阳1,化,...,佩}之间的主动探测或被动监视获得的性能参数估计。 State matrix W can be defined to include a set of nodes from a male, of, ..., performance parameters active or passive monitoring probe} between Pei estimate obtained. 在表示多于一个性能参数的情况中,可能并行地使用若干状态矩阵,每个性能参数对应于一个状态矩阵,或单个状态矩阵中每对节点对应于若干元素。 Representing more than one performance parameter of cases, it may be used in parallel several state matrix, each performance parameter corresponds to a state matrix, or each pair of nodes corresponding to a single state in a number of matrix elements. 性能参数的示例是端到端可用容量、往返时间(RTT)、抖动、IP层拓扑、路径可用容量和丢失(loss)。 Example-end performance parameters is the available capacity, round-trip time (RTT), jitter, the IP layer topology, the path loss and the available capacity (loss). 图2示出状态矩阵的示例,其中元素沿骗述在两个节点Ni和的之间观察到的特定性能参数方面的路径特征。 Figure 2 shows an example of the state of the matrix, where the elements lie along said path between two nodes Ni features and observed performance parameters of specific aspects. 注意,路径Ni至Nj 可W具有与反向路径叫至Ni相比不同的特性。 Note that the path Ni to Nj W may have a reverse path called different characteristics compared to Ni.

[0038] 根据一示范实施例,状态矩阵中的每个元素Xu包括至少如下参数:性能参数估计、 指示何时进行测量的时间戳W及指示何时测量估计到期的超时阔值。 Performance parameter estimation, time stamp indicates when the measurement W and indicates when measuring the timeout threshold value estimate of expire: [0038] According to an exemplary embodiment, the state of each element in the matrix Xu include at least the following parameters. 可选地,该状态估计可W包括估计的方差。 Alternatively, the state estimated W including estimated variance. 可W使用估计的方差为状态矩阵的用户提供有关其可信度的信息。 W can be estimated using the variance matrix of the state to provide users with information about their credibility. 如稍后将论述的,可W使用其它函数来提供此信息。 As will be discussed, W can use other functions to provide this information later.

[0039] 能够采用各种方式来填充状态矩阵中的元素。 [0039] can be employed in various ways to fill the state matrix elements. 下文描述=个示例。 The following describes one example =. 但是,正如本领域技术人员将认识到的,还存在其它可能性。 However, as those skilled in the art will appreciate, there are other possibilities. 第一和第二示例是基于主动探测,并且可W提供如RTT、抖动、延迟、路径可用容量和丢失的性能参数。 The first and second example is based on the active detection, and can provide W as RTT, jitter, delay, path loss available capacity and performance parameters. 第S示例是基于被动监视,并且可W提供例如链路负载、队列长度和丢弃分组的其它性能参数。 The first S example is based on passive monitoring, and can provide, for example W link load, queue length and the packet is discarded in other performance parameters.

[0040] 接下来要论述的示例可W使用本领域中已知的技术。 [0040] Example Next W can be discussed using techniques known in the art. 为了完整性,现在简要地论述运些技术。 For completeness, we briefly discuss some transport technology. 主动探测是通过从发送方向接收方发送分组来表征网络链路或路径的一种通用技术。 Active detection by receiving from a sender sends packets to characterize the network link or path of a common technology. 图3中示出了此方法。 Figure 3 illustrates this method. 更确切地来说,发送方30经由网络32向接收方34发送分组。 More specifically, the sender 30 via the network 32 sends a packet to the receiver 34. 发送方30将分组间间隔36随机化。 The sender packet interval 30 36 randomized. 接收方34计算新的分组间间隔38,其然后被分析W确定期望的性能参数。 Receiver 34 calculates the interval between the new packet 38, which is then analyzed to determine the expected performance parameter W. 能够用此方法测量的性能参数例如是抖动、分组丢失、往返时间和路径可用带宽。 Performance parameters can be measured by this method, for example, jitter, packet loss, round-trip time and available bandwidth. 测量运类性能参数的传统工具例如是各种版本的试通程序(ping)。 Measuring performance parameters of transport category is traditional tools such as various versions of the test through the program (ping). 测量可用容量是更为复杂的问题。 Measurement of available capacity is a more complex issue. 一个运种测量技术是实时可用带宽(BART),并且下一个段落中对此进行描述。 Yun is a kind of real-time measurement techniques available bandwidth (BART), and the next paragraph be described.

[0041] 作为本领域中已知的技术,BART是一种用于估计分组交换网络路径上实时的路径可用容量和其它容量有关参数的方法。 [0041] As known in the art of technology, BART is a method for estimation of available capacity and other parameters related to the capacity of a packet-switched network path method of real-time path. 图4中示出性能参数容量、利用和可用容量的概念, 其示出作为网络路径的一部分的=个链路40、42和44。 Figure 4 shows the performance parameters of capacity, and the use of available capacity in the concept, which is illustrated as a = a network path 40, 42 and 44 of the link portion. 每个链路具有定义能够发送IP业务的最大速率的标称容量。 Each link has a defined capable of transmitting IP traffic maximum rate of nominal capacity. 在任何给定时间点,可W按小于其最大容量来利用链路。 At any given point in time, you can press W less than its maximum capacity to take advantage of the link. 运由与每个链路关联的阴影区域来举例说明。 Transported by the shaded area associated with each link to illustrate. 对每个链路将可用带宽计算为标称链路带宽与估计的链路利用之差。 Available bandwidth for each link will be calculated as the difference between the nominal link bandwidth estimation of link utilization.

