Patent Application
20060126618
This invention generally relates to transmitting communications in networks. More specifically, it relates to transmitting communications efficiently within these networks.
The Transport Control Protocol/Internet Protocol (TCP/IP) is a suite of protocols used in Internet applications to route messages between different points in a network. One protocol from the suite is the Transport Control Protocol (TCP) protocol, which is responsible for controlling the movement of information between applications. Other examples of protocols in the suite include the User Datagram Protocol (UDP) and the Internet Protocol (IP).
As mentioned above, the TCP protocol specifies procedures for the exchange of data between applications such as between a remote client and a server. Before sending data from a server and a remote host using the TCP protocol, a TCP connection must first be established. In order to establish the connection, messages are exchanged between the server and the remote host. These may include synchronization (SYN) and acknowledgement (ACK) messages. Once the connection is established, data can be exchanged between the server and the remote host.
In previous systems, the client was unaware of the existence of multiple servers and a single IP address maps to a single entity. Since the same IP address corresponds to all possible destinations, an intermediate entity is typically used to distinguish between destinations and route the message to the appropriate destination. In many previous systems, a front end processor (FEP) is situated between the remote clients and the servers to modify the IP address in the packet so that the packet can be routed to the correct destination server.
Unfortunately, several problems occur in these previous systems because of the need to modify the contents of the packet at the front end processor. For instance, because the difficulty of performing the actions at the front end processor, line rates were limited. In addition, a time-consuming reassembly of the packets is required and the FEP needed to be aware of the application involved. Previous approaches also often require application-specific processing to process the IP addresses that are embedded in the IP packets as the packets are disassembled. Also, once encryption was used, it was difficult for the FEP to be able to modify the contents of the packets. All of these considerations may significantly increase the cost of the system and contribute to a slow-down in the processing of packets in the system.
Skilled artisans will appreciate that elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale. For example, the dimensions and/or relative positioning of some of the elements in the figures may be exaggerated relative to other elements to help to improve understanding of various embodiments of the present invention. Also, common but well-understood elements that are useful or necessary in a commercially feasible embodiment are often not depicted in order to facilitate a less obstructed view of these various embodiments of the present invention. It will also be understood that the terms and expressions used herein have the ordinary meaning as is accorded to such terms and expressions with respect to their corresponding respective areas of inquiry and study except where specific meanings have otherwise been set forth herein.
A system for automatically routing a message from a remote client to a server without modifying the contents of the message uses a front end processor to perform the routing. The front end processor utilizes a mapping relationship to translate a message identifier into an identifier corresponding to a particular destination server or servers. In this way, packets are quickly routed to a destination server without having to disassemble or modify the packets.
In many of these embodiments, messages are routed between one or more servers and a remote device. A message is received from the remote device at the front end processor. In one example, the remote device may have a remote Internet Protocol (IP) address that is included in the message. This remote IP server address is mapped to an address corresponding to a selected one of the servers. The message can then be routed to one (or more) of the servers using the server address. By using this approach, the packet and its contents remain unmodified.
The mapping of the remote IP address may map the IP address to a physical port number of the selected one of the plurality of servers. The packet may have a Layer 3 (L3) IP address. Other types of addressing schemes may also be used.
Thus, the present approaches avoid modifying the message or its contents. In these approaches, the message is quickly and efficiently mapped to a destination server or servers. The source of the message need not modify the message in any way since the front end processor performs all mapping and routing functions. Consequently, the time to route a message from a source to a destination is significantly reduced.
Referring now to
The servers 104, 106, and 108 comprise processor boards that manage and process messages according to programmed criteria. The servers 104, 106, and 108 may be identical or they may process the messages differently according to, for example, the type of message received or the source of the message.
The front end processor 110 may include any type of controller that processes ingress messages (i.e., messages received from the remote clients 114, 116, or 118) and egress messages (i.e., messages sent from one of the servers 104, 106, and 108 to one of the remote clients 114, 116, or 118). Communications between the server arrangement 102 and the remote clients 114, 116, and 118 is performed according to the TCP/IP protocols.
The remote clients 114, 116, and 118 may represent any number of applications or services. For example, the remote clients 114, 116, and 118 may be other servers and/or supply various types of services for the system.
In one example of the operation of the system shown in
The mapping relationship 111 may be any number of procedures and/or data structures that identify a destination server or servers based upon an identifier in the message. For example, the mapping relationship 111 may be a table that maps a range of addresses into a destination port number. An example of such a table is described with respect to
After the mapping relationship 111 is used to determine the identity of the destination server 104, 106, or 108, the message 112 is routed to the appropriate server. The IP address and the remaining contents of the message 112 are not changed or modified by the processing and routing performed at the front end processor 110. Consequently, the source of the message (i.e., remote client 116) only needs to know the IP address of the servers and does not need to know the detailed routing relationship between the source and the destination.
Referring now to
In the ingress direction (from remote client to server), the receiver/transmitter 210 receives a message, for instance, a data packet. The receiver/transmitter 210 examines the message for an identifier to indicate the destination of the message. In one example, the identifier may be an IP address. The controller 208 then retrieves the mapping relationship 204 that is stored in the memory 202 and applies it to the identifier. The result of the application is a destination identifier. For example, the destination identifier may be a port number of a destination server. The receiver/transmitter 206 is then used to transmit the message to the correct destination server. As described above, the contents of the message are preferably not modified or altered in any way.
In the egress direction (from server to remote client), the receiver/transmitter 206 receives a message. The message has an identifier that specifies the destination of the message. The controller 208 obtains the message and, since the message is an egress message, does not apply the mapping relationship 204 to the message. The message is sent to the receiver/transmitter 210 where it is forwarded to the appropriate destination, in this case, one (or more) of the client servers. Again, the contents of the message are not modified by the front end processor 200.
Referring now to
At step 306, the front end processor obtains a mapping relationship from a memory. Numerous examples of mapping relationships are possible. For example, as described below with respect to
At step 308, the front end processor applies the mapping relationship to the identifier to obtain the identity of the destination server. In one example, an IP address is applied to a lookup table and the result of the application gives a physical port number of the destination server. In another approach, a Layer 2 (L2) address may be used to choose the correct server while not modifying the Layer 3 (L3) address. At step 310, the packet is routed to the destination server using the identity that was determined using the mapping relationship.
Referring now to
The second column represents the identity of the server that a particular address range maps. For example, addresses between A1 and A2 map to port 1; addresses between A2 and A3 map to port 2; addresses between A3 and A4 map to port 3; and the addresses between A4 and A5 map to port 4. A controller in the front end processor obtains the table 400 from memory and uses the table to identify a port number or port numbers that correspond to an IP address.
Thus, modification of messages and/or their contents is avoided by the front end processor. The message, for example, the data packet, is quickly and efficiently mapped to a destination server or servers. Since the source of the message need take no actions or modify the message, the time needed to route a message from a source client to a destination server is significantly reduced.
Those skilled in the art will recognize that a wide variety of modifications, alterations, and combinations can be made with respect to the above described embodiments without departing from the spirit and scope of the invention, and that such modifications, alterations, and combinations are to be viewed as being within the ambit of the inventive concept.
