Internet protocol optimizer

• US 8,553,572 B2
• Filed: 09/09/2004
• Issued: 10/08/2013
• Est. Priority Date: 09/10/2003
• Status: Expired due to Fees

• Current Assignee: Network Executive Software, Inc.
• Original Assignee: Hyperdata Technologies, Inc.
• Inventors: Farrell, Richard S., Reiland, David J., Signorelli, James R.
• Primary Examiner(s): Hailu, Kibrom T
• Assistant Examiner(s): Beyen, Zewdu A
• Application Number:
  US10/571,189
• Publication Number:
  US 20070110046A1
• Time in Patent Office: 3,316 Days
• Field of Search: 370/338, 370/912, 370/902, 370/446, 370/463, 370/535, 370/401, 370/402, 370/469, 709/222, 709/225
• US Class Current: 370/252
• CPC Class Codes: H04L 67/14 Session management for real-time applications in data packet communications networks H04L65/1066
• CPC Class Codes: H04L 67/146 Markers for unambiguous identification of a particular session, e.g. session cookie or URL-encoding
• CPC Class Codes: H04L 69/04 Protocols for data compression, e.g. ROHC
• CPC Class Codes: H04L 69/16 Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
• CPC Class Codes: H04L 69/161 Implementation details of TCP/IP or UDP/IP stack architecture; Specification of modified or new header fields
• CPC Class Codes: H04L 69/163 In-band adaptation of TCP data exchange; In-band control procedures
• CPC Class Codes: H04L 69/165 Combined use of TCP and UDP protocols; selection criteria therefor
• CPC Class Codes: H04L 69/32 Architecture of open systems interconnection [OSI] 7-layer type protocol stacks, e.g. the interfaces between the data link level and the physical level
• CPC Class Codes: H04L 69/326 in the transport layer [OSI layer 4] H04L69/16 takes precedence

Abstract

A method for optimizing the throughput of TCP/IP applications by aggregating user application data and consolidating multiple TCP/IP connection streams into a single optimized stream for delivery to a destination application. Optimization of the internet protocol uses a packet interceptor to intercept packets from a source application, a packet driver to aggregate the intercepted packets, a data mover to transport the aggregated packets to another data mover at the destination, a destination packet driver to disaggregate the transported aggregated packets, and a destination end processor to deliver the disaggregated IP packets to the destination application.

Claims

1. A method of transport protocol optimization of an internet protocol for efficiently moving large amounts of data, comprising the steps of:

(a) using a source packet interceptor at a source network location to intercept an IP packet from a source application, the source packet interceptor examines an IP header of the IP packet to determine if it is an IP packet to be intercepted,

(b) using a source edge process at the source network location to act as the new destination for the source application resulting in terminating the connection between the source application and a destination application such that the source edge process becomes the new endpoint for the connection from the source application, and generating transport protocol acknowledgment messages with the source edge process and sending them to the source application, wherein such transport protocol acknowledgement messages would otherwise be generated on behalf of the destination application,

(c) using a source packet driver at the source network location to aggregate the intercepted IP packets from the source application while retaining all IP header routing information for the intercepted IP packets,

(d) using a source compression engine at the source network location to compress the aggregated IP packets,

(e) using a source data mover at the source network location to transport the compressed and aggregated IP packets over a communication link to a destination data mover at a destination network location,

wherein the source packet interceptor, source edge process, source packet driver, source compression engine and source data mover are integrated into a source transport protocol optimization system and the source transport protocol system is attached to a source LAN, the packets from the source application transported over the source LAN to the source transport protocol system;

(f) using a destination compression engine at the destination network location to decompress the compressed and aggregated IP packets,

(g) using a destination packet driver at the destination network location to disaggregate the transported aggregated packets, and

(h) using a destination edge process at the destination network location to deliver the disaggregated IP packets to the destination application;

wherein the destination compression engine, destination packet driver, destination edge process are integrated into a destination transport protocol optimization system and the destination transport protocol system is attached to a destination LAN, the packets from the destination transport protocol optimization system transported over the destination LAN to the destination application.

2. The transport protocol optimization method of claim 1, comprising the step of using IP routing.

3. The transport protocol optimization method of claim 1, the IP packet comprising a TCP, UDP, ICMP, or other type of IP packet.

4. The transport protocol optimization method of claim 1, the step of intercepting an IP packet from the source application comprises the steps of comparing the IP packet'"'"'s address to packet addresses in a look-up table and (b) intercepting only those source packets with the same addresses as those stored in the look-up table.

5. The transport protocol optimization method of claim 4 comprising the step of modifying the destination address of the IP packets accepted for interception to be the address of the source packet interceptor.

6. The transport protocol optimization method of claim 1, the address of the IP packet comprises the packet'"'"'s source IP address, source port number, destination IP address, destination port number, and protocol type.

7. The transport protocol optimization method of claim 1, the step of intercepting an IP packet from the source application comprises the step of routing the IP packet to an edge process that is unique to the address of the IP packet.

8. The transport protocol optimization method of claim 7, comprising the step of combining a routing header field, a message header field, and the intercepted IP packet data from the edge process.