[0042] -个容量有关性能参数是端到端可用容量。 [0042] - a performance parameter is related to capacity-end available capacity. 此参数定义为一组连续链路的最小链路容量。 This parameter is defined as a set of minimum link capacity continuous link. 在图4中,链路40是最小端到端可用容量。 In Figure 4, the link end 40 is the minimum available capacity. BART方法依赖于在网络路径上主动发送探测业务W便确定在哪个探测速率路径显示拥塞的征兆。 BART method relies on a network path sends probe traffic W will determine which path detection rate in the display of signs of congestion. 此速率定义可用容量,即,未被IP 业务利用的容量的部分。 This rate is defined portion of the available capacity, that is, not business use of IP capacity. 如图3中所示,BART发送方30正在按随机化的分组间间隔36向接收方34传送IP分组。 As shown in Figure 3, BART is 30 according to the sender of the packet interval between randomization 36 transmit IP packets to the receiver 34. 该间隔受共享网络的其它IP业务的影响。 Affected by the shared network other IP services that interval. 接收方36对每个入局IP分组标记时间戳,并且计算新的分组间间隔38。 Receiver 36 pairs each incoming IP packet-stamp, and calculates the new inter-packet interval 38. 可W通过卡尔曼过滤器化alman filter)分析在发送方30和接收方34的分组间间隔,运是一种实时地跟踪不可直接观察的特性的统计方法, 并且从该分析的输出是可用容量。 W can Kalman filter of alman filter) packet analysis between the sender 30 and receiver 34 intervals, tracking operation is not directly observable statistical properties of a real-time, and output from this analysis is available capacity .

[0043] 可W由上文提到的第一示例,即自动测量系统使用上文论述的主动探测。 [0043] W may be the first example from the above mentioned, namely automatic measurement using active detection system discussed above. 运种系统的特性是使得给定某个时间段两个节点W某个概率来在彼此之间执行测量。 Characteristics of transport such systems is such that a given probability for a period two nodes W to perform measurements at each other. 该测量系统在参与节点上作为后台进程来运行测量。 The measuring system as a background process to run measurements on participating nodes. 如图5中所示,在给定时间点,节点B触发与节点D 的测量,而同时节点E触发与节点A的测量。 As shown in FIG. 5, at a given time point, a Node B triggering and measurement node D, while the measurement node E and the node A trigger. 可W触发网络的所有节点执行运类测量,并且最终W每个节点的数据填充状态矩阵。 W can trigger the execution of transport category measure all nodes of the network, and eventually fill state data W matrix for each node.

[0044] 仍基于主动监视的填充状态矩阵的第二示例,使用应用数据作为测量探测。 [0044] The second example is still based on proactive monitoring of the filling state matrix, using the application data as the measurement probe. 例如, 在其中缓存彼此往返地迁移数据的CDN系统中两个节点交换数据时,运些节点能够在应用数据分组中梢带探测信息。 For example, CDN caching system in which each other and from the migration of data when exchanging data in both nodes, those nodes can be transported with the tip probe information in the application data packets. 当运种方法要求在探测业务方面的更多开销时,该方法可W用于估计端到端可用容量。 When the transport method requires more overhead in terms of probing operations, this method can be used to estimate the W-end available capacity. 例如,试通程序和RlT无需再用应用数据。 For example, test procedures and pass RlT eliminates the need for application data.

[0045] 填充状态矩阵的第=示例是基于被动监视,而非主动监视。 [0045] The first filling state matrix = example is based on passive surveillance, rather than proactive monitoring. 此方法使用路由器、缓存和其它节点中的业务的被动观察W便确定负载、丢失和相似参数。 This method uses a router, caching and other nodes in the business of passive observation W will determine the load loss and similar parameters. 此方法需要对中间网络节点的访问权W及用于检索数据的标准化协议(例如,简单网络管理协议(SNMP))。 This method requires access to intermediate network node W and a standardized protocol for retrieving data (e.g., Simple Network Management Protocol (SNMP)). 因此, 运些网络单元需要驻留在运营商自己的域内或运营商需要对所设及的网络单元的访问权。 Therefore, some transport network elements reside on the operator's own domain or operator needs to be established by the network elements and access.

[0046] 所有运=个示例W及关联的方法均可W用于填充状态矩阵。 [0046] example of W = all transport and associated method can be used to fill the W state matrix. 根据图6中所示的一示范实施例,状态矩阵60使用(i)自动测量方法62来接收第一组参数,例如RTT、抖动、延迟、 丢失、路径可用带宽等(ii)应用数据测量探测方法64W用于确定第二组参数,例如,RTT、抖动、延迟、丢失、路径可用带宽等W及(iii)被动监视方法66W用于确定第=组参数,例如, 链路负载、丢弃分组等。 According to an exemplary embodiment shown in FIG. 6 embodiment, the state matrix 60 (i) a method of automatic measurement 62 receives the first set of parameters, such as the RTT, jitter, delay, loss, available bandwidth, etc. (ii) application of the measurement data detection 64W method for determining a second set of parameters, e.g., the RTT, jitter, delay, loss, and available bandwidth W, etc. (iii) passive monitoring method for determining 66W = first set of parameters, such as link load, the packet is discarded, etc. .

