Based Content Switching Techniques
4/26/2009
i
Content Based Switching Techniques
4/26/2009
A Study on Content based Switching
By Baba Anem Ganesh Kumar Sudhakar Kasireddy Date : 12/07/2000
CS522 - Computer Communications Dr. Edward Chow
ii
Content Based Switching Techniques
4/26/2009
iii
Content Based Switching Techniques
4/26/2009
1.
Project Goals:
The goals are to study the design and implementation of content-based switch.
2.
Introduction
The Internet has radically changed the way users and corporations use networks. In particular, the emergence of the Web changes the way information is located and accessed. With the client/server model, content is managed by proprietary servers and is only accessible by a small community of users. With the Webbased computing model, all content managed by Web servers is accessible to all users unless they are explicitly prohibited. If a large number of users want to access a "hot" content area at the same time, "flash" network crowds occur. This phenomenon creates sporadic traffic loads on the Internet that can stress the current infrastructure beyond its limits. The Internet is also changing how information is delivered from passive content retrieval to proactive content delivery based on network policies and user identity. This proactive information delivery model is best represented by the emerging "push" technologies that deliver information to users as it becomes available. The passive retrieval model requires a network infrastructure that is built for predictable network and server loads. The proactive delivery model requires that content be intelligently distributed closer to clients and network access points to better cope with sporadic network loads driven by hot content.
1
Content Based Switching Techniques
4/26/2009
3.
Effective Information Management
Just as there are many types of jungles, so there are many types of data networks. And jungles and networks have some striking similarities in the way they are organized. In the jungle, the parts of the whole are called ecosystems; in the network, they are called layers. Each subsystem, or layer, is often quite distinct from others within the same system or network, but depends upon access to the others for its survival. Call it the food chain or call it the protocol stack. Knowledge of layering is crucial for the strategic and tactical deployment of both networking and information technology in an organization. Many people view layering as an academic exercise in which Layer 2 represents switching and Layer 3 represents routing. Such shortsighted thinking leaves many organizations at the mercy of the performance constraints of their collapsed backbone routers. Understanding the capabilities and limits of each layer is the foundation for information management. Strategic decisions must be made about application deployment, network scalability, performance, and cost of ownership. Tactical decisions must be made about which products to apply as part of an overall solution. This methodology becomes even more important as voice, video, and data networks continue to converge, blurring the once clear demarcation between data communications and telecommunications.
3.1.
Overview of layering
Layering schemes provide guidelines, rather than strict rules, for delegating networking functionality. Figure 1 shows the basic principles of layering. Elements at the same layer, shown on the horizontal, are known as peers and communicate via a well-known (and documented) protocol. Messages are exchanged among peers, the protocol defining the format, syntax, semantics, and sequencing. Elements within the same stack, shown on the vertical, communicate via an internal interface. This interface, though usually not well documented nor standard, often exhibits the same
2
Content Based Switching Techniques
4/26/2009
characteristics as a protocol, the only difference being that the interface protocol between Layer n and Layer n+1 on stack 1 may be wholly different from that of stack 2.
Figure 1. Layering Reference Model
As mentioned, communication within one stack may be different from that within other stacks and entirely proprietary, but communication between peers in different stacks must be open and consistent. The notion of open systems has been a major factor in the growth and operation of the Internet, along with those of institutional organizations. It is also important to note that an element at a particular layer may be further broken down into additional layers. This is most clearly seen with Asynchronous Transfer Mode (ATM) models. Finally, in certain models, higher layers may share information with lower layers to either conserve system resources or improve performance. The Internet Engineering Task Force (IETF) Next-Hop Resolution Protocol (NHRP) is an example of this intralayer communication, allowing Layer 3 "shortcuts." This concept will be discussed later.
3.2.
Contemporary Layering Model
For many years, the OSI model (Figure 2) was the reference layering paradigm for data networking. The OSI model was an extremely powerful architecture that included well-defined Layer n/Layer n+1 protocols in addition to rich peer-to-peer protocols. Unfortunately, much of this model
3
Content Based Switching Techniques
4/26/2009
succumbed to the complexity of the protocols and the effects of an overly rigorous standardization process. Since only a few elements survived to become part of the contemporary networking model, no further analysis will be made of this model.
