SI20229A - Internet caching system - Google Patents

Abstract

The present invention relates to a method, a system and a server for caching Internet information content. According to the invention, there is provided a set of geographically distributed cache servers generally serving different geographical regions, wherein Internet information derived in relation to the operation of one of the said cache servers is distributed to essentially all of the said servers.

Description

INTERNET CEMETERY SYSTEM

FIELD OF THE INVENTION

The object of the present invention is a caching method, a system, and a server designed to cache online information content.

BACKGROUND OF THE INVENTION

Over the past few years, the Internet has evolved into the fastest growing medium of communication and is expected to be the main information dissemination and dissemination system in the future. The internet is gaining popularity and the number of its users is growing at an incredible rate. Anyone can put information on the web in any format - text, images, audio or video - and then make it accessible to users anywhere in the world. This is due to the incredible expanse of the Internet and its nowadays most used component of the World Wide Web. with the abbreviation: WWW.

However, the mass popularity of the Internet, and especially the increasing use of its aforementioned application, the World Wide Web, severely pressures on the performance of the Internet itself. Traffic on the internet doubles every three months and there is nothing to indicate that this pace is slowing down. This kind of exponential growth will be introduced at some point, but several factors suggest that this point is still a long way off.

Today, only a minority of PCs have access to the Internet. However, they predict that most people who either own or purchase a PC will soon have access to the Internet. In addition, the speed of data exchange between the user and the Internet infrastructure is increasing rapidly. New technologies such as ISDN, cable TV modems and xDSL are emerging. Manufacturers of home electronics are already sending cheap devices to the market that allow access to the World Wide Web on the Internet with a home television set.

These and other factors are making access to online information increasingly difficult as the system is generally congested. In many places around the world, one of the major problems is insufficient bandwidth. However, the flow of information between the US and Europe is limited by the permeability of the transatlantic cable. Its throughput is simply too small to cope with the explosion of internet use.

There are basically two known solutions to this problem. The first solution is to increase bandwidth and switching capacities, which is already being done, but this solution is hampered by huge costs and significant technological barriers.

Another solution is to use so-called caching techniques. Caching basically means keeping track of traffic on the Internet and storing copies of commonly used World Wide Web files somewhere closer to the user than keeping the original files. This means that we keep a local copy of the particular site we are interested in. For example, a copy of a CNN homepage, the original of which is located in the United States, can be temporarily stored in the cache of a European "proxy" server, so that European Internet users do not have access to the CNN homepage without having to communicating via transatlantic cable, on the one hand, achieving shorter access times and, on the other, reducing network load. The total amount of information on the World Wide Web is virtually immeasurable, but the bulk of traffic actually generates a relatively small amount of this information. Around 1020 gigabytes of large cache reduces traffic in a given community by 30-50 feet depending on the size and homogeneity of the community itself (data for spring 1997).

Despite today's caching techniques, the far too low bandwidth of available communication systems in many corners of the globe, such as in southern and eastern Europe, southern America, India and east Asia, simply does not allow users to use the internet the way they want.

Summary of the Invention

The object of the invention is to reduce the performance constraints of the Internet, which is achieved by caching, which is solved in such a way as to reduce the time of actual presence on the Internet and at the same time allow significantly greater expansion of both the number of users and the flow of information on the Internet, especially in areas where the bandwidth is width low.

According to the present invention, said purpose is achieved by a caching method, system, and server for caching the contents of the Internet information according to the attached claims.

The invention is based on the recognition that in many languages and culturally homogeneous geographical areas, neither in size nor in form, does it coincide with the area that could be economically and technically optimally covered by a single cache server. As a result, such an area covers multiple geographically distributed servers. In addition, the invention of servers according to the invention, geographic areas, is also based on the realization that users in such a linguistically and culturally homogeneous geographical area can also be assigned a homogeneous selection of the websites visited, and it is therefore appropriate that the information stored by one user upon request these cache servers are also stored by other cache servers in the said area, based on the principle that if a particular user is interested in a given information, that information is likely to be of interest to other people in the same linguistically and culturally homogeneous geographical area. The larger the end-user community, the greater the likelihood that another file member will want to view the file.