[0047] 根据一示范实施例,可W使用方法62、64和66中的仅一种方法或其组合来填充状态矩阵60。 [0047] According to an exemplary embodiment, the use of W 62, 64 and 66, only one method or a combination to fill the state matrix 60. 但是,因为来自运些方法的测量可能不是同时进行的,即运些测量可能具有宽时间段内的不同时间戳,所W执行测量的时间和使用测量的时间能够在时间窗口内从数秒改变到数分钟或数小时。 However, because the measurement from the operation of these methods may not be performed simultaneously, i.e. these measurements may be shipped with a wide period of time stamps differ, the execution time measured W and using the measured time within the time window can be changed from several seconds to minutes or hours. 还可能的情况是对于特殊路径未实行测量。 Also possible is a special path for failure to implement the measure. 因此,根据一示范实施例,引入可靠性函数W提供在状态矩阵中考虑测量的时间相关性的方式。 Thus, according to an exemplary embodiment, the introduction of reliability function W provided with measurement of the correlation matrix state in the way of time. 换言之,引入可信度函数,其配置成滤出基于时间的相关性。 In other words, the credibility of the introduction of a function that is configured to filter out the time-based correlation.

[0048] 此函数可W基于估计的方差,并且确保性能参数估计的关联性随着时间降低。 [0048] This function W based on the estimated variance, and ensure the performance of parameter estimation correlation decreases over time. 降低的速率取决于感兴趣的性能参数。 Reduced interest rate depends on the performance parameters. 例如,RTT估计在时间上比可用容量的估计更稳定。 For example, RTT is estimated at more stable over time than the estimate of available capacity. 可信度可W由函数f(P,To, T)表示,其中T是当前时间,To是超时时间,W及P是特定性能参数类型,例如RTT或可用带宽。 Reliability at the W by the function F. (P, To, T), where T is the current time, To is the timeout period, and W is P type specific performance parameters, such as available bandwidth or RTT. 函数f的特定定义可W取决于应用而改变,正如接下来将论述的。 Specific definition of the function f W may vary depending on the application, as will be discussed next.

[0049] 图7中更详细地示出状态矩阵60的元素Xu。 [0049] FIG. 7 shows in more detail the elements Xu state matrix 60. 每个元素可W包括给定参数的估计/测量值70、与值70关联且指示测量或估计该值的时间的时间戳72、与估计值/测量值70关联的超时时间To 74W及估计/测量值70的方差76。 Each element can be estimated to include W / measured value 70, 70 is associated with and is indicative of the value measured or estimated value of the timestamp of the time 72 of the given parameters, the estimated value / measured value 70 associated timeout To 74W and estimates / measurements of variances 7670. 该方差能够是用于指定图12中的可信度函数的附加参数。 The variance can be used to specify the function of the credibility of FIG. 12 additional parameters. 如果方差高,则静态成本线上方或下方的曲线更快速地向此线收敛。 If the variance is high, the cost side or the static curve below the line more quickly converge to this line. 例如,高方差可W由静态线上方的曲线表示,而低方差能够由静态线下方的曲线来举例说明。 For example, the variance can be high by curve W represents the static line side, low variance can be curved by the static line below to illustrate.

[0050] 在论述如何将成本函数应用于测量的参数W及该新的智能网络如何能够确定用户所要求内容的最佳缓存之前,论述再一个组成。 [0050] In the discussion on how the cost function is applied to the measurement of parameters W and the new intelligent network before the user can determine how content requires the best cache, and then discusses a composition. 运设及上文引入的静态成本。 Transport and set up the introduction of the above static costs. 为了确定网络节点之间的链路的静态成本,可W使用分级拓扑逼近技术。 In order to determine the cost of static links between network nodes, W using a hierarchical topology approximation technique. 分级拓扑逼近技术将图1中所示的真实网络转换成图8中所示的在任何边缘节点对之间具有相同路径成本的分级网络。 Hierarchical network topology in real approximation technique shown in Figure 1 have the same mapping converting hierarchical network cost path between any pair of edge nodes in FIG. 8. 分级网络的生成在本领域中是已知的,并且可W用两个步骤来实现。 Hierarchical network generation are known in the art, and W can be achieved in two steps.

[0051] 第一步骤是创建分级虚拟拓扑,W及第二步骤是W将真实网络与虚拟网络之间的路径成本上的差减到最小的方式为虚拟拓扑中的每个链路生成成本。 [0051] The first step is to create a virtual hierarchical topology, W, and W is the second step of the difference between the real cost of the path network and virtual networks on the way to minimize the virtual topology generation cost for each link. 因为图8的虚拟网络是分级的,所W能够将其建模成类似树形。 Because the virtual network Figure 8 is hierarchical, the W can be modeled to resemble a tree. 另外,虚拟网络中任何边缘节点对之间正好存在一个路径。 In addition, there is just one path between any virtual network edge node pair. 运两个特征使得使用较不复杂的算法来计算边缘节点之间的静态成本成为可能。 Yun two features makes the use of less complex algorithms to calculate the static costs between the edge node possible.

[0052] 在图8中所示的虚拟网络80中,边缘节点14保持与图1的真实网络10中相同,而中间节点82不具有与真实网络10的中间节点12的直接对应性。 [0052] In the virtual network shown in FIG. 880, the edge node 14 to maintain a real network of FIG. 10 of the same, while the intermediate node 82 does not have a real network 10 directly to intermediate node 12 in correspondence. 创建虚拟网络80的特殊方式取决于真实网络、应用和可用技术,如本领域技术人员会认识到的。 Create a virtual network in a special way 80 depends on real network, applications and available technologies, as those skilled in the art will recognize. 但是,真实网络由路由器孤岛组成,且一个孤岛中的路由器较其它孤岛中的路由器更加彼此靠近。 However, the real network composed by routers islands and an island in the router closer to each other than the other islands in the router. 运类孤岛可W对应于分级拓扑中的中间节点。 W can transport category classification island topology corresponds to an intermediate node. 更好地理解虚拟网络的比拟是想象真实社区和街道/道路,W 及一个级别上的虚拟节点可W对应于真实城市,下一更高级别上的虚拟节点可W对应于真实地区,并且再下一更高级别上的虚拟节点可W对应于国家的虚拟模型。 To better understand the analogy is to imagine a virtual network community and real street / road, W and virtual nodes on a level W may correspond to the real city, a virtual node on the next higher level W may correspond to the real area and then virtual node on the next higher level W may correspond to a virtual model of the country. 生成分级网络的此第一步骤可W由网络操作员手动实现,或通过参照包括所需信息的数据库的计算程序来实现。 This first step of generating hierarchical network can be implemented by W network operator manually, or by referring to the calculation program includes a database of information needed to achieve.