Application Presentation Session Transport Network Data Link Physical Figure 2. OSI Layering Model
The contemporary network layer architecture is much simpler than its OSI counterpart. Originating from various research and defense initiatives, the contemporary model was intended to be supplanted by OSI. Instead, it became the de facto networking standard, especially through IP. As mentioned, both IPX and AppleTalk are quite similar to IP, but are slowly becoming less prominent as IP dominance continues to grow. This discussion will emphasize IP, but the methods discussed can easily be applied to environments using NetWare and Apple protocols. Figure 3 shows the contemporary networking model based upon IP. Network participants, whether infrastructure equipment (switches and routers) or end systems (clients and servers), may include some or all of the protocol stack.
4
Content Based Switching Techniques
4/26/2009
Application Transport Routing Switching Interface Figure 3. Contemporary Layering Model
3.2.1. Layer 1 This layer, known as the interface layer, is responsible for device connectivity. Though usually represented by well-known network types-- Ethernet, Fast Ethernet, Gigabit Ethernet, Token Ring, FDDI, ATM, SONET/SDH, etc.--Layer 1 also covers the subtypes. For example, Fast Ethernet provides physical connectivity over copper media (100BASETX) and over fiber media (100BASE-FX). Fiber can be further divided into multimode or single mode, with single mode further partitioned based on its "reach," the distance over which it can transmit. Certain technologies are actually used as a pure Layer 1 element (SONET/SDH) or provide a virtual Layer 1 element (ATM with SONET/SDH). While the various types of Ethernet are rather straightforward, FDDI , ATM, and SONET/SDH add more complexity, while providing extended Layer 1 capabilities such as fault tolerance and support for physical multiplexing to support distinct traffic flows such as voice and data. With these added capabilities comes added cost, and sometimes slower performance. 3.2.2. Layer 2 This layer, known as the switching layer, allows end station addressing and attachment. Because architectures up to Layer 2 allow end station connectivity, it is often practical to construct a Layer 2-only network,
5
Content Based Switching Techniques
4/26/2009
providing simple, inexpensive, high-performance connectivity for hundreds or even thousands of end stations. The past five years have seen the extraordinary success of the "flat" network topologies provided by Layer 2
switches connected to other Layer 2 switches or ATM switches.
Layer 2 switching, also called bridging, forwards packets based on the unique Media Access Control (MAC) address of each end station. Data packets consist of both infrastructure content, such as MAC addresses and other information, and end-user content. At Layer 2, generally no modification is required to packet infrastructure content when going between like Layer 1 interfaces, like Ethernet to Fast Ethernet. However, minor changes to infrastructure content--not end-user data content--may occur when bridging between unlike types such as FDDI and Ethernet. Either way, processing impact is minimal and so is configuration complexity. Layer 2 deployment has seen the most striking infrastructure change over the past decade. Shared Ethernet, represented by particular cable types or contained within shared hubs, offered a very simple, and even more inexpensive, approach for Layer 2. Though still quite popular, shared technology, where all stations use the same bandwidth slice, has very limited scaling capabilities. Depending upon the applications being used, shared networks of more than one hundred users are becoming less common. Many network designers have "tiered" their infrastructure by feeding shared Layer 2 into switched Layer 2 or even Layer 3. Switched Layer 3 apportions each station--or port--its own dedicated bandwidth segment. Recent enhancements at Layer 2 provide packet prioritization capabilities for the application of network policies. The new IEEE 802.1p standard defines Class of Service (CoS) policies capabilities for Layer 2 segments.
