WO2002042874A2 - Method and architecture for serving and caching web objects on the modern internet - Google Patents

Method and architecture for serving and caching web objects on the modern internet Download PDF

Info

Publication number
WO2002042874A2
WO2002042874A2 PCT/US2001/043594 US0143594W WO0242874A2 WO 2002042874 A2 WO2002042874 A2 WO 2002042874A2 US 0143594 W US0143594 W US 0143594W WO 0242874 A2 WO0242874 A2 WO 0242874A2
Authority
WO
WIPO (PCT)
Prior art keywords
proxy cache
client
server
cache
request
Prior art date
Application number
PCT/US2001/043594
Other languages
French (fr)
Inventor
Gregory M. Mcgregor
Jeremy Zullo
Mark A. Shaw
Original Assignee
Azoth Technologies, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Azoth Technologies, Inc. filed Critical Azoth Technologies, Inc.
Priority to AU2002241497A priority Critical patent/AU2002241497A1/en
Publication of WO2002042874A2 publication Critical patent/WO2002042874A2/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/2866Architectures; Arrangements
    • H04L67/2876Pairs of inter-processing entities at each side of the network, e.g. split proxies
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/957Browsing optimisation, e.g. caching or content distillation
    • G06F16/9574Browsing optimisation, e.g. caching or content distillation of access to content, e.g. by caching
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/50Network services
    • H04L67/56Provisioning of proxy services
    • H04L67/568Storing data temporarily at an intermediate stage, e.g. caching
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/04Protocols for data compression, e.g. ROHC
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/30Definitions, standards or architectural aspects of layered protocol stacks
    • H04L69/32Architecture of open systems interconnection [OSI] 7-layer type protocol stacks, e.g. the interfaces between the data link level and the physical level
    • H04L69/322Intralayer communication protocols among peer entities or protocol data unit [PDU] definitions
    • H04L69/329Intralayer communication protocols among peer entities or protocol data unit [PDU] definitions in the application layer [OSI layer 7]