[0053] 在第二步骤中,可W使用优化程序来W将分级与真实网络之间的边缘到边缘路径成本上的差最小化的方式对虚拟网络中的链路指派成本。 [0053] In the second step, the optimizer can use the W to W edge between the grading and real network to the edge of the path difference between the cost of a way that minimizes the cost assigned to the virtual network link. 所使用的特殊优化方法取决于真实网络和与运营商有关的其它考虑,正如本领域中已知的各种技术。 Special optimization method used depends on real network with carriers and other considerations relevant, as known in the art of various techniques. 一种方式是通过梯度捜索(gradient search)查找误差函数的最小值。 One way is to find the minimum error function gradient Dissatisfied by cable (gradient search). 该误差函数能够定义为所有路径,即所有可能的边缘节点对之间的所有路径的路径成本上的差的平方和。 The error function can be defined as all the paths that all squared differences path cost of all possible paths between the edge nodes and on. 一旦定义了分级树,应用哪个特殊的树算法就再次是基于问题的更细分细节进行的挑选和审慎选择的问题。 Once you have defined a hierarchical tree, which is a special tree algorithm is once again pick and choose carefully the details of the problem is more segmentation based undertaken.

[0054] 系统的分级树表示使得CDN的运营商/提供商能够指派系统的节点之间的静态成本。 [0054] hierarchical tree representation system makes the CDN operators / providers to assign a static cost between the nodes of the system. 运些静态成本/权重被CDN用在决定提取对象/资产的最佳缓存中。 These static transportation cost / weights are used in determining the extracted object CDN / Asset optimal cache. 运些静态成本是数值,并且可W基于底层链路的一些特征在逻辑上推导。 These transport costs are static value, and may be based on some characteristics of the underlying link W derivation logic. 例如,称为开放最短路径优先(OSPF) 的技术基于链路的带宽来确定静态成本。 For example, the technique known as Open Shortest Path First (OSPF) link bandwidth is determined based on the static costs. 通过将参考带宽(缺省为100 Mbps)除W接口带宽来计算OSPF中的链路的静态成本。 By reference bandwidth (default is 100 Mbps) in addition to static cost W interface bandwidth to calculate the OSPF link. 因此,至目的地的总静态成本是至目的地的路径中的个体链路的静态成本之和。 Therefore, the total cost to the destination of the static cost of a static route to a destination in and of individual links.

[0055] 因为静态成本是逻辑值,所W通过有意向地修改链路的静态成本而不考虑链路的底层特征,运营商可W影响使用该链路的方式。 [0055] Because static cost is a logical value, the W by the intention to modify the static link cost regardless of the underlying characteristics of the link, operators can affect the use of the W link mode. 此能力为运营商运作网络提供更多灵活性。 This ability to provide more flexibility for operators functioning network. 因此,能够将基于底层物理链路的固定特征进行静态成本的初始指派视为指派的成本。 Accordingly, based on the characteristics of the underlying physical link fixed initial cost as a cost assigned statically assigned. 如果静态成本是基于运营商内部策略,则可W将此静态成本视为基于策略的成本。 If the static costs are based on operator internal policies can be seen as the cost of this static W strategy based on cost. 基于策略的静态成本也是指派的静态成本。 Cost based on static static cost strategy is assigned. 但是,静态成本未将网络可能受到的实况修改纳入考虑, 并且因此在决定使用哪个缓存来响应用户的请求时不是精确的参数。 However, the cost is not the network may still be a live modified into account, and thus to determine which buffer in response to a user's request is not an exact parameters.

[0056] 根据一示范实施例,引入对例如由于业务导致的网络的改变敏感的动态成本。 [0056] According to an exemplary embodiment, for example, the introduction of lead due to changes in network traffic sensitive dynamic cost. 可W将动态成本与静态成本组合,从而得到上文提到的新成本函数。 W can be static and dynamic cost costs combined to obtain a new cost function mentioned above. 可W将总成本定义为取决于静态成本和动态成本。 W can be defined as the total cost depends on the cost of the static and dynamic costs. 可W通过引入上文论述的可信度函数f来使总成本C为时间相关的。 W discussed above can be introduced by the credibility of the function f to make the total cost C for the time-related. 在一个特殊实施例中,由(i)静态成本W及(ii)动态成本与可信度函数之间的积的和来给出时间相关的总成本C(t)。 In one particular embodiment, the (i) W static costs and (ii) the cost and reliability of dynamic function between the product and to give time-dependent total cost C (t). 接下来更详细地论述运些量。 Next discussed in more detail some amount of luck.