6
Content Based Switching Techniques
4/26/2009
Note that Layer 2 does not ordinarily extend beyond the corporate boundary. To connect to the Internet usually requires a router; in other words, scaling a Layer 2 network requires Layer 3 capabilities. 3.2.3. Layer 3 This layer, known as the routing layer, provides logical partitioning of sub networks, scalability, security, and Quality of Service (QoS). QoS, a recent enhancement to Layer 3, goes beyond the simple packet prioritization found in CoS by providing bandwidth reservation and packet delay bounding. The backbone of the Internet, along with those of many large organizations, is built upon a Layer 3 foundation. IP is the premier Layer 3 protocol. In addition to Layer 2 MAC addresses, each IP packet also contains source and destination IP addresses. For an intranet packet, one IP address addresses the client, the other the server. IP in itself is not a particularly complex protocol; extensive capabilities are supplied by the other components of the IP suite. The Domain Name System (DNS) removes the burden of remembering IP addresses by associating them with real names. The Dynamic Host Configuration Protocol (DHCP) eases the administration of IP addresses and is used extensively by network administrators and Internet service providers (ISPs). Routing protocols such as Open Shortest Path First (OSPF), Routing Information Protocol (RIP), and Border Gateway Protocol (BGP) provide information for Layer 3 devices to direct data traffic to the intended destination. IP Security (IPsec) furnishes elements necessary for security, such as authentication and encryption. IP not only allows for user-to-user communication, but also for efficient dissemination over point-to-multipoint flows, known as IP Multicast. Higher-layer protocols, discussed later in this paper, provide even greater versatility for content distribution.
7
Content Based Switching Techniques
4/26/2009
Although
many
organizations
received
tremendous
performance
advantages by converting routed and shared networks to Layer 2 switching, it became apparent that some level of partitioning was still required. Consequently, routers maintained a presence at many points within a corporate network. For a while this presented minimal problems, since a majority of the data traffic stayed local to the subnet, which was increasingly being serviced by a Layer 2 switch. But concurrent with the increasing acceptance of Layer 2 switching as an essential component of network infrastructure were two other developments: the migration of servers to server farms for increased security and management of data resources; and the deployment of intranets, organization-wide client/server communications based on Web technology. These factors began moving data flows off local subnets and onto the routed network, where the limitations of router performance increasingly led to bottlenecks. With the routers causing information flow constriction, IT managers became increasingly reluctant to deploy new, enabling technologies, such as multicast-based applications and middleware. Even the migration of desktops to higher-performance media connections, such as 100 Mbps Fast Ethernet, were scrutinized as long as 10 Mbps router interface funnels were in place. Router vendors attempted to respond by offering higher-performance interface cards, but throughput was fundamentally bounded by centralized, software-based architectures that simply could not go any faster. The same software responsible for managing WAN links, X.25, and asynchronous terminal lines was now expected to handle next-generation gigabit networks. Router vendors tried distributing functionality to improve performance, resulting in a hodgepodge collection of route processing and interface cards. Was the device still routing, or was it performing some other packet forwarding scheme?
8
Content Based Switching Techniques
4/26/2009
Emerging QoS was even more suspect. The IETF was moving forward on Resource Reservation Protocol (RSVP), a signaling method to set up bandwidth and delay control on packet-based internetworks. Monitoring RSVP flows, using a process know as policing, required extensive software support on already overburdened legacy routers. Could QoS be practical on a contemporary LAN? Meanwhile, standards bodies such as the ATM Forum were working on methods to offload the Layer 3 bottleneck by exploiting the capabilities of the lower layers. One result was the Multiprotocol over ATM (MPOA) specification, which uses Layer 3 routing information and the IETF's NHRP protocol to offload the routers and provide forwarding at the physical (ATM) layer. A Layer 3 switch can route at Layer 3 or utilize MPOA; the performance is identical. 3.2.4. Layer 4 This layer, known as the transport layer, is the communication path between user applications and the network infrastructure and defines the method of communicating. TCP and UDP are well-known examples of elements at the transport layer. TCP is a "connection-oriented" protocol, requiring the establishment of parameters for transmission prior to the exchange of data. Web technology is based on TCP. UDP is "connectionless" and requires no connection setup, which is especially important for multicast flows. Elements at this level also differ in the amount of error recovery provided and whether or not it is visible to the user application. Both TCP and UDP are layered on IP, which has minimal error recovery and detection mechanisms, leaving the burden at Layer 4 or higher. TCP forces retransmission of data that was lost by the lower layers, whereas UDP makes the application responsible. A major enhancement to multimedia support at Layer 4 is the Real Time Protocol (RTP). RTP works in conjunction with UDP, and provides
9
Content Based Switching Techniques
4/26/2009
services necessary for packet timing and sequencing. Many time-sensitive applications running over IP networks now actually include both UDP and RTP. 3.2.5. Layer 5 This layer, known as the application layer, provides access to either the end user or some type of information repository such as a database or data warehouse. Users communicate with the application, which in turn delivers data to the transport layer. Applications do not usually communicate with the lower layers; rather, they are written to interface with a specific communication library, like the popular WinSock library available in Windows-based workstations. When defining the behavior of the applications they are writing, developers decide on the type of transport mechanism necessary. For example, database or Web access requires robust, error-free access and would demand TCP, though it could be implemented with more code and in a more cumbersome manner with UDP. Multimedia, on the other hand, cannot tolerate the overhead of connection-oriented traffic and will commonly make use of UDP. For prioritization, either TCP nor UDP can be selected, depending on the application or other parameters such as time of day. Any assistance that a network device can provide in terms of prioritization of the application would be extremely beneficial to the network manager, particularly during times of traffic volume from the LAN to the WAN.