Definitions

  • a proxy is an intermediary program that acts as both a server and a client for the purpose of making requests on behalf of other clients Requests are serviced internally or may be passed to another server
  • the use of a cache attached to the proxy lets the proxy provide static data from local storage rather than acquiring the data through the WAN for each request
  • a proxy cache 130 on the client side 100 of the connection some improvements are introduced First, because the client 100 and proxy cache 130 are typically connected via a LAN (not shown) instead of a WAN, the client 100 no longer has to have direct internet access via the WAN 110 The client 100 only needs to be able to connect locally (through the LAN) and talk to the proxy cache 130. The proxy cache 130 will then broker the transaction, as necessary, with the server 120 through the LAN. This allows the client 100 to stay within the security of the LAN and not be exposed to the internet. The proxy cache 130 needs to be able to directly access the server 120. This allows the networking infrastructure to be optimized for web access to the proxy cache 130, and not every single client 100.
  • the use of a cache attached to a proxy lets the proxy provide static data from the cache rather than acquiring the data through the WAN for each request. This is done through a reactive technique.
  • the client makes a request, if the request exists in the cache it is served directly from the proxy, with no access to the origin server. This serving is done at the speed of the LAN or WAN, and with very little delay, or latency.
  • One drawback is that the origin server has no knowledge or log of this transaction taking place. If the object that the client request does not exist in the cache then the proxy contacts the origin server and retrieves the object. The proxy stores this object in the cache and in parallel serves the object to the client. Now, every request made for this object should be fetched from the cache if possible.
  • the invention comprises a method and architecture for serving and caching web objects comprising the steps of requesting, from a client, a request over a network; and receiving and fulfilling the request by a proxy cache located on a client side of the network; wherein the fulfillment of the request by the proxy cache is enabled by information passed to the proxy cache by at least one of a server and a reverse proxy cache located at a server side of the network via a pi-bonded connection with the proxy cache.
  • Figure 1 depicts a known server/client architecture
  • Figure 2 depicts a known server/client/proxy cache architecture
  • Figure 3 depicts a serving and caching architecture in accordance with the present invention
  • Figure 4 depicts the architecture illustrated in Figure 3, and further illustrates the use of a pi-bonded connection between the reverse proxy cache and the proxy cache;
  • Figure 5 depicts the architecture illustrated in Figure 4, however with a pi-bonded connection between the server and the proxy cache.
  • FIG. 3 Unlike the architecture shown in prior art Figure 2, Figure 3, illustrates a presently preferred embodiment of the present invention wherein a reverse proxy cache on the server side of the architecture. Specifically, cache 300, WAN, 310, server 320, and proxy cache 330 are provided, as shown. As used herein, the term "reverse proxy cache” means a server that proxies content on the server side of the network.
  • the client requests that is, requests from either or both of client 300 and proxy cache 330
  • the reverse proxy cache 340 thus acts as a virtual server from the point of view of the proxy cache 330 and the client 300. That is, the reverse proxy cache 340 brokers this transaction between the server 320 and the client side request, following the same or similar rules of the proxy cache 330. This allows the actual server 320 to be sitting behind the private network, thus more secure, than having it fully exposed to the internet.
  • the reverse proxy cache 340 further allows the server 320 to serve the content at LAN speeds to the proxy cache 330, and frees up resources faster, than having to wait for the request to finish at WAN speeds.
  • Origin servers such as server 320, need to be powerful, and thus an expensive resource, to handle dynamic content. Dynamic content comes
  • a pi-bonded connection 350 is used to connect proxy cache 330 with reverse proxy cache 340
  • the term "pi-bonded connection” means the communication between the server-side proxy cache and the client-side proxy cache, such that the two caches can intelligently share content to streamline the overall caching process In this fashion, as further discussed below the advantages of compression can be better exploited over traditional caching schemes
  • the pi-bond connection between devices can extend current protocols such as HTTP and RTSP or other communications means such as XML Other protocols such as UDP or multicast may also be supported by the pi-bond configuration of the present invention
  • XML also is a good carrier of information and could carry HTTP and RTSP within the XML
  • the pi-bond not only carries control information but it also controls and/or carries information that aids the originating suurce in determining whether the cache or request is permissible
  • proxy cache 330 and reverse proxy cache 340 may alternatively be connected via other configurations
  • present invention will be further discussed in the context of a pi-bonded connection
  • Another advantage resulting from the present invention is the ability to push objects to client-side proxy caches 330 that would normally not be cacheable because they would need to be tracked by the logs As will be understood by those of skill in the art, this may be achieved because traditionally people tag content as "not cacheable” because they lose control of the content and are uncomfortable with that As such, proxies won't cache content that is marked non-cacheable
  • the present invention allows those entities to keep some measure of control even when the content is served from third party networks
  • the pi-bonded control channel 350 allows the server 320 (as the owner or hoster of the content) to keep control of the flow and caching of the cached data Server 320 may thus control of the content and the benefits in terms of cost and performance of allowing the client-side proxy cache 330 to serve the content
  • the pi-bonded connection 350 may utilize special rules or protocols about what is and is not compressed
  • one of the benefits of compression is to be able to decrease the amount of redundant data
  • the full advantage of compression has not previously been realized
  • the present invention permits groups of small web-objects to be compressed together, thus taking full advantages of the inter-object redundancies of data and achieving far greater compression ratios
  • the present invention makes use of the fact that when browsing a web page, usually the browsers tend to request objects on the page in a similar pattern from someone else browsing the page That is, by having pi- bonded connection 350, the present invention enables the caching of content and gate access via the pi-bond
  • This architecture can be exploited according to the present invention Specifically, when a request is made, between pi-bonded enabled proxies 330 and 340, according to the present invention extra information may be sent down the pi-bonded channel to the client-side proxy cache 330 about what is the statistically highest requested object or objects after the presently-requeste ⁇ object
  • other useful statistical information may be sent as well to the client-side proxy cache 330, such as, for example, client side bandwidth, location (in terms of country, area code, etc ), personal data, and others
  • the reverse proxy cache 340 can start to send these additional objects, identified through statistics or other intelligence resident in server 320, before the objects

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Security & Cryptography (AREA)
  • Information Transfer Between Computers (AREA)
  • Computer And Data Communications (AREA)

Abstract

A new and novel method and architecture for serving and caching web objects on the modern internet is provided. A client requests a request over a network. The request is received and fulfilled by a proxy cache located on the client side of the network. The proxy cache is enabled to fulfill the request by information passed to the proxy cache by at least one of a server and a reverse proxy cache located at a server side of the network via a pi-bonded connection with the proxy cache.