[0057] 正如已经论述的,状态矩阵可W存储与关联链路的成本不同的一个或多个测量。 [0057] As already discussed, the state matrix W can be stored with the associated costs of different links with one or more measurements. 因此,可W使用归一化函数将状态矩阵的元素Xij转换成动态成本。 Therefore, using W normalization function elemental state matrix Xij is converted into dynamic cost. 换言之,归一化函数将测量的性能参数映射到可W添加到静态成本的动态成本。 In other words, the performance parameters of the normalization function can be mapped to W measured dynamic cost added to the cost of the static. 例如,图9示出链路的可用容量及其关联的动态成本。 For example, Figure 9 shows a dynamic link costs associated with its available capacity. 通过主动测量方法(例如BART)来测量可用容量。 By actively measuring methods (such as BART) to measure the available capacity. 生成第二列中的动态成本的归一化函数可W具有各种形式,其取决于应用。 Generating a second column dynamic cost normalized function W in various forms, depending on the application. 图9中所示的动态成本对应于单个测量参数。 Dynamic Cost shown in Figure 9 correspond to the individual measurement parameters. 如果使用多个参数,则可W确定多个动态成本。 If you use multiple parameters, you can define a plurality of dynamic W cost. 当链路的性能参数变化时,动态成本变化,并且出于此原因,动态成本更好地反映真实网络的成本。 When performance parameters of the link, the dynamic changes in cost, and for this reason, the dynamic costs to better reflect the true cost of the network. 图10在第一列中示出实际容量W及在第二列中示出与此链路容量关联/指派的静态定价。 10 In the first column shows the actual capacity W and the second column shows the capacity associated with this link / assignment of static pricing. 可W使用其它归一化函数。 W can use other normalization function.

[0058] 基于静态成本和动态成本,可W计算两个成本之差A,如图11中第5列中所示。 [0058] based on the static and dynamic cost cost, the cost difference between the two calculated W A, as shown in column 5 of 11 in FIG. 更确切地来说,对于相同的链路,将差A计算为从固定网络拓扑特征获得的链路容量与从主动测量获得的实际带宽之间的成本上的差。 More specifically, for a same link, will be calculated as the difference between the A link capacity obtained from the fixed network topological characteristics and the difference from the actual cost of bandwidth active measurement obtained between. 成本差A可W是负的或正的,如图11中所示。 A cost difference W can be negative or positive, as shown in Figure 11. 因此,成本差A取决于静态(指派的成本)与动态成本二者。 Therefore, the cost depends on the difference between A static (assigned costs) and dynamic cost both.

[0059] 该测量系统和状态矩阵提供在操作中时网络的实时性能参数。 [0059] The measuring system and the state matrix provided in real-time operation of the network performance parameters. 基于上文论述的归一化函数,成本差A实时地标识节点之间的实际逻辑成本。 Based on the above discussed normalization functions, the actual cost of the logical difference between the cost of the node A real-time identification. 运使得CDN能够基于网络的实际状态提供更好的响应。 Yun makes CDN can provide better response based on the actual state of the network.

[0060] 在理想状况下,无论何时CDN逻辑需要节点之间的更新读取,测量系统就能够即时地提供该更新读取。 [0060] In an ideal situation, between whenever CDN node update logic needs to read the measurement system can provide real-time updates to the reading. 但是,在真实世界中,测量花费时间,并且CDN系统可能不等待直到最新结果可用。 However, in the real world, measurement takes time, and CDN system may not wait until the results of the latest available. 因此,CDN逻辑需要依赖于上次或最近进行的测量。 Therefore, CDN logic need to rely on recent or last measurement. 目前时间与进行测量的时间之间的时间差可能改变,从数秒到很长时间或完全无时间差。 At present time and the time difference between the time of measurement may change, from a few seconds to complete a very long time or no time difference.

[0061] 因此,根据一示范实施例,引入实现差A的可信度或可靠性的函数,W便提供在CDN中使用时将测量的信息的时间相关性纳入考虑的方式。 [0061] Thus, according to an exemplary embodiment, is introduced to realize the function of the confidence or reliability of the difference A, W will provide time information will be used when measured in the CDN into account the correlation mode. 上文已论述运种函数f。 The foregoing has discussed shipped species function f. 此函数f 提供使得实际逻辑成本为时间相关的方式。 This logical function f so that the actual cost of providing a time-dependent manner. 现在更详细地论述此概念。 This concept is now discussed in more detail.

[0062] 此函数使用来自网络拓扑的指派成本(Cf)作为基线,如图12中所示。 [0062] This function is used to assign costs (OF Cf) from the network topology as a baseline, as shown in FIG. 如果没有测量信息可用,则基线用作逻辑固定成本。 If there is no measurement information is available, the baseline is used as a logical fixed costs. Tl是当前时间(即时时间)W及To(超时时间)是自此后测量值太旧而无法使用的时间。 Tl is the current time (real-time) W and To (time-out) is the measured value has since time is too old to use. 图12提供可信度函数f(T,To)的示例。 Figure 12 provides a confidence function f (T, To) of the sample. 按运营商认为适合的W及正如本领域技术人员会认识到的,可W使用其它函数。 Press operators may think fit and W as skilled in the art will recognize, you can use the other functions W. 可信度函数的形状可W取决于估计方差或取决于所使用的特定性能参数的固有特性。 W shape credibility function may depend on the inherent characteristics of the estimated variance or specific performance parameters depend on the use.