9. The transport protocol optimization method of claim 7, comprising the step of creating a edge process for each TCP application connection;

a UDP edge process for each UDP intercept; and

a ICMP edge process for a ICMP intercept.

10. The transport protocol optimization method of claim 1, the step of intercepting an IP packet from the source application comprises the steps of a source edge process (a) reading the data contained in the routed IP packets and (b) forming a message header field for the routed IP packets.

11. The transport protocol optimization method of claim 10, the message header comprises a version field, a length of header field, a message function type field, a message flag field, a protocol type field, a sequence number field, a source IP address field, a destination IP address field, a source IP port number field, a destination IP port number field, a length of data field, and a status field.

12. The transport protocol optimization method of claim 1, comprising the step of the packet driver forming a packet driver message.

13. The transport protocol optimization method of claim 12, the packet driver message comprises the message header field and intercepted IP packet data from the source edge process.

14. The transport protocol optimization method of claim 13, comprising the step of forming a plurality of packet driver messages.

15. The transport protocol optimization method of claim 14, comprising the step of aggregating multiple packet driver messages into a packet driver buffer.

16. The transport protocol optimization method of claim 15, the size of the aggregated packet driver messages is less than or equal to a predetermined maximum size of the buffer.

17. The transport protocol optimization method of claim 16, comprising the step of the packet driver forming a routing header in the packet driver buffer that precedes a first packet driver message.

18. The transport protocol optimization method of claim 17, the routing header comprises a function type field, a number of packet driver messages field, and a data length field.

19. The transport protocol optimization method of claim 1, comprising the step of routing the aggregated packet driver messages to the source data mover.

20. The transport protocol optimization method of claim 19, further comprising transmitting a packet driver buffer over a communication link comprising one or more of the steps of (a) inserting data mover fields into the start of the packet driver buffer;

(b) reducing the size of the packet driver buffer by breaking the buffer into multiple segments, with each segment being no greater than the size specified in the configuration file;

(c) using standard UDP socket calls to interface with the TCP stack for UDP delivery of the segments over the network.

21. The transport protocol optimization method of claim 20, the communication link comprising a TCP, UDP, or other TCP/IP link.

22. The transport protocol optimization method of claim 21, the TCP, UDP, or other TCP/IP link for transporting the stored packets is over a WAN.

23. The transport protocol optimization method of claim 20, the data mover protocol comprising (a) data mover transport data subfield, and (b) data mover transport acknowledgement subfield.

24. The transport protocol optimization method of claim 23, the data mover transport data subfield comprising the length of the entire subfield, the subfield type code, the logical sequence number of the packet driver buffer or packet drive buffer segment, and the physical sequence number of the packet driver buffer or packet driver buffer segment.

25. The transport protocol optimization method of claim 23, the data mover transport acknowledgement subfield comprising the length of the entire subfield, the subfield type code, the highest physical block number sent from the source packet driver to the destination packet driver over a communication link, the highest physical block number received by the source packet driver that was sent by the destination packet driver over a communication link, the bit-significant flags representing the blocks received, and the rate of data delivery to the destination packet driver.

26. The transport protocol optimization method of claim 1, the packets intercepted by an operating system exit point.

27. The transport protocol optimization method of claim 1, comprising the step of opening a connection between a source application and a destination data application.

28. The transport protocol optimization method of claim 27, comprising the steps of (a) opening a connection between the source application and the source edge processor and (b) opening a connection between the destination edge processor and the destination application.

29. The transport protocol optimization method of claim 28, comprising the steps of (a) transporting packets from the source application to the source packet interceptor over a source LAN and (b) transporting packets delivered to a destination data mover to a destination application over a destination LAN.

30. A method of internet protocol optimization and efficiently moving large amounts of data, comprising the steps of:

(a) using a source packet interceptor at a source network location to intercept IP packets identified in a look-up table as having a specified source address, source port number, destination address, destination port number, and protocol type;

(b) terminating the connection between the source application and a destination application and generating and sending acknowledgement messages from a source edge process to the source application that would otherwise be generated and sent on behalf of a destination application;

(c) using a source packet driver at the source network location to encapsulate the IP packets into a packet driver message, and to aggregate packet driver messages while retaining all IP header routing information for the intercepted IP packets;

(d) using a source compression engine at the source network location to compress the aggregated packet driver messages;

(e) using a source data mover at the source network location to route via IP routing the aggregated packet driver messages to a destination data mover at a destination network location over a UDP communication link;

(f) using a destination compression engine at the destination network location to decompress the aggregated packet driver messages;

(g) using a destination packet driver at the destination network location to disaggregate the packet driver messages and to dis-encapsulate the IP packets; and

(h) using a destination edge process at the destination network location to deliver the dis-encapsulated IP packets to their respective destinations;

wherein the source packet interceptor, source edge process, source packet driver, source compression engine and source data mover are integrated into a source transport protocol optimization system and the source transport protocol system is attached to a source LAN, the packets from the source application transported over the source LAN to the source transport protocol system; and

wherein the destination compression engine, destination packet driver, destination edge process are integrated into a destination transport protocol optimization system and the destination transport protocol system is attached to a destination LAN, the packets from the destination transport protocol optimization system transported over the destination LAN to the destination application.