KR20070003920A - How to Schedule Distribution of Cache Server Network and Content Files - Google Patents

Abstract

Techniques for scheduling the distribution of content files within a content delivery network and a content delivery network adapted to carry out the technology are disclosed. The technique includes scheduling distribution of content files based on delivery location, service time of content request, and cache server layer. The multicasting tree that delivers each content file is dynamically established in the content delivery network taking into account location and service time.

Description

How to schedule distribution of cache server network and content files {CACHE SERVER NETWORK AND METHOD OF SCHEDULING THE DISTRIBUTION OF CONTENT FILES}

FIELD OF THE INVENTION The present invention generally relates to the field of data communications and cache server networks, and more particularly, to systems and methods for scheduling multicasting distribution of content within a content delivery network.

For large amounts of content, such as movies, content clients generally tolerate some delay for better quality. The client can watch the downloaded video in high quality at a planned time in the future, rather than immediately watching the low quality streaming video. For example, a mobile user may pre-order a video while he is in a cellular mobile network, and download the ordered video while that user is in a hotspot wireless LAN. As such, the mobile user can enjoy high quality content at low cost.

Recently, content delivery network (CDN) technology has been spread to the Internet and used to improve download of web pages. The CDN consists of cache servers at different geographical locations, i.e. network nodes having storage and forwarding capabilities. The basic premise of CDN technology is that the link between cache server and client has low cost and high bandwidth. When a client requests a content file, the download will be faster if the content file is stored in a cache near the cache server. If not, the client may experience a longer delay. Therefore, it is desirable for the client to download the content file from the nearest server. The technique of finding a cache server close to a client is called request-routing. There is a process for redirecting content requests to a closer cache server. For example, modifying a URL from the original URL to any one URL is predetermined by the cache server. In another application, by extending request-routing with content timing, the request is redirected to a closer cache server based on the future availability of the requested content on the cache server.

Typically, a client can tolerate a delay for a large amount of content file up to the expected service time that it specifies by the time it wants to retrieve the content file. Thus, even if the requested content file is not currently stored in the cache server close to the client, the client will not experience a delay if the download system can transfer the content file to the cache server before the expected service time. It is an industry goal to reduce this delay by properly scheduling the download of content files requested for client retrieval to the appropriate cache server.

Multicasting content delivery can be requested from different cache servers. Prior to content download, due to the availability of enhanced content request information, it is possible to optimize content distribution in the CDN through multicasting techniques. Typically, the download service requires a CDN to distribute the content file to the cache server closest to where the client request for the content file comes. The content file must be stored on its cache server and ready to be downloaded to the client at a time no later than the client's expected service time. As such, a need exists for an improved system and method for scheduling distribution of content files to cache servers associated with requests for content files.

Briefly, the present invention relates to a method for scheduling distribution of content files within a cached network environment. These methods include: receiving a request for content to be delivered at service time, associating a content file with a particular cache server, dynamically establishing a multicasting tree of the cache server, and requesting the content of the cache server. Passing from the multicasting tree at service time.

1 is a schematic diagram of a content delivery network according to one embodiment of the invention.

FIG. 2 is a schematic diagram of the content delivery network of FIG. 1 showing the download and associated structure resulting from multiple user requests for content files in accordance with an embodiment of the present invention in view of the proximity of a cache server.

3 is a flow chart of a method for scheduling distribution and download of content files in a content delivery network according to an embodiment of the present invention in consideration of proximity of a cache server.

Multicasting distribution can be implemented at the transport layer or at the application layer. Since there are many defects associated with transport layer multicasting, the present invention only considers application layer multicasting. Transport layer multicasting requires a multicasting enabled transport network. The Internet generally does not have such a transport network. In addition, even if a multicasting enabled transport network is available, it must be transmitted simultaneously on all branches of the multicasting tree. Because of this, it is not possible for any network node (ie cache server) on the multicasting tree to have transfer or cache capability in any period of the multicasting session. However, application layer multicasting can be more flexible in the transmission schedule from node to node on the multicasting tree. For download services with multiple download requests of different expected service times, application layer multicasting may be more appropriate. As used herein, application layer multicasting is defined as a store / forward action at each network node on the multicasting tree. The store includes caching on intermediate nodes and forward means transmission to multiple ports at the same or different time.