A series of geographically distributed caches that serve different connections, however, have a common feature, such as serving a cultural or therefore, it is updated with essentially the same information each time one of the above servers transmits the information from elsewhere because it has been requested by an end user, etc.

Such cultural and linguistic homogeneity can be defined by the national border, the border of a particular language, the border of a religion, the border of a technological or developmental level, the border of an economic region, as well as their various combinations, etc.

Therefore, according to one embodiment of the invention, said set of geographically located cache servers is deployed within a linguistically and culturally defined global geographical area or geographically defined community. In addition, each of the above cache servers is preferably arranged to serve its local region within the specified geographical area or community.

Thus, e.g. assume that German-speaking parts of Germany, Austria, Switzerland and Italy form a culturally and linguistically homogeneous global area. Therefore, a set of cache servers operating according to the invention is deployed within this range. This set of servers located within the global area defined above may be arranged to say, for example, one (or more) serving the first region comprising the northern part of Germany, the other (or more) serving the second region comprising the eastern portion Germany, a third (or more) serves a third region comprising central Germany, a fourth (or more) serves a fourth region comprising Switzerland and northern Italy, and finally a fifth (or more) serves a fifth region comprising Austria.

Importantly, the defined geographical area of the invention does not necessarily mean a homogeneous, continuous area surrounded by a boundary, but may also represent a group of geographically distinct territories that together form the area of the invention.

According to estimates made by the inventor himself, in some places around the world, e.g. in India and South America, the bandwidth of a telecommunications network is so small that it is actually more efficient to distribute a certain packet of information to all cache servers within a defined global area, based on a single request received by the network from a single user. for another user to search the same packet of information next time around the network. However, to decide when to implement distributed caching according to the invention and when not, it is possible to set the rules in different ways, depending on the particular environment of use, as will be explained below.

The dissemination of information to a set of cache servers according to the invention can take many different forms. From one point of view, we could only use the internet for this purpose. However, such a solution would further burden the network's throughput capacity. Therefore, the information according to one of the preferred embodiments of the invention is disseminated to said set of cache servers using multicast communication techniques, preferably via dedicated data channels separate from internet communication as such.

According to one of the preferred embodiments, said information is disseminated to said set of cache servers by satellite. In this case, each of the servers preferably has access to an outbound connection to send said information, and to an incoming connection that receives information transmitted by any of the servers. The hardware and software needed to manage this kind of distribution may reside on the cache server itself, for example. with your ISP or satellite communications provider. Similarly, software for controlling information flow and for decision making within the system can be installed on the cache servers themselves, but it can be installed with Internet service providers or in any other suitable location, as is clearly understood by one of ordinary skill in the art. If we use a satellite system, we can also use the satellite link to transmit or transmit claims to other ends of the globe, say from Europe to the US and vice versa.

In an alternative embodiment, all cache servers in the system are connected via a data channel to a central control unit. Such a central control unit may e.g. makes decisions whether or not to store certain information in the cache, how to distribute it, etc.

Depending on the size of the community, the capacity of the cache servers, the available network communication capabilities, the purposes of the cache provider, etc. the distributed caching technique of the invention may be adapted to be distributed or distributed. does not distribute information under different rules. According to a preferred embodiment of the invention, only the requests or frames relating to a particular communication format or application corresponding to the type of information services provided by the local information storage medium are stored in the cache. One such example is that caches only store i.e. TCP queries addressed to WWW ports. Another example is that the cache server provider only stores some of the incoming addresses (such as .com or .org addresses only), or only if so many requests have been made to the system (say, after only three separate requests), etc.

The processing resources required for this kind of decision-making can be installed at each cache server, can be installed at a central control unit as mentioned above, at an internet service provider, or at some similar location connected to cache servers, etc.