10
Content Based Switching Techniques
4/26/2009
4.
Content Aware Paradigm
In the new content-aware paradigm, a desktop client deals with content in the network directly the network is simply a background enabling technology that provides content-aware delivery. In other words, a content-aware desktop requires a Content Smart network environment that facilitates the distribution of content requests to locations where the content was available most recently. This eliminates unnecessary network loads due to content requests being dropped, causing frequent re-transmissions. A Content Smart network allocates network bandwidth based on the content and user requirements; safeguards content based on defined access policies; and is composed of the following services: Content Location Services Content Distribution and Replication Content Caching Content Redirection and Forwarding
4.1.
Content Location Services
A Content Smart network requires the resolution of a content request to a specific server based on information about the content's location and service requirements. Ideally, the location of a content item should be discovered dynamically by querying a Content Location Service. A Content Location Service is a name server that identifies a content request by its URL (and possibly other URL and URI extensions in the future), and resolves the URL to the server that is best suited to serve the content at that moment in time. Each Content Location Service manages a set of servers and server resources that comprise a given community of interest. The Content Location Service is transparent to the client because a content request is directed either to a Content Location Service or to the target server depending on how the content request is resolved.
11
Content Based Switching Techniques
4/26/2009
A Content Location Service agent relays content requests to, and exchanges content information with, other Content Location Service agents for request resolution. That is, Content Location Service agents that manage a common URL name space exchange information to form a logical URL name in effect, an alias. It is desirable that the URL name space is organized in a hierarchical manner similar to that of DNS names to achieve scalability in content location searches. The most fundamental building block for content location services to be developed is a globally unique scheme for content identification. The ability to uniquely identify content allows the network to become contentaware without necessarily looking into the content itself. The combination of IP Address, Domain Name, Port, and URL is sufficient to provide this content unique identification. Much like the composition of a file can be inferred by examining its associated file attributes, the content pointed to by this unique content identifier can be deduced by examining attributes embedded in an HTTP message, as well as attributes stored by the Content Location Service. While this content ID determines the location and attributes of the content, it does not assure that two instances of content bearing the same name are identical. As a possible solution to this problem, the URL could be extended to include a message digest attribute. A message digest can be thought of as a digital "fingerprint" of the larger document. Many message digest algorithms exist and have been widely used in cryptography. The MD5 algorithm is well known for computing a 128-bit checksum for any file or object. Since the likelihood of two different content items producing the same MD5 checksum is very small (about 1 in 264), the MD5 checksum of a content item could be used to construct a reliable content identifier that uniquely locates and identifies content. This optimization would simplify the task of checking the consistency of content.
12
Content Based Switching Techniques
4/26/2009
5.