[0063] 可信度函数f随时间推移使差A的值减少。 [0063] The function f credibility over time so that the difference between the value of A is reduced. 如果上次测量刚好在当前时间Tl之前进行,则计算的差A能够从0改变至无限大,如图12中所示。 If just the last measurement carried out before the current time Tl, the calculated difference A can change from 0 to infinity, as shown in FIG. 随着测量时间与目前时间之间的间隙变得越来越大,由于测量的可信度减少,所WA的上和下可能值减小。 With the measurement time and the time gap between the current becomes larger, due to the reduction of confidence in measurements, up and down the WA may decrease. 在时间To之后,A不再适用,并且总成本与指派成本Cf相同。 After time To, A is no longer applicable, and the total cost and the cost of assigning the same Cf. A对总成本的影响如何改变的说明仅是一个示例,并且实际实现取决于网络。 A on how to change the total cost of the description is only an example, and the actual implementation depends on the network. 总成本的另一个示例可W包括(i)静态/指派成本乘W (1 - f)与(ii)动态成本乘Wf之和,其中可信度函数f在0与1之间改变。 Another example of the total cost at the W include (i) Static / Cost assigned multiplying W (1 - f) and (ii) dynamic cost of multiplication and Wf, wherein the function f confidence between 0 and 1 change. 在此情况中,如果f =0,德尔塔(delta)不可信,则总成本等于指派成本,而如果f = 1,德尔塔充分可信,则总成本等于动态成本。 In this case, if f = 0, delta (delta) can not be trusted, then the total cost is equal to the cost assignment, and if f = 1, Delta fully credible, then the total cost is equal to the cost of moving.

[0064]因此,真实网络10的两个边缘节点i和j之间的总成本Cu-般可W基于上文定义的个体量来表达,并且具有如下表达式: [0064] Thus, the total cost of two real network edge node 10 i and j at the W Cu- as an individual based on the amount defined above to express, and has the following expression:

[00 化] [Of 00]

Figure CN102640472BD00121

[0066] 其中A是指派成本,A是通过归一化函数从状态矩阵获得的动态成本,W及f是可信度函数。 [0066] wherein A is to assign costs, A dynamic costs through normalization function obtained from the state matrix, W, and f is a function of confidence. 运算符0指示可W使用的各种数学运算。 Operators 0 W indicates various mathematical operations can be used. 上文已论述Cu的特定示例。 The foregoing has discussed a specific example of Cu. 但是,正如本领域技术人员会认识到的,可W使用其它函数。 However, as those skilled in the art will appreciate, W can use other functions.

[0067] 根据一示范实施例,关于图13论述用于选择资源的过程。 [0067] According to an exemplary embodiment of the process, discussed in relation to Figure 13 for selecting resources. 在图13中,用户130在步骤150中向CDN 132请求资产XdCDN 132可W包括重定向引擎134、成本估计器单元136、虚拟拓扑单元138、状态矩阵单元140等。 In Figure 13, the user request 130 to the CDN 132 in step 150 include asset XdCDN 132 at the W redirection engine 134, the cost estimation unit 136, a virtual topology unit 138, state matrix unit 140 and the like. 可W用软件、硬件或其组合来实现运些单元。 W can be achieved shipped more units in software, hardware or a combination thereof. 在一个应用中,可W在单个节点中的处理器中实现所有运些单元。 In one application, the W in a single node processor to achieve all these transport units. 在另一个应用中,运些单元可W分布在CDN的各种节点,各种处理器中。 In another application, these units can be transported in a variety of distribution node CDN W, a variety of processors.

[0068] 重定向引擎134在步骤152中确定CDN 132中是否存在资产X,并且确定是否多个缓存中存在该资产。 [0068] redirection engine 134 determines whether there are assets X CDN 132 in step 152, and determines whether the asset exists in more than one cache. 在步骤154中,重定向引擎134查询成本估计器136对于用户130哪个是最佳缓存。 In step 154, the re-estimated cost of the query engine 134 136 130 which is the best for the user cache. 成本估计器136在步骤156中构建至具有资产X的缓存的路径,并将此信息提供到虚拟拓扑单元138。 Cost estimator 136 to build path has assets X cached in step 156, and provides this information to the virtual topology unit 138. 虚拟拓扑单元138将真实网络转换成分级虚拟网络,正如上文论述的。 Virtual real network topology unit 138 will convert a hierarchical virtual network, as discussed above. 注意,此步骤是可选的,并且CDN网络可W无需虚拟拓扑单元138而执行最佳缓存的评估。 Note that this step is optional and may be W CDN network without a virtual topology unit 138 performs optimal cache evaluation. 在步骤160中将来自虚拟拓扑单元138的信息传递到状态矩阵单元140,状态矩阵单元140还从网络的测量系统接收测量数据。 Virtual topology information from a unit 138 in step 160 to the state in the matrix unit 140, unit 140 also receives state matrix from the measurement data of the measuring system network. 基于测量数据和虚拟网络,状态矩阵单元140在步骤162中更新矩阵值,并在步骤164中将此数据提供到成本估计器单元136。 162 update a matrix of values based on the measured data and virtual networks, state matrix unit in step 140, and provided to a cost estimator unit 136 in step 164 in this data. 成本估计器单元136还在步骤166中接收虚拟网络数据,并在步骤168中计算来自用户130连接到的边缘节点和具有资产X的边缘节点的每个路径的总成本。 Cost estimator unit 136 is also received in step 166 the virtual network data, and calculate the total cost of each path from the user edge node 130 is connected to the edge node and having the property of X in step 168.

[0069] 基于总成本对具有资产X的边缘节点评级,并在步骤170中将最佳缓存信息(例如, 具有最低成本的缓存)提供到重定向引擎134。 [0069] Based on the total cost of the edge nodes having a rating of asset X, and in step 170 the best information in the cache (e.g., cache with the lowest cost) to the redirection engine 134. 基于此数据,在步骤172中,重定向引擎134响应用户130的查询向该用户提供最佳缓存的地址。 Based on this data, in step 172, redirection engine 134 in response to a user's query to the user 130 with the best cache address. 用户130然后从最佳边缘节点提取资产X 或由CDN 132命令用户130连接到的边缘节点从最佳边缘节点提取资产X并将该资产X提供到用户130。 User 130 is then extracted from the best assets X edge node from the CDN 132 or 130 instructs the user to connect to the edge node to extract the best from the edge node and the property assets X X 130 is provided to the user.