Referring now to FIG. 1, one embodiment of a content delivery system 100 of the present invention is shown. System 100 comprises a content server S and a CDN comprising a cache server A, a cache server B and a cache server C. Although a particular architecture for a CDN with only three cache servers is shown, the invention is not limited to any particular network architecture or configuration. Regardless of the structure of the (flat or hierarchical) CDN network, a multicasting tree can be established for content file distribution according to the present invention. Unless requests for the same content file have the same expected service time, the download and associated structure (i.e., the multicasting tree structure) will depend on the gap between the different expected service times as well as the distance between the source and destination.

Referring now to FIG. 2, there is shown a content delivery system 100 that connects the components shown by the data flow arrows, and FIG. 3 is an embodiment of the present invention, in which step 300, a client / user A1, B1 and C1) each make a request for the same content file. Each request for a content file has an estimated service time. As used herein, the service time of a request is the time the client requests that the content file is available for download to the client's electronic device, i.e., a computer, a mobile phone, or a personal digital assistant. The service time is specified by the clients A1, B1 and C1 when the request is made. In this example, the service times for requests generated by clients A1, B1 and C1 are 7 PM, 5 PM and 8 PM, respectively. Although the service time has a chronological order of B1-A1-C1, the request is not limited to the order generated by the client and / or received by the system 100.

The request generated by the clients A1, B1 and C1 (referred to as requests A1, B1 and C1 for simplicity of understanding) relates to each of the cache servers A, B and C, and step 310 is completed. do. The relationship between requests A1, B1 and C1 and cache servers A, B and C is designated as lines 1, 2 and 3 in FIG. Relationships are determined by the CDN through a static hierarchical structure or a dynamic request-routing process. The choice of which cache server is associated with each request is determined by the cache server's proximity to the client. The request is preferably associated with a cache server closest to the client making the request for the content file. As used herein, the determination is made using physical locality and / or network locality, taking into account bandwidth cost and / or congestion. The technique used to perform such a relationship is known as a request-route technique. Determination of which cache server is associated with the request can be statically, i.e., a pre-established hierarchical structure at multiple proxy servers, or can be learned dynamically. Alternatively, the client may specify a particular cache server to which the client request relates, such as the remote site download function at the hotspot.

If the relationship of the request to the cache server is dynamically determined by the request-routing technique, an extended request-routing technique should be used. In this case, even if the requested content file is not currently available on the cache server, the request-pathing can still relate the request to the cache server because the relationship means delivering the content file to the cache server in the future. have.

Requests A1, B1 and C1 are sent to content server S in the order B1, C1 and A1. The multicasting tree will initially have only one node, the content server (S). Since the request B1 is the first request sent to the content server S, step 320 is executed first with respect to the cache server B. In step 320, it is determined whether cache server B is on the multicasting tree. If the answer is NO (in this case), the system adds node B to the multicasting tree and proceeds to step 330. In step 330, the system checks for the presence of the nearest upstream cache server and finds the upstream cache server C in this case. This is done through static hierarchy or request-pathing. Request routing is shown. Request B1 is then associated with cache server c and completes step 340. The relationship of the request B1 to the cache server C is shown by line 4 in FIG.

Step 320 is then executed for the cache server C. According to step 320, it is determined whether cache server C is on the multicasting tree. If the answer is NO in this case, then the system adds node C to the multicasting tree. The cache server C then finds the closest upstream node that is the content server S and completes step 330. Next, request C1 is associated with content server S at step 340, which is shown by line 5 in FIG.

Next, step 320 is executed for the content server S. FIG. According to step 320, since the content server S is on the multicasting tree, the answer is YES and proceeds to step 350. Since the current server is a content server, the answer at step 350 is NO and then proceeds to process the request.

Considering request C1, request C1 is now generated (following request B1) at step 300 and associated with cache server C at step 310. Since node C has already been added to the multicasting tree when executing the process for request B1, the answer at step 320 is YES and the process proceeds to step 350. Since the service time of C1 (8 PM) is later than the service time of 5 (5 PM), the answer at step 350 is NO. The process then restarts and processes the next request.

Referring now to the request A1 received following request C1, request A1 is generated at step 300 and associated with cache server A at step 310. According to step 320, it is determined whether cache server A is on the multicasting tree. In this case, the answer is NO and the process continues at step 330. At this time, node A is preferentially added to the multicasting tree, and cache server A finds upstream cache server B, and completes step 330. Next, in step 340, request A1 relates to cache server B. This is illustrated by line 6 in FIG. 2. Since node B is already on the multicasting tree and the service time of A1 is later than the service time on node B, the answer at step 350 is NO. The process then restarts and processes the next request.