Content Distribution and Replication
A Content Smart network is keenly aware of how critical content is replicated and distributed. The objective is for content to be pushed as close to a target client group as possible. One way this can be achieved is by caching content pages on proxy or cache servers located at a customer premise or an ISP before the traffic reaches a network access point where network traffic is exchanged. A Content Smart network makes content distribution and replication completely transparent from a requesting client. That is, a content request can be made to a content-aware switch that seamlessly redirects the content request to the "best-fit" server in the network based on the content intelligence it learns from the network and from other Content Location Services in the network. In other words, unlike a content-blind network where packets are directed to the fixed network address contained in the packet header, a unique content identifier constructed from the content's URL and other information identifies the content being requested. The request is directed transparently to the optimal server based on dynamic network conditions, server load, and content-specific Quality of Service (QoS) requirements. Since a content item could become hot instantly, it needs to become accessible at network locations where the most content requests are expected. A hot content item can be detected by a Content Smart network by tracking content hit rates and sharing those statistics among fully- or partially-meshed network devices. Content caching based on content hit count and last-access-time insures a hot content item is more likely to be cached than other pages. This is in contrast with traditional caching models based on least-recently-used (LRU) algorithms. A content item can be versioned so that only changes to the content since it was last accessed will be distributed. This content intelligence, along with the ability to push content to multiple, distributed, mirrored Web servers or caches using multicasting, forms the basis for how content will be distributed in a Content Smart network. Complex content could consist of a number of identifiable sub-
13
Content Based Switching Techniques
4/26/2009
content items that can be distributed and replicated freely. The majority of content actually contains static information that does not change over time, while a small percentage of the content is derived dynamically. The ability to do differential loads of complex content by only loading the small percentage of sub-content which has changed is key to the future of the Content Smart network. The ability to do differential loads along with the ability to send a copy of content as close to a local distribution point as possible will drive how content is distributed and replicated in the future.
14
Content Based Switching Techniques
4/26/2009
6.
Content Caching
Within a Content Smart network, content must be cached within an infrastructure that distributes intelligent cache servers along major network paths between clients, ISPs, and backbones (see Figure 4). A collaborative caching scheme involving client-side caching, network-side caching, and server-side caching allows each to serve a specific cache context. In this way, a client or its proxy requests information from the closest content location and extends the search further a field if it is unavailable.
Figure 4 : Content Caching
The problem of content caching is different depending on whether a caching point is near a client population, a content farm, or a network exchange point. Proxies near a client population must keep their caches populated with content having a
15
Content Based Switching Techniques
4/26/2009
high hit ratio specific to the client population's common characteristics. Proxy cache servers that are closer to content farms are specialized to handle specific content based on content hit count and content type. Network cache servers are typically located at a customer premise or in the ISP's data centers, and are used to prevent unnecessary traffic from passing across a potentially congested network exchange point.
16
Content Based Switching Techniques
4/26/2009
7.
Content Forwarding and Redirection
A Content Smart network must be able to respond to a request for content based on the availability and location of the content, available servers where the requested content can be found, and the network conditions at the time of the request. Since these attributes are dynamic in nature, the best server to handle a request might be different at any given time. Depending on the conditions mentioned above, a content request in a Content Smart network actually points to a target server that will either service the request locally or redirect the client to other servers transparently. This is in contrast with today's passive network, where requests are directed to a fixed network address without any awareness of network conditions or where duplicate content may be located. With transparent content redirection, a content server farm lets an enterprise use all of its content sites as though they were part of a single system, while at the same time letting each individual content site stay under the control of its local owners. To be able to dynamically determine which local site to visit, and the most cost-effective way to reach that site, the following intelligence must be integrated into a single Web traffic management framework:
An Internet-attached Web content router will determine the "best site" for the URL request and forward the request to that site. At the target site, either software intelligence or a local traffic director device is used to manage multiple local servers in order to provide load balancing and redundancy in case of a server failure. Content management is used to affect how content is distributed across the Internet and locally within a Web site. It is also used as a form of traffic partitioning based on URL content. At each site, application proxy services that provide application-level authentication provide secure Web services, such as banking, transaction processing, and public document management.
17
Content Based Switching Techniques
4/26/2009
8.
The Elements of a Content Smart Network
It is clear that Web traffic behavior is significantly different from today's client/server paradigm. The next generation of Internet device must address Web traffic analysis and handling in an intelligent, efficient manner. In order to achieve the promise of the Content Smart network, this new class of networking devices must fully-address the challenges of a content-aware network.
8.1.