[0070] 根据图14中所示的一示范实施例,可W在CDN 132中实现决策引擎142 W便基于(i)动态成本144和Qi )静态(指派)成本146来决定内容提取和各种缓存之间的资产迁移。 [0070] According to an exemplary embodiment shown in FIG. 14 cases, CDN 132 W in the implementation decisions will be based on engine 142 W (i) 144 and dynamic cost Qi) Static (assigned) to determine the cost of 146 and a variety of content extraction migration between assets cache.

[0071] 根据图15中所示的一示范实施例,有一种方法用于选择在通信网络中连接的多个缓存可用的资源。 [0071] FIG. 15 according to an exemplary embodiment shown, there is a method for selecting a plurality of buffers in a communications network connection available resources. 该方法包括从用户接收对资源的请求的步骤1500、标识多个缓存中存储该资源的一个或多个缓存的步骤1502、确定用户与存储该资源的一个或多个缓存中的每个缓存之间的路径所关联的成本的步骤1504(其中该成本包括反映通信网络的拓扑的静态成本和指示通信网络的变化状况的动态成本)、W及基于至一个缓存的最低成本从存储该资源的一个或多个缓存选择该一个缓存的步骤1506。 The method includes the step of receiving a request for resources from the user's 1500, identifying a plurality of cache memory cache one or more steps of the resource 1502, to determine the user with one or more of the resource cache memory in each cache of path between the associated cost of step 1504 (wherein the cost comprises a change situation reflects the communications network topology static cost and instructs the communication network of the dynamic cost), W, and the lowest cost on to a cache from storing the resource a Select one or more steps of the cache a cache 1506.

[0072] 较之如仅使用指派成本的传统方法的情况,一个或多个论述的示范实施例可W为内容迁移和检索提供精确且动态的成本估计。 [0072] As compared with the case of using only traditional methods to assign the cost of one or more exemplary embodiments may be discussed W content migration and retrieval to provide accurate cost estimates and dynamic. 再有,公开一种确定网络的测量信息的可信度的新方法、一种用于通过将基于例如时间的可能改变纳入考虑来表示和汇总分布式系统中的网络测量数据的新方法。 Further, disclosed a new method of measuring the reliability of the network information is determined, for example, by the time-based change may be taken into account and represent a new method in a distributed system network measurement data summary. 该新方法提供网络状况的动态画面W便在更新的网络状态信息对于整体性能有益的如CDN的数据交换网络中使用。 The new method provides a dynamic picture of network conditions W will be used to update the network status information as useful for the overall performance of the CDN data exchange network. 一个或多个实施例能够用作用于将动态网络状况纳入考虑来实行运营商策略的输入,并且可W使得如BART的测量工具能够被利用来为改善分布式系统中的性能提供建议。 One or more embodiments can be used as a dynamic network conditions into account to implement input operator policies, and may make such BART W measurement tools can be utilized to provide recommendations for improving the performance of distributed systems.

[0073] 图16中示出根据示范实施例的能够实行操作的代表性结构的示例。 [0073] FIG. 16 illustrates an example of the operation of the representative structure can be implemented in accordance with an exemplary embodiment. 可W使用硬件、固件、软件或其组合来执行本文描述的各种步骤和操作。 W can use the hardware, firmware, software, or combination described herein to perform various steps and operations. 适于执行示范实施例中描述的活动的示范装置1600可W包括服务器1601,服务器1601可W包括图13中所示的单元134、 136、138和140中的一个或多个单元。 Adapted to perform the exemplary embodiments described demonstration activities 1600 W may include server 1601, server 1601 may include W unit shown in FIG. 13 134, 136, 138 and 140 in one or more units. 运种服务器1601可W包括禪合到随机存取存储器(RAM)1604和只读存储器(R0M)1606的中央处理器(CPU)1602dR0M 1606还可W是用于存储程序的其它类型的存储媒体,如可编程ROM(PROM)、可擦PROM(EPROM)等。 Species transport server 1601 may include Zen W bonded to a random access memory (RAM) 1604 and read-only memory (R0M) 1606 central processing unit (CPU) 1602dR0M 1606 W may also be other types of storage media for storing the program, such as programmable ROM (PROM), erasable PROM (EPROM) and the like. 处理器1602可W通过输入/输出(I/O)电路系统1608和总线连接(bussing)1610与其它内部和外部组件通信, W提供控制信号W及诸如此类。 Processors 1602 and 1608 can be W circuit system bus connection (bussing) 1610 with other internal and external components communicate, W provides a control signal W and the like through the input / output (I / O). 处理器1602按软件和/或固件指令所指示的,实行如本领域中已知的各种各样的功能。 Processor 1602 according to the software and / or firmware instructions indicated, as known in the art to implement various functions.