The algorithm used to determine the distance between cache servers is based on geographic distance as well as other factors such as cache capacity, load balancing of network links, and the like. For example, because the cost of caching content at node B from 5 PM to 7 PM may be greater than the cost difference between link 7 and link 6, node A is node C, which is its upstream node. Find).

While the invention has been described and illustrated in sufficient detail to enable those skilled in the art to readily construct and use the invention, it will be apparent that various substitutions, modifications, and improvements may be made without departing from the spirit and scope of the invention.

Claims (17)

A method of processing a request for a content file from a content delivery network system, the method comprising: Receiving a request for content to be delivered at service time, Associating the content file with a particular cache server; Dynamically establishing a multicasting tree of cache servers, and Delivering the requested content from the multicasting tree of cache servers at the service time How to include. The method of claim 1, The associating further comprises associating the request with the nearest cache server. The method of claim 1, The dynamically establishing further includes adding a cache server associated with the request if the cache server is not yet associated with the multicasting tree. The method of claim 1, The associating further comprises associating the request with the nearest server if the request has a faster service time than a previous request. A method of processing a request for a content file from a content delivery network system. (a) receiving a first request for a content file having a first service time, (b) associating the first request with a cache server for retrieval, (c) determining whether the relevant cache server is on a multicasting tree originating in a content server that is the origin of the content file, (d) if it is determined that the relevant cache server is not on the multicasting tree, add the relevant cache server to the multicasting tree, find an upstream cache server destined for the content server, and find the found upstream cache server and the Associating a first request such that the upstream cache server becomes the associated cache server and repeating step (c) until reaching the content server and the first request is associated with the content server. 1 the request relates to a content server and processes the next request for the content file starting at step (a)-, (e) if it is determined that the relevant cache server is on the multicasting tree, determining if the first service time is earlier than all service times of a request for a content file that already exists on the related cache server, (f) if it is determined that the first service time is not earlier than all other service times of requests already present on the associated cache server, the first request is associated with the cache server and the content file begins at step (a). Processing the next request for, and (g) if it is determined that the first service time is earlier than all other service times of a request already present on the associated cache server, in association with a cache server determined to be an upstream cache server directed to a content server on the multicasting tree, Causing this cache server to be the associated cache server and returning to step (c) until the first request is associated with the content server, if the first request is associated with the content server, step (a) Processed the next request for a content file, starting at- How to include. The method of claim 5, Finding the upstream cache server comprises finding the nearest upstream cache server using a request routing process. The method of claim 6, The proximity is determined using at least one factor selected from the group consisting of geographic distance, cache occupancy, and load balance of the network link. The method of claim 5, The step of finding an upstream cache server is a method of finding an upstream cache server using a hierarchical relationship. A content delivery network system comprising a content server and a CDN network and processing a request for a content file, (a) receive a first request for a content file from a client, (b) associate the first request with a cache server for retrieval, (c) determine whether the associated cache server is on a multicasting tree, and At least one cache server adapted to relate the cache server to a multicasting tree, if not Content delivery network system comprising a. The method of claim 9, Determine if the first service time is earlier than all other service times of requests already present on the related cache server on the multicasting tree, and if not fast, associate the first request with a cache server on the tree, and If so, locate an upstream cache server, and wherein the first service time is earlier than all other service times of a request already present on the associated cache server on the multicasting tree or until the first request is associated with a content server. And a means for associating the 1 request to the upstream cache server. The method of claim 9, Means for associating the first request with a cache server determined to be on the multicasting tree if a first service time is not earlier than all other service times of a request already present on the associated cache server on the tree. Network system. The method of claim 9, Means for finding the nearest upstream cache server by request routing. The method of claim 12, Means for finding the nearest upstream cache server using at least one factor selected from the group consisting of geographic distance, cache occupancy, and load balance of the network link. The method of claim 9, And a means for locating an upstream cache server using a hierarchical relationship. The method of claim 9, And further comprising a content delivery network broker adapted to provide information for request routing so that the result is used by the content server, the information being requested on one or more cache servers in the content delivery network. A content delivery network system that relates to the availability of a file. The method of claim 9, And further comprising a content delivery network broker adapted to provide information for request routing so that the results are used by the content server, the information using the requested content file on one or more cache servers in the content delivery network. A content delivery network system that relates to the possibility or to schedule future availability of a content file at one or more cache servers (13). The method of claim 9, Further comprising one or more cache servers and a content delivery network broker, further comprising means for the broker, cache server, and / or a content server for determining a future time period and a server from which a client can request the file. Content Delivery Network System.

Images