Also, a claim concerning an information provider located in a region served by one of the cache servers of said set of cache servers may be processed in various ways in accordance with various embodiments of the invention, which will be discussed below.

After the first two embodiments of the invention, it is assumed that the request for local information provided by the end user to an online content provider located in the same region as the user itself is of local importance only. Therefore: a) we will not store the information in cache at all, since it will always be easy for a local user to gain access to a local internet content provider; or b) store the information only on the local cache server and do not distribute the information to the entire set of geographically located cache servers, since the request is likely to be of local importance only. In this context, therefore, the term "local" denotes the common property of those objects located within the same region.

In the second embodiment, it is assumed that the information request made by the end user to a content provider located in the same region as the user will not store the cache server serving the specified region, but will send it to all other cache servers within system, as it will always be easy for a local user to gain access to a local internet content provider (so they do not need to be stored in local caches), while an out-of-region user will not have so easy access to a local internet content provider (so they need to be stored in caches located outside the region).

Upon further implementation, it is assumed that the capacity of the cache server is large enough to handle all requests, so there is no need to handle information regarding local content providers differently from any other information, and as a result, any information stored by a single cache server on the system will also be stored by all other cache servers on the system.

. After the second embodiment of the invention, a list of addresses of internet content providers that should not be stored in the cache is compiled based on the rules we have chosen to make decisions within the system. Then each cache is arranged to compare the incoming information requests with the specified list and to determine whether or not to store the individual request in the cache.

Depending on the intended purpose or. The usability of the system will be arranged by the cache servers to store the information for a certain period of time. Caches can e.g. they operate on a first-in-first-out basis (FIFO or "the sooner in, the sooner out"). In another embodiment, certain types of information may be provided with up-to-date conditions that determine the maximum time interval of use in which the copied information is still held. However, with respect to these and other aspects, the invention can be combined with different types of known caching methods, as is readily apparent to one skilled in the art.

'In addition, each of the above cache servers may consist of several lower level servers that are preferably connected via a high-speed switch to the cache manager that controls them. Furthermore, there are no restrictions on the number of cache servers that can be used. In addition, each server is made available to more than one ISP.

If one of the above cache servers fails, the system can quickly bypass it and initiate an emergency operation during the time that the cache server is repaired or replaced. At this time, the functionality of the cache may be provided by another server in another region within the system. The advantage of the invention is that in such a case, the replacement cache server makes available the same stored contents as the disconnected server, thereby ensuring that the cached information is still relevant to end users in the disconnected server region.

Although the invention has been described on the basis of an intercommunication system, it will be apparent to one of ordinary skill in the art that the invention can also be used in other similar systems of global exchange of information that may occur in the future and which will be subject to communication problems of a similar nature. Therefore, of course, the invention is by no means limited to use on the Internet.

Furthermore, although the invention has been described primarily on the basis of use on the World Wide Web, it is also clear that the invention can also be used in many other types of network and is not limited to the Internet.

Brief description of the drawings

Further aspects, features and advantages of the invention will be explained in the preferred embodiments by means of the accompanying drawings showing:

Figure 1 schematically shows an embodiment of an on-line cache system of the present invention;

Figure 2 schematically shows the arrangement of the cache server that is part of the system shown in FIG. 1;

Figure 3 schematically shows an alternatively arranged cache server;

Figure 4 schematically shows another alternatively cached server;

Figure 5 schematically shows a flowchart of the operations performed by the cache server in Figure 4;

system

Figure 6 schematically shows a flowchart of the operations performed by the interceptor in Figure 3;

Figure 7 schematically shows another embodiment of the internet cache of the present invention; and

Figure 8 schematically illustrates the internet cache of the present invention.

one system implementation after

Detailed description of preferred embodiments

Now with reference to FIG. 1 described the first embodiment of an internet cache system according to the invention. In FIG. 1 shows the preferred language or. a culturally defined geographical area 200 comprising four geographical regions A, B, C and D. Area 200 could, e.g. would represent Great Britain, while Regions A, B, C and D would represent Ireland; Wales; Scotland with northern England; and southern England.