Content and User Identity
Within the Content Smart networking paradigm, a browser becomes the access point for all content accessible on the network. This leads to the inevitable conclusion that user identity is passed to the network in the form of HTTP authentication messages in a Content Smart network. It also follows that a content item is identified by the URL a user specifies plus other associated URL attributes. Thus, a Content Smart network must be HTTP-aware in order to direct content flows to the optimal proxy or content servers. In other words, a Content Smart network must be keenly aware of content flows and the policies of those flows related to bandwidth, access privileges, and priority. Edge flow switching devices that are closer to clients or content servers are more likely to deal with per-flow policies. The core of the Content Smart network facilitates the transport of flows based on "flow classes," which aggregate flows of similar QoS requirements and common destinations into flow bundles.
8.2.
Web-Aware Server and Storage Systems
In a Content Smart network, Web servers are quickly becoming synonymous with content servers. As Web servers become the de facto application development platform, general-purpose servers will give way to specialized Web servers optimized for Web transaction processing. These Web servers will be armed with intelligent Web caching hardware and software, intelligent switched storage systems tightly integrated with the server, and an optimized Web development kernel that replaces today's general purpose operating systems. In a Content
18
Content Based Switching Techniques
4/26/2009
Smart network, content servers and proxy cache servers must work in concert across the network to facilitate the distribution and replication of content. Web caching differs from traditional server caching in several aspects. Web pages are often dynamically constructed from reusable components and some real-time data. As such, they must be kept together in cache or aged out altogether. Web page access distribution is non-uniform, in that it is heavily weighted in favor of index pages and instant hot content pages. As such, Web caching needs to be highly sensitive to content hit count, as well as last access time as determined by LRU algorithms. Since a content database could be served by a group of servers, the concept of a logical cache array to preserve inter-server cache coherency becomes an important issue. In a Content Smart network, content servers must exchange content information to form a global content space within a content overlay network. More importantly, content servers must participate in the propagation of updated content and redistribution of hot content to local distribution points.
8.3.
Content Smart Network Infrastructure
The role of the Internet core in a Content Smart network is to enable high-speed content delivery via high-speed switching technologies. Tags, VCs, and IP precedence bit settings are used to identify a flow or a class of flows. A Content Smart network functions as a logical overlay on top of the physical, high-speed switching core providing the necessary content identification, flow classification, and tagging required by the "flow-class-aware" core switched infrastructure (see Figure 5).
19
Content Based Switching Techniques
4/26/2009
Figure 5. Content Smart Network Infrastructure
A Content Autonomous System (CAS) is an independent system that shares a content administrative domain. Many content autonomous systems exist within a larger Content Smart network, each functioning as a Content Smart intelligent Virtual Private Network (VPN). A Content Smart VPN is a Content Smart overlay network over the Internet where all content is shared and freely exchanged within the VPN. All content servers within a Content Smart VPN can be perceived as a unified logical content server for clients within and outside of the VPN. Thus, for any client request, any server within a content-aware VPN can serve as the proxy server for that content. The ability to segment a Content Smart network based on the notion of content autonomous system allows a hierarchical content structure be built that significantly reduces the amount of traffic that has to flow from one CAS to another. Content-aware devices that reside within a CAS communicate with each
20
Content Based Switching Techniques
4/26/2009
other in an effort to maintain consistent network states to facilitate distribution of contents and redirection of content requests.
8.4.
Content Switching
The most important technology foundation to build a Content Smart network is the ability to direct a content request to the best-fit server dynamically so that content can be automatically located, distributed to required access points, and replicated for redundancy and load balancing all without client knowledge. This requires that content be uniquely identified, which allows the network to dynamically resolve it to the network address of the best-fit server. Content switching differs from LAN switching (Layer 2) and IP switching (Layer 3) in one important aspect both LAN switching and IP switching forward information flows based on addresses. In the case of LAN switching, the MAC address and header information are used. In the case of IP switching, the destination IP address is used. Content-aware flow switching forwards information flows based on both the destination IP address, the associated TCP/UDP port number, and the URL/URI of the content. It uses the combination of content-induced QoS, server loading, and network path optimization to determine how to direct flows to the best fit server. With content switching, a server farm may more gracefully absorb a content request spike beyond the capacity of the farm by directing overflow capacity to other sites. This enables the emergence of a business model based on providing mirrored sites located in distributed data centers, with overflow content delivery capacity and backup in the case of a partial communications failure. Additionally, overflow content capacity intelligence minimizes the need to "build out" to handle sporadic releases of highly-requested content.