Description

Method and Architecture for Serving and Caching Web Objects on the Modern Internet
Cross Reference to Related Application
[001] This application is related to and claims priority under 35 U S C
§119(e) from U S Provisional Application No 60/252,542, filed November 22, 2000, the entire contents of which are hereby incorporated by reference
Background of the Invention
[002] Currently, when a user desires to request and retrieve an object from the internet, an application program (i e , a "client" 100) in the user's computer (not shown) establishes a connection with a network such as a WAN 110, to a server 120 on which the requested object resides or is to be created This simple architecture is shown in Figure 1
[003] As a function of the server/client architecture illustrated in Figure 1 , all requests and connections are made directly from the client 100 through the WAN 110 to the server 120 That is, every object that is needed by the client 100 is requested by the client 100 from the server 120 each time Similarly, the requested object, which is either resident on or created by the server 120, is passed by server 120 to client 100 back through WAN 110 This architecture puts a large burden on the server 120, since every request from client 100 must be individually serviced by the server 120 This architecture also suffers from a large consumption of bandwidth because the server 120 must send every byte of every request into the WAN 1 10 The client 100 must also receive every byte of every request from the WAN 110 As a result, web object requests are fetched at the speed of the slowest link between the origin server 120 and client 100
[004] In an effort to address and improve the above drawbacks of the traditional architecture illustrated in Figure 1 , it is also known to provide a proxy cache 130 on the client side of the WAN This architecture is illustrated in Figure 2
[005] A proxy is an intermediary program that acts as both a server and a client for the purpose of making requests on behalf of other clients Requests are serviced internally or may be passed to another server The use of a cache attached to the proxy lets the proxy provide static data from local storage rather than acquiring the data through the WAN for each request By introducing a proxy cache 130 on the client side 100 of the connection some improvements are introduced First, because the client 100 and proxy cache 130 are typically connected via a LAN (not shown) instead of a WAN, the client 100 no longer has to have direct internet access via the WAN 110 The client 100 only needs to be able to connect locally (through the LAN) and talk to the proxy cache 130. The proxy cache 130 will then broker the transaction, as necessary, with the server 120 through the LAN. This allows the client 100 to stay within the security of the LAN and not be exposed to the internet. The proxy cache 130 needs to be able to directly access the server 120. This allows the networking infrastructure to be optimized for web access to the proxy cache 130, and not every single client 100.
[006] As stated above, the use of a cache attached to a proxy lets the proxy provide static data from the cache rather than acquiring the data through the WAN for each request. This is done through a reactive technique. The client makes a request, if the request exists in the cache it is served directly from the proxy, with no access to the origin server. This serving is done at the speed of the LAN or WAN, and with very little delay, or latency. One drawback, is that the origin server has no knowledge or log of this transaction taking place. If the object that the client request does not exist in the cache then the proxy contacts the origin server and retrieves the object. The proxy stores this object in the cache and in parallel serves the object to the client. Now, every request made for this object should be fetched from the cache if possible.
[007] Some of the benefits of this configuration are that is allows a decrease in the use of WAN bandwidth for fetching web objects. This configuration also speeds of the clients requests by serving the content at LAN speeds, which usually are significantly faster, than WAN speed.
[008] However, because of the dynamic nature of some web objects, or the need to log the transactions, certain objects cannot, or should not, be cached. These objects just pass through the proxy and are not stored in the cache. As a result, there remains a need for an architecture that will realize the improvements offered by the use of proxy caches while permitting the caching of object not previously cacheable.
Summary of the Invention
[009] To address the foregoing and other disadvantages of the various architectures previously known in the art and to fulfill the above-stated needs in the industry, applicants have invented a new and novel method and architecture for serving and caching web objects on the modern internet. As set forth and described in detail below, the invention comprises a method and architecture for serving and caching web objects comprising the steps of requesting, from a client, a request over a network; and receiving and fulfilling the request by a proxy cache located on a client side of the network; wherein the fulfillment of the request by the proxy cache is enabled by information passed to the proxy cache by at least one of a server and a reverse proxy cache located at a server side of the network via a pi-bonded connection with the proxy cache. Brief Description of the Drawings
[010] The present invention will now be described in detail with reference to the accompanying drawings, in which:
[011] Figure 1 depicts a known server/client architecture;
[012] Figure 2 depicts a known server/client/proxy cache architecture;