In each of the individual regions A, B, C and D there is at least one HOa, IlOb, HOc and lOOd internet service provider providing their region with access to the Internet 100.

Each of the HOa, IlOb, HOc, and HOd internet service providers therefore provides a certain number of end users 120a, 120b, 120c, and 120d and a certain number of internet content providers 130a, 130b, 130c, and 130d access to the internet. Although in FIG. 1 end users 120 displayed separately from content providers 130, it is understood that the end user may also be a content provider.

Each service provider is connected to at least one internet cache server 140a, 140b, 140c and 140d. Each cache server 140 therefore provides caching to end users 120 within its region. Each of the cache servers 140a, 140b, 140c and 140d is also connected to a satellite dish 150a, 150b, 150c and 150d and can transmit information to all other servers within range 200 via satellite 160.

We will now describe an example of the operation of the system shown in FIG. 1. When e.g. HOa service provider in region A receives from the end-user 120 a request for information provided by content provider 130e located in region E, in which case it lies outside 200 (for example region E, for example, North America), HOa service provider checks , or the cache assigned to it 140a stores a contemporary copy of the information requested. If so, the requested information is retrieved from the cache server 140a and then passed to the end user. However, if the cache server 140a does not retain a copy of the requested information, the HOa service provider then transmits the request on the internet 100 to the original content provider 130e. When the requested information arrives on the Internet 100 from the content provider 130e, the HOa service provider transmits the information to the end user and updates the cache server 140a with it. The update process also includes the distribution of updated information via satellite dish 150e and satellite 160 to cache servers 140b, 140c, 140d in other regions B, C, D so that all servers are updated with the same information. After this update is performed, none of the users 120 in the region 200 need to visit the original content provider 130e then the Internet, but can use the copy stored in the associated cache server. If any of the cache servers are already full during the update, it will erase the information that has been stored on it for the longest time (that is, the longest time since it was last used by a user). Each server will also ensure that accurate and up-to-date information is stored on a single server by one of a number of methods, such that they represent reliable copies of the original websites.

In the following drawings, for ease of understanding, the elements will perform functions similar to those already described in FIG. 1, denoted by the same numbers.

Now, with the help of Figure 2, we will describe the implementation of the cache server, which is an integral part of FIG. 1 of the system shown. In FIG. 2, the three end user terminals or stations 120 are connected to a multiplex unit 170, which may be a modem system or a LAN router, depending on the type of user system. The multiplex unit 170 is connected to an interceptor 110 installed by an internet service provider that provides access to the internet 100.

The multiplex unit 170 multiplexes communications to and from user terminals 120.

According to this scheme, it is the user 120 who wants to obtain certain information, e.g. made available on the homepage or site by an internet content provider (not shown) located on the original publication site somewhere on the Internet 100, generally sends an information request, a specific HTTP request to the WWW port, via the multiplex unit 170 to the internet service provider 110, where the request is redirected to the Internet 100 and finally reaches the content provider. The content provider then responds to the request, that is, sends the requested information over the Internet 100 back to the user 120 then to the service provider 110.

However, according to the scheme shown in FIG. 2, user information request 120 is reviewed by the service provider interceptor 110. The interceptor then verifies that the requested information, such as information made available by that content provider, exists in the form of a copy, etc. in the local cache server 140, which is directly connected to the interceptor 110. If a copy, etc. The requested information is stored in the local cache server 140, the request is redirected to the server 140, which then transmits the requested information to the user. In this case, the information already exists in all the cache servers in the range 200 shown in FIG. 1, so it does not need to be distributed to other caches.

However, if the interceptor 110 determines that no copy, etc., exists in the local server 140. the requested information, the information request of user 120 is forwarded to the Internet 100 as in the normal use of Internet communication. The interceptor then follows the response of the network 100 to user 120, and when it receives the requested information, it also updates the cache server 140.