Content switching allows implicit detection of Web flows based on the arrival of TCP SYN or HTTP GET packets. It allows flow distribution based on embedded information in these flows that give implicit indications of their QoS, bandwidth,
21
Content Based Switching Techniques
4/26/2009
and content location requirements. The ability to distribute flows based on both the implicit content hints, server loads, and network conditions is unique in that it combines network route statistics, server metrics, and server locations to make the most optimal flow distribution decision. This combination of functionality is uniquely identified as "content-aware flow switching."
8.5.
Content Location Services
A Content Location Service performs server name resolution based on policy or configuration information associated with a client or stored in a directory service database. Dynamic content requires real-time updates to the Content Location Service much like when a DNS server is updated when an IP address is assigned to an existing domain name. Content Smart networks will facilitate the convergence of server-based directory services and network-based directory services into a new content-based directory service infrastructure that promises to integrate desktop, network, and server into Content Smart networks of the future.
8.6.
Content Usage Monitoring
Several important factors affect how frequently a content item is requested, when it is requested, and where it is requested. Localized content is highly specialized and is only of interest to the geographic areas it serves. On the other hand, content items that have broad appeal are often driven by real time events. In a Content Smart network, the ability for the network to be alerted before content-derived network congestion is about to occur is important. The ability to detect content congestion and redirect content requests to uncongested areas without client awareness form the foundation for a Content Smart network that can stand up to the true vision of electronic commerce.
8.7.
Content Management
Content management deals with distribution, replication, location, and versioning of contents. In a Content Smart network, content items may be distributed to clients directly from content servers or from intermediate proxy cache servers. Content is distributed based on network topological information, as well as target
22
Content Based Switching Techniques
4/26/2009
client groups, and can be customized into channels that are specifically designed to suit client needs. A distributed content server farm could consist of a number of geographically dispersed and local content servers. The locations of these servers, their contents, access paths, and access control rules are the type of information that should be centrally managed and controlled. This could take the form of a separate policy server or be integrated within the Content Location Service which functions as a directory service agent to access policy or configuration data.
23
Content Based Switching Techniques
4/26/2009
9.
Implementation of Content Based Switch
We tried to implement the content based switching on Linux operating system (kernel 2.4). We tried to find the path of the incoming IP packets, though we could not find exact places for intercepting IP packets. The most favorable places we felt for the interception of incoming IP packets (in kernel 2.4) are ip_rcv() subroutine in the ip_input.c, which is the main IP Receive routine ip_local_deliver() subroutine in the ip_input.c, where the incoming packet is sent to the higher level protocols. ip_fw_demasquerade() subrotinue in the ip_masq.c, where we can find the machine address to which it is being forwarded. Further research on this can be done in the future examining net filter of the Linux operating system's kernel- 2.4.
24
Content Based Switching Techniques
4/26/2009
10.
Summary
The advent of the Web has fundamentally changed how networks will be used. In this information age, content is king. Not only has it become the driving force behind recent innovations and advances in desktop, server, and networking technologies, it also has the potential of collapsing the functional boundaries among them. A desktop device is network-aware and a network device is contentaware. Today's networks are passive and content blind. The active networks of tomorrow will be Content Smart, converging from today's value-added Internet networks into proactive Content Smart networks. Content Smart networks proactively facilitate the distribution, replication, and management of content. Ultimately, the Content Smart network is just a smart agent for the optimum delivery of content. Once Content Smart networks understand the identity of clients and the contents they attempt to access, companies will no longer need artificial boundaries between internal users, partners, and WANs. Distinctions among Intranets, Extranets, and the Content Smart Internet will finally disappear, leaving one IPbased active network to handle all data communication for the company
25
Content Based Switching Techniques
4/26/2009
11.
References
1. Design, Implementation and Performance of a content-Based switch by George Apostolopoulos, David Aubespin, Vinod peris, Prashant Pradhan, Debanjan Saha 2. Linux open source code of Kernel-2.2, 2.4 3. http://linuxdocs.org/Adv-Routing-HOWTO.html 4. http://www.arrowpoint.com/solutions/white_papers/contentsmart.html 5. http://www.3com.com/technology/tech_net/white_papers/500660.html
26