[013] Figure 3 depicts a serving and caching architecture in accordance with the present invention;
[014] Figure 4 depicts the architecture illustrated in Figure 3, and further illustrates the use of a pi-bonded connection between the reverse proxy cache and the proxy cache; and
[015] Figure 5 depicts the architecture illustrated in Figure 4, however with a pi-bonded connection between the server and the proxy cache.
Detailed Description of the Invention
[016] Reference is now made to Figure 3. Unlike the architecture shown in prior art Figure 2, Figure 3, illustrates a presently preferred embodiment of the present invention wherein a reverse proxy cache on the server side of the architecture. Specifically, cache 300, WAN, 310, server 320, and proxy cache 330 are provided, as shown. As used herein, the term "reverse proxy cache" means a server that proxies content on the server side of the network.
[017] The introduction of a reverse proxy cache 340 in front of the server
320 allows a significant load to be removed from the server 320. Specifically, the client requests, that is, requests from either or both of client 300 and proxy cache 330, are sent to the reverse proxy cache 340. The reverse proxy cache 340 thus acts as a virtual server from the point of view of the proxy cache 330 and the client 300. That is, the reverse proxy cache 340 brokers this transaction between the server 320 and the client side request, following the same or similar rules of the proxy cache 330. This allows the actual server 320 to be sitting behind the private network, thus more secure, than having it fully exposed to the internet. The reverse proxy cache 340 further allows the server 320 to serve the content at LAN speeds to the proxy cache 330, and frees up resources faster, than having to wait for the request to finish at WAN speeds.
[018] Origin servers, such as server 320, need to be powerful, and thus an expensive resource, to handle dynamic content. Dynamic content comes
j - from a multitude of source, databases, scripts, applications, etc By putting a reverse proxy cache 340 in front of these servers, it allows the offloading of the non-dynamic, thus cacheable, objects to be served from the reverse proxy cache 340 instead of tying up the valuable resources of the origin server Allowing the origin server to have more cycles of dynamic content to spend, means more dynamic content can be served
[019] Further according to the present invention, all the requests are still request in a reactive manner That is, a web object is only cached if the object is requested
[020] Additional new and novel features of the present invention will now be discussed in connection with Figure 4 As shown therein, a pi-bonded connection 350 is used to connect proxy cache 330 with reverse proxy cache 340 As used herein, the term "pi-bonded connection" means the communication between the server-side proxy cache and the client-side proxy cache, such that the two caches can intelligently share content to streamline the overall caching process In this fashion, as further discussed below the advantages of compression can be better exploited over traditional caching schemes According to the present invention, the pi-bond connection between devices can extend current protocols such as HTTP and RTSP or other communications means such as XML Other protocols such as UDP or multicast may also be supported by the pi-bond configuration of the present invention XML also is a good carrier of information and could carry HTTP and RTSP within the XML In addition, it should be noted that as described herein the pi-bond not only carries control information but it also controls and/or carries information that aids the originating suurce in determining whether the cache or request is permissible
[021 ] Of course, as will be understood by those skilled in the art, the present invention is not limited to pi-bonded connections For example, proxy cache 330 and reverse proxy cache 340 may alternatively be connected via other configurations For discussion purposes only, however the present invention will be further discussed in the context of a pi-bonded connection
[022] By virtue of the architecture of the present invention a special level intelligence may be created that can be passed between the pi-bonded proxies and servers, reverse, forward, and transparent, better optimizations can be achieved between the proxies and/or serving solutions, than simple reactive caching That is, with traditional find and fetch configurations, there is little or no time to inject user defined requirements By virtue of the construction of the present invention, however, the ultimate originating source can have input for control purposes at a third party location As a result, certain features not heretofore available can be enabled For example, once the reverse proxy cache 340 has identified a possible client-side proxy cache 330 with which it may establish a pi-bonded connection, the reverse proxy cache 340 may choose to pin a subset of it's content to the proxy cache 330 By "pinning" the content it is meant a manual function of passing the content to 'proxy cache 330 Thus, as a result of the present invention, content can be put in the proxy cache 330 either by reverse proxy cache 340 or directly from server 320 (as shown in Figure 5) before even the first user requests the data That is, even if an edge device (such as proxy cache 330) has not seen an object requested the origin server has and can give information so that the requesting device knows that if one object has been requested certain others are likely to follow Such other likely objects therefore can be requested in front of the actual client side request This