The caching of new or updated information in the cache server 140 of the invention is also followed by the distribution of new or updated information to other cache servers in the area 200 shown in FIG. 1. This distribution is triggered by either the cache server 140 or e.g. intercepts 110 at the service provider and reaches other cache servers in the area through an outbound satellite connection to satellite 160. Similarly, the cache server 140 receives such information via incoming satellite in the opposite case when receiving new or updated information from other cache servers in his area.

It should be noted that in this case, the interceptor 110 intercepts only communication packets of certain types. For example, it can intercept only online information requests from users 120. Other types of communication services, such as e-mail, etc., leave the interceptor 110 completely alone.

Now, based on Figure 3, we will describe an alternative arrangement of the cache server, similar to the one shown in FIG. 2. The arrangement of FIG. 3 is different from that in FIG. 2 in that the interceptor 110 is not only connected to one cache server 140 by the service provider, but also to the memory cache 112 with fast access. Compared to the cache server 140 of the invention, which provides e.g. with hundreds of gigabytes of storage, the memory cache with fast access in this case has a much smaller capacity, for example around one gigabyte. The memory cache 112 is thus used to store essentially all the information passing through the interceptor 110, except that each packet of information is subsequently stored for only a relatively short time.

In FIG. 3, a high-speed memory cache comprising memory areas 112 and an index 111, which states what information is currently stored in said memory areas, is used as the working memory of the interceptor 110. Thus, when it receives an information request from user 120, the interceptor first verify that the desired information is contained in its fast access memory cache 111, 112. Only if it does not find the requested information in the memory cache 112 will it look in the cache server 140. The procedure that follows will be the same as that described in FIG. . 2. But in FIG. 3, all information relating to all user information requests 120, both information received from the network 100 and that received from the cache server 140, is temporarily stored in the high-speed memory cache 112, while the cache server is updated only if when it comes to new information received from the Internet 100. Furthermore, memory cache can also be used to cache information other than that stored on server 140. In addition to storing information from the World Wide Web just like the cache server 140. , memory cache, e.g. it also stores information regarding other types of queries. Further, information stored in the memory cache 112 will not be automatically transmitted to other caches in the area 200 of Figure 1.

Now with reference to FIG. 4 describes another alternative arrangement of the cache server, which is also similar to the one shown in FIG. 2. In FIG. 4, service provider 110 transmits all web requests to the cache server 140. The cache server then either responds to the request itself if the requested information is stored there, or retrieves the information on the Internet 100 via a satellite connection or other port 141 to the Internet. When it receives the requested information, the cache server 140 updates itself and, via satellite 160, other caches based on its own choices and then transmits the information to the end user 120 via service provider 110.

With respect to FIG. 4, it should be noted that the service provider can continue to independently manage other types of services, such as e-mail, etc., without forwarding such information packets to the cache server.

Now, based on the flowchart shown in FIG. 5, describe an example of operations performed by the cache server 140 in Figure 4. After the operation is started in step Sl, the cache server in step S2 receives an end user information request addressed to a specific content provider. Then, in step S3, the cache server determines whether or not a copy of the requested information is stored in it. If so, it passes the requested information to the end user in step S4 and the operation ends with step S10. (The final step S10 may, of course, trigger a return to the initial step Sl).

However, if it is determined in step S3 that a copy of the requested information is not stored on the cache server, then the procedure is continued with step S5, in which the cache server sends a request for that information to the specified content provider over the Internet. The cache server then receives the requested information in step S6 and transmits it to the end user in step S7. In step S8, the received information is stored in the cache server, and then in step S9 is distributed to all other cache servers in the area 200 shown in Figure 1. Now, the received new, previously unsaved information is stored essentially in all cache servers in the area and In this way, it has become easily accessible to all users in the area concerned.