[0074] 服务器1601还可W包括一个或多个数据存储装置,包括硬盘驱动器和软盘驱动器1612XD-R0M驱动器1614和能够读取和/或存储如DVD等的信息的其它硬件。 [0074] W server 1601 may also include one or more data storage devices, including hard disk drives and floppy disk drives 1612XD-R0M drive 1614 and the ability to read and / or storage of other hardware, such as DVD and other information. 在一个实施例中,可W将用于实行上文论述的步骤的软件存储和分布在CD-ROM 1616、软磁盘1618或能够便携地存储信息的其它形式的媒体上。 In one embodiment, W can be used to store the software implementation of the steps discussed above and distributed on CD-ROM 1616, flexible disk 1618, or other form capable of portably storing information medium. 运些存储媒体可W插入到如CD-ROM驱动器1614、盘驱动器1612等的装置中并被其读取。 W shipped some storage medium may be inserted into the CD-ROM drive, such as 1614, 1612 and so the disk drive apparatus and read it. 服务器1601可W禪合到显示器1620,显示器1620可W 是任何类型的已知显示器或呈示屏幕,如LCD显示器、等离子显示器、阴极射线管(CRT)等。 Server 1601 can be combined to display Zen W 1620 W monitor 1620 may be any type of known display or render screen, such as LCD displays, plasma displays, cathode ray tube (CRT) and the like. 提供一种用户输入接口1622,其包括如鼠标、键盘、麦克风(mi crophone )、触控板、触摸屏、 语音识别系统等的一个或多个用户接口机制。 Providing a user input interface 1622, including as a mouse, a keyboard, a microphone (mi crophone), touch pad, touch screen, voice recognition system, one or more user interface mechanisms.

[0075] 服务器1601可W经由网络禪合到其它计算装置,如陆上通信线路(Iandline)和/ 或无线终端和关联的观看器(watcher)应用。 [0075] W server 1601 can be bonded to other computing devices via a network of Zen, such as landline (Iandline) and / or wireless terminals and associated viewer (watcher) applications. 该服务器可W是如全球区域网络(GAN)(例如因特网1628)中的更大网络配置的一部分,其允许最终连接到各种陆上通信线路和/或移动客户端/观看器装置。 The server can be as W Global Area Network (GAN) (such as the Internet 1628) a part of a larger network configuration, which allows the final connection to the various landline and / or mobile client / viewer apparatus.

[0076] 公开的示范实施例提供用于确定资源的节点、方法和计算机程序产品。 [0076] disclosed exemplary embodiments provide for determining node resources, methods and computer program products. 应该理解本文描述不打算限制本发明。 It should be understood described herein are not intended to limit the present invention. 相反,示范实施例打算涵盖在所附权利要求书定义的本发明的精神和范围中包括的备选、修改和等效。 On the contrary, the exemplary embodiment is intended to cover within the spirit and scope of the appended claims define the invention includes alternatives, modifications and equivalents. 另外,在示范实施例的详细描述中,阐述了许多特定细节W便提供对要求保护的本发明的全面理解。 Further, in the detailed description of exemplary embodiments, numerous specific details are set forth to provide a W then claimed a thorough understanding of the present invention. 但是,本领域技术人员会理解,没有运类特定细节仍可W实施各种实施例。 However, those skilled in the art will appreciate that specific details are still not running class W implement various embodiments.

[0077] 正如本领域技术人员还将领会的,可W在计算机程序产品中或作为方法在电信网络、无线通信装置中实施示范实施例。 [0077] As those skilled in the art will appreciate, you can W in a computer program product or as a method in a telecommunications network, a wireless communication device implemented exemplary embodiment. 相应地,示范实施例可W采用完全硬件实施例的形式或组合硬件和软件方面的实施例的形式。 Accordingly, the exemplary embodiments may employ W entirely hardware embodiment or a combination of hardware and software aspects of the embodiments. 另外,示范实施例可W采取计算机可读存储介质上存储的计算机程序产品的形式,该计算机可读存储介质具有实施在该介质中的计算机可读指令。 In addition, the exemplary embodiments may take the form of W stored on a computer readable storage medium of a computer program product, the computer-readable storage medium having computer-implemented in the medium-readable instructions. 可W利用任何适合的计算机可读介质,包括硬盘、CD-ROM、数字多用光盘(DVD)、光存储装置或如软盘或磁带的磁存储装置。 At the W using any suitable computer-readable media, including hard disks, CD-ROM, digital versatile disc (DVD), optical storage devices, or magnetic storage devices such as a floppy disk or magnetic tape. 计算机可读媒体的其它非限制性示例包括闪存型存储器或其它已知的存储器。 Other non-limiting examples of computer readable media include flash-type memories or other known memories.

[0078] 虽然在实施例中W特殊组合描述了本发明示范实施例的特性和单元,但是每个特性或单元能够在没有实施例的其它特性和单元的情况下单独地使用或W具有或不具有本文公开的其它特性和单元的各种组合来使用。 [0078] Although in the embodiment described in W special combination of features and an exemplary embodiment of the present invention is an example of unit, but each feature or element can be used alone or with a W in the absence of other features and elements of the embodiments or without other features and with various combinations of elements disclosed herein for use. 本申请中提供的方法或流程图可W在计算机可读存储介质中有形地实施的计算机程序、软件或固件中实现W由专口编程的计算机或处理器来运行。 The implementation method of the present application provides a flowchart of a computer program or a computer can be W-readable storage medium tangibly implemented in software or firmware run by W E I programmed computer or processor.

Patent Citations
Cited PatentFiling datePublication dateApplicantTitle
*US2005188073 Title not available
*US2009172167 Title not available
WO2002025463A1 *19 Sep 200128 Mar 2002Conxion CorporationMethod and apparatus for dynamic determination of optimum connection of a client to content servers
Classifications
International ClassificationH04L12/70, H04L29/08
Cooperative ClassificationH04L67/1025, H04L67/1097, H04L45/124, H04L67/1002, H04L67/101, H04L45/00, H04L67/2852, H04L45/125, H04L45/121
Legal Events
DateCodeEventDescription
15 Aug 2012C06Publication
26 Dec 2012C10Entry into substantive examination
14 Sep 2016C14Grant of patent or utility model