intelligence may also be advantageous at the cache side with regard to client requests There are many advantages resulting from this configuration For example, if the content provider behind server 320 knows that a spike of requests might be coming it's way from clients, server resources may be saved by pinning some of the content into client-side proxy caches 330 to help smooth out the initial spike of requests, thus significantly reducing the time necessary for the client 300 fetch the requests Notably, the enhancement of the end-user experience resulting from the increased speed and efficiency of the present invention apply even for the first users to fetch the requested objects This is especially attractive for bandwidth-intensive objects, such as video and other heavy objects, as they can be thus be cached at the proxy cache 330 and accessed locally by client 300 by a LAN (not shown)
[023] Another advantage resulting from the present invention is the ability to push objects to client-side proxy caches 330 that would normally not be cacheable because they would need to be tracked by the logs As will be understood by those of skill in the art, this may be achieved because traditionally people tag content as "not cacheable" because they lose control of the content and are uncomfortable with that As such, proxies won't cache content that is marked non-cacheable The present invention, however, allows those entities to keep some measure of control even when the content is served from third party networks Additionally, the pi-bonded control channel 350 allows the server 320 (as the owner or hoster of the content) to keep control of the flow and caching of the cached data Server 320 may thus control of the content and the benefits in terms of cost and performance of allowing the client-side proxy cache 330 to serve the content
[024] By virtue of the structure and function of the present invention, including the provision of pi-bonded connection 350 between the server-side and client-side proxy caches, a richer set of compression is possible between caches More specifically, the pi-bonded connection 350 may utilize special rules or protocols about what is and is not compressed Of course, one of the benefits of compression is to be able to decrease the amount of redundant data But with individual small web objects the full advantage of compression has not previously been realized By using pi-bonding, as explained above, the present invention permits groups of small web-objects to be compressed together, thus taking full advantages of the inter-object redundancies of data and achieving far greater compression ratios
- [025] Additionally, as will be apparent to those skilled in the art, with some added intelligence and management on the reverse proxy cache 340, very intelligent and complex cache pushing can be achieved by the present invention, such as may be useful in special high-speed, highly optimized and fast web surfing Established jobs can be done at specified time windows, with a maximum pipe size specified by the lesser of the server and receiver These can be done with any type of web object
[026] First, a table a statistics is maintained by the proxy, such as proxy cache 330 Next, the present invention makes use of the fact that when browsing a web page, usually the browsers tend to request objects on the page in a similar pattern from someone else browsing the page That is, by having pi- bonded connection 350, the present invention enables the caching of content and gate access via the pi-bond This architecture can be exploited according to the present invention Specifically, when a request is made, between pi-bonded enabled proxies 330 and 340, according to the present invention extra information may be sent down the pi-bonded channel to the client-side proxy cache 330 about what is the statistically highest requested object or objects after the presently-requesteα object Of course, other useful statistical information may be sent as well to the client-side proxy cache 330, such as, for example, client side bandwidth, location (in terms of country, area code, etc ), personal data, and others The reverse proxy cache 340 can start to send these additional objects, identified through statistics or other intelligence resident in server 320, before the objects are even requested by the user from the server 330, through a simple cache pin, as described above Then when the client goes to make the request, the requested object or objects are already in the client-side proxy cache 330 and thus readily and quickly available to client 300 via the LAN Further, the proxy cache 330 can also start to prepare for the potential request, even before the request is made by the client 300, as discussed above, for example, in paragraph 22 Significantly, according to the present invention, this also can be done for objects that are not traditionally treated as cacheable, since the cached objects will only reside in the client-side proxy cache 330 for a finite amount of time, and are pinned to the proxy cache 330 just shortly before the client 300 would normally request the objects As a result, the present invention permits some forms of dynamic data to get all of the speed benefits of a cache, without "actually" caching the object End-users, through clients 300, can thus surf the internet at speeds that exceed the theoretical limits possible without the pi-bonding configuration of the present invention
[027] The present invention has been described above in the context of the presently preferred embodiments known to the inventors Of course, other embodiments and variations will be apparent to those of skill in the art without departing from the spirit or scope of the present invention
Claims