Now, based on the flowchart shown in FIG. 6, describe an example of operations performed by the interceptor or service provider 110 in Figure 3. After the operation is activated in step Sil, the interceptor 110 in step S12 reads an end user information request addressed to a specific content provider. In step S13, the interceptor 110 examines the contents index 111 of the fast-access memory cache 112 to determine if a copy of the requested information is already in the memory cache. If located, the operation is continued with step S15, described below. If not, the operation proceeds with step S14, in which the interceptor queries the cache server 140 to determine if a copy of the requested information already exists in it. If so, proceed to step S15 intercepting the end-user request and step S16, where the requested information is transmitted either from the memory cache 112 or from the cache server 140. However, if the response of the cache server is negative, the process continues with step S17, in which the interceptor determines that the user's request is not intercepted but forwarded to the specified content provider via the internet. The process is then continued with step S18, in which the interceptor reviews the response that the content provider sends to the end user. When such information passes through the interceptor on its path to the end user, it stores the previously unreachable information in step S19 to the cache server 140 and distributes it to all other cache servers in the area 200 of FIG. 1. Now, the new, previously unsaved information received is now stored essentially on all the cache servers in the area and thus becomes easily accessible to all users in the area. It should be emphasized that whenever a piece of information is received from a network or a cache server, a copy of that information is preferably stored in a fast access cache server 112.

Now with reference to FIG. 7 describes another embodiment of an internet cache system of the present invention similar to the one shown in the figure

1. The system of FIG. 7 is from that of FIG. 1 distinguishes that the content provider response, in this case located in region E outside the system area 200, to the received end user information request 120a is not sent back to the user 100 over the network 100. Instead, the response is sent directly by satellite to servers 140a, 140b, 140c and 140e, so that all cache servers 140a are automatically updated. The cache server 140a and the HOa service provider then deliver that information to the end user 120a.

Now, based on Fig. 8, we will describe another embodiment of an interconnect cache system of the present invention similar to the one shown in FIG. 1. The cache servers in range 200, each serving its region A, B, C and D, are connected in Fig. 8 to a central control unit 190 via data channels. The central control unit 190 is responsible for deciding whether information saves or not to the individual cache server and how to distribute it to other caches in the 200 range. In this case, the information can be distributed to individual caches either via satellite or through incoming and outgoing data channel connections to the central control unit 190.

From the above figures, in particular Figures 1, 7 and 8, it is further understood that decisions about how caches can handle information related to "local" requests - such as the end-user information request 120a discussed in the Summary of the Invention, addressed to the content provider 130a, located in the same region A, and implemented various schemas, including making the necessary decisions, either with an individual cache server 140, or with an individual service provider 110, or e.g. at the central control unit 190.

It is understood that the composition and function of the elements described on the basis of the figures are clear to the person skilled in the art.

Although the invention is described on the basis of specific embodiments, it allows for various changes, modifications, etc. that are apparent to one skilled in the art. The embodiments described thus do not in any way limit the scope and scope of the invention as set forth in the appended claims. _

Claims (34)

PATENT APPLICATIONS 1. A method for caching online information, comprising the following two steps: taking over the information that the content provider sends to the end user in response to the information request received from him; and distributing said information to a set of geographically located cache servers. 2. The method of claim 1, wherein said information retrieval step is performed in conjunction with a cache server serving said end user, said end user being within the geographical area served by said cache server. The method of claim 2, which comprises storing said information in said cache server. The method of claim 3, which involves storing said information in said cache server only if said content provider is located outside the specified region. 5. The method of claim 3, which includes storing said information in said cache server only if said content provider is located within said region. 6. The method of claim 2, comprising the said broadcasting step only if said content provider is located outside the said region. 7. The method of claim 2, comprising the said broadcasting step only if said content provider is located within said region. A method according to any one of the preceding claims, comprising transmitting said information to said set of geographically located cache servers using multicast communication. The method of claim 8, which comprises transmitting said information to said set of geographically located cache servers via satellite. A method according to any one of the preceding claims, which involves the dissemination of said information only if so is the rule that specifies for each information received whether or not to distribute said set of geographically located cache servers. Method according to any one of the preceding claims, characterized in that said set of geographically distributed servers serve different geographical regions. Method according to any one of the preceding claims, characterized in that said set of geographically located cache servers is arranged within a linguistically or culturally defined area. 13. The method of claim 11, wherein said set of geographically located cache servers is arranged within a linguistically or culturally defined area comprising said regions. 14. An online cache system that includes: 15. 16. 17. 18. a set of geographically located cache servers; means for retrieving the on-line information that is retrieved in connection with the operation of one of the cache servers listed; and means for disseminating said retrieved information to substantially all of said cache servers. An internet caching system according to claim 14, characterized in that said means of receiving information are arranged such that said information in connection with the sending is provided by the internet content provider in response to a specific information request, at the end user located in which to serve said information retrieve information to the end user of the user receiving the specified geographic region, the cache server from the specified set of servers. An interconnect cache system according to claim 15, characterized in that said cache server of said set is arranged to store said information. An inter-cache system according to claim 16, characterized in that said cache server of said set is arranged to store said information only if said content provider is located outside the specified region. An inter-cache system according to claim 16, characterized in that said cache server of said set is arranged to store said information only if said content provider is located within said region. 19. An online cache system according to claim 15, characterized in that said means of distribution are arranged to distribute said information to said set of cache servers only if said provider is located outside the said region. 20. An online cache system according to claim 15, characterized in that said means of distribution are arranged to distribute said information to said set of cache servers only if said provider is located within said region. 21. An internet cache system according to any one of claims 14 to 20, characterized in that said means of distribution are arranged to distribute said information to said set of cache servers using multicast communication. 22. An internet cache system according to claim 21, characterized in that said broadcasting means are arranged to distribute said information to said set of cache servers via satellite. 23. An internet cache system according to any one of claims 14 to 22, which includes processing means for deciding whether or not to transmit the transmitted information to said set of cache servers. An interconnect cache system according to any one of claims 14 to 23, characterized in that said geographically distributed cache servers are arranged to serve different geographical regions. 25. An online cache system according to any one of; 'of claims 14 to 24, characterized in that said geographically located cache servers are distributed within a linguistically or culturally defined geographical area. 26. The internet caching system of claims 14 to 24, characterized in that said geographically located cache servers are distributed within a linguistically or culturally defined geographical area comprising the said regions. 27. An online cache server arranged in a geographical region for the purpose of serving primarily users in that region, including: means for receiving information sent by an online content provider to an end user located in the same region, in response to a request for information received from the user; and means for disseminating said information to a set of geographically located cache servers, with said servers serving multiple geographic regions. 28. An internet cache server according to claim 27, characterized in that said cache servers are arranged within a linguistically or culturally defined geographical area. 29. An internet cache server according to claims 2Ί or 28, which includes processing means for deciding whether or not to store the specified retrieved information. 30 / An internet cache server according to claims 27, 28 or 29, which includes processing means for deciding whether or not to transmit certain retrieved information to said set of geographically located cache servers. 31. The internet cache server of claim 29 or 30, characterized in that the decisions made by said processing aids are based on whether or not the said content provider is located in the said region. 32. An online cache server according to claims 29, 30, or 31, characterized in that said processing means are arranged to instruct said server not to store said information and not to distribute it to said set of geographically located cache servers if said final the user and the specified internet content provider are both located in the same region as the specified server. 33. An online cache server according to claims 29, 30, or 31, characterized in that said processing means are arranged to instruct said server to store said information but not to send it to said set of geographically located cache servers if specified the ISP is located in the same region as the specified server. 34. An online cache server according to claims 29, 30, or 31, characterized in that said processing means are arranged to instruct said server not to store said information but to distribute it to said set of geographically located cache servers if the specified Internet content provider is located in the same region as the specified server.