Claims

What is claimed is:
1. A method of serving and caching web objects comprising the steps of: requesting, from a client, a request over a network; receiving and fulfilling said request by a proxy cache located on a client side of said network; wherein the fulfillment of said request by said proxy cache is enabled by information passed to the proxy cache by at least one of a server and a reverse proxy cache located at a server side of said network via a pi-bonded connection with said proxy cache.
PCT/US2001/043594 2000-11-22 2001-11-23 Method and architecture for serving and caching web objects on the modern internet WO2002042874A2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU2002241497A AU2002241497A1 (en) 2000-11-22 2001-11-23 Method and architecture for serving and caching web objects on the modern internet

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US25254200P 2000-11-22 2000-11-22
US60/252,542 2000-11-22

Publications (1)

Publication Number Publication Date
WO2002042874A2 true WO2002042874A2 (en) 2002-05-30

Family

ID=22956450

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2001/043594 WO2002042874A2 (en) 2000-11-22 2001-11-23 Method and architecture for serving and caching web objects on the modern internet

Country Status (2)

Country Link
AU (1) AU2002241497A1 (en)
WO (1) WO2002042874A2 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7171469B2 (en) * 2002-09-16 2007-01-30 Network Appliance, Inc. Apparatus and method for storing data in a proxy cache in a network
US7284030B2 (en) 2002-09-16 2007-10-16 Network Appliance, Inc. Apparatus and method for processing data in a network
US7552223B1 (en) 2002-09-16 2009-06-23 Netapp, Inc. Apparatus and method for data consistency in a proxy cache

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7171469B2 (en) * 2002-09-16 2007-01-30 Network Appliance, Inc. Apparatus and method for storing data in a proxy cache in a network
US7284030B2 (en) 2002-09-16 2007-10-16 Network Appliance, Inc. Apparatus and method for processing data in a network
US7552223B1 (en) 2002-09-16 2009-06-23 Netapp, Inc. Apparatus and method for data consistency in a proxy cache

Also Published As

Publication number Publication date
AU2002241497A1 (en) 2002-06-03

Similar Documents

Publication Publication Date Title
US9497284B2 (en) Apparatus and method for caching of compressed content in a content delivery network
Touch et al. LSAM proxy cache: a multicast distributed virtual cache
Davison A web caching primer
US6701415B1 (en) Selecting a cache for a request for information
US6959318B1 (en) Method of proxy-assisted predictive pre-fetching with transcoding
KR100571059B1 (en) Distributed Systems and Methods for Prefetching
US6892218B2 (en) Extending network services using mobile agents
EP1175651B1 (en) Handling a request for information provided by a network site
US6587928B1 (en) Scheme for segregating cacheable and non-cacheable by port designation
US20020007404A1 (en) System and method for network caching
US20110219109A1 (en) System and method for sharing transparent proxy between isp and cdn
US8166198B2 (en) Method and system for accelerating browsing sessions
JP2000013779A (en) Data distributing method and proxy internet server
US6807572B1 (en) Accessing network databases
US20180302489A1 (en) Architecture for proactively providing bundled content items to client devices
WO2019052299A1 (en) Sdn switch, and application and management method for sdn switch
WO2002042874A2 (en) Method and architecture for serving and caching web objects on the modern internet
Eden et al. Web latency reduction via client-side prefetching
Selvakumar et al. Implementation and comparison of distributed caching schemes
Chi et al. Proxy-cache aware object bundling for web access acceleration
KR100394189B1 (en) Method for servicing web contents by using a local area network
Rangarajan et al. A technique for user specific request redirection in a content delivery network
Shen et al. Streaming media caching with transcoding-enabled proxies
Ahuja et al. Cache on demand
Hussain Intelligent Prefetching

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A2

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT TZ UA UG US UZ VN YU ZA ZW

AL Designated countries for regional patents

Kind code of ref document: A2

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR

121 Ep: the epo has been informed by wipo that ep was designated in this application
WA Withdrawal of international application
REG Reference to national code

Ref country code: DE

Ref legal event code: 8642