Pattern-driven, message-oriented compression apparatus and method

• US 7,321,322 B2
• Filed: 11/07/2005
• Issued: 01/22/2008
• Est. Priority Date: 05/08/2003
• Status: Active Grant

• Current Assignee: Sap Portals Israel Ltd (SAP SE)
• Original Assignee: Sap Portals Israel Ltd (SAP SE)
• Inventors: Drobinsky, Alex, Keren, Guy, Helfman, Nadav Binyamin
• Primary Examiner(s): Nguyen; Khai M.
• Application Number:
  US11/269,148
• Publication Number:
  US 20060139187A1
• Time in Patent Office: 806 Days
• Field of Search: 341/51, 341/87, 341/106
• US Class Current: 341/87
• CPC Class Codes: H03M 7/3088 employing the use of a dictionary, e.g. LZ78

Abstract

A compression and decompression method and apparatus comprising at least one data source providing a stream of data to at least one data destination, employing at least one pattern classifier processing the stream of data of the at least one data source into a single stream of messages and generating at least one pattern event, a message encoder and a message decoder changing an internal state in response to the at least one pattern event.

Claims

1. Within a computerized environment having at least one central processing unit for processing incoming data to perform data compression, a method comprising:

providing a stream of incoming data from at least one data source directed to at least one data destination;

processing the stream of incoming data into a single stream of messages;

analyzing the single stream of messages to detect one or more session events in the single stream of messages; and

compressing the single stream of messages with an encoder to generate a stream of encoded messages representing a compressed version of the incoming data stream, wherein the encoder is configured to encode the single stream of messages with respect to one or more internal states of the encoder associated with the one or more session events.

2. The method of claim 1, wherein the single stream of messages comprises content segments continuous in one or more of time, application layer or proximity.

3. The method of claim 1, further comprising matching messages from a store for most recent messages within the encoder internal data structure with strings stored in a string dictionary.

4. The method of claim 3, wherein matching comprises matching a hash value of a fixed size prefix within a matched context.

5. The method of claim 1, further comprising detecting a pattern event in the data stream using the pattern classifier.

6. The method of claim 5, wherein the pattern event comprises one or more of a silence in the session event or an end session event.

7. The method of claim 1, further comprising processing a badly compressed message segments store within the encoder into new dictionary strings in response to a silence in session event.

8. The method of claim 1 further comprising:

receiving the stream of encoded messages with a decoder; and

decompressing the stream of encoded messages into an uncompressed stream of messages comprising uncompressed message content of the single stream of messages, wherein the decoder is configured to decode the single stream of messages with respect to one or more internal states of the decoder associated with the one or more session events.

9. The method of claim 8, further comprising processing a badly compressed message segments store within the decoder into new dictionary strings in response to a silence in session event.

10. Within a computerized environment having at least one central processing unit for processing incoming data to perform data compression, an apparatus comprising:

at least one data source that provides a stream of incoming data directed to at least one data destination;

a pattern classifier for processing the stream of incoming data into a single stream of messages, wherein the processing comprises detecting one or more session events in the single stream of messages; and

an encoder configured to compress the single stream of messages into a stream of encoded messages representing a compressed version of the incoming data stream, wherein the encoder is configured to encode the single stream of messages with respect to one or more internal states of the encoder associated with the one or more session events.

11. The apparatus of claim 10, wherein the single stream of messages comprises content segments continuous in one or more of time, application layer or proximity.

12. The apparatus of claim 10, wherein an encoder internal data structure comprises:

at least one string dictionary, and

a store for most recent messages comprising at least one most recent message.

13. The apparatus of claim 12, wherein the at least one most recent message is matched with at least one string within the at least one string dictionary.

14. The apparatus of claim 10, wherein the pattern classifier is a pattern classifier that detects pattern events in the incoming data stream.

15. The apparatus of claim 14, wherein pattern events include one or more of a silence in a session event or an end session event.

16. The apparatus of claim 10, wherein the encoder further comprises a badly compressed message segments store for processing at lea one badly compressed message into at least one new dictionary string in response to a silence in session event.

17. The apparatus of claim 10, further comprising:

a decoder configured to receive the stream of encoded messages, wherein the decoder is configured to decompress the stream of encoded messages into an uncompressed stream of messages comprising uncompressed message content of the single stream of messages, and wherein the decoder is configured to decode the single stream of messages with respect to one or more internal states of the decoder associated with the one or more session events.

18. The apparatus of claim 17, wherein the decoder further comprises a badly compressed message segments store for processing at least one badly compressed message into at least one new dictionary string in response to a silence in session event.

19. Within a computerized environment having at least one central processing unit for processing incoming data to perform one of compression or decompression, a method comprising:

providing a stream of incoming data from at least one data source directed to at least one data destination;

processing the stream of incoming data using a pattern classifier into a single stream of messages;

generating one or more pattern events;

for compression, encoding messages using a message encoder that changes an internal state in response to the at least one pattern event;

for decompression, decoding messages using a message decoder that changes an internal state in response to the at least one pattern event; and

processing a badly compressed message segments store within the encoder or decoder into new dictionary strings in response to a silence in session event.

20. The method of claim 19, wherein the single stream of messages comprises content segments continuous in one or more of time, application layer or proximity.

21. The method of claim 19, further comprising matching messages from a store for most recent messages within the encoder internal data structure with strings stored in a string dictionary.

22. The method of claim 21, wherein matching comprises matching a hash value of a fixed size prefix within a matched context.

23. The method of claim 19, further comprising detecting a pattern event in the data stream using the pattern classifier.

24. The method of claim 23, wherein the pattern event comprises one or more of a silence in the session event or an end session event.

25. Within a computerized environment having at least one central processing unit for processing incoming data to perform one of compression or decompression, a method comprising:

providing a stream of incoming data from at least one data source directed to at least one data destination;

processing the stream of incoming data using a pattern classifier into a single stream of messages;

generating one or more pattern events;

for compression, encoding messages using a message encoder that changes an internal state in response to the at least one pattern event, wherein encoding messages includes matching messages from a store for most recent messages within an encoder internal data structure with strings stored in a string dictionary; and

for decompression, decoding messages using a message decoder that changes an internal state in response to the at least one pattern event.

26. Within a computerized environment having at least one central processing unit for processing incoming data to perform one of compression or decompression, an apparatus comprising:

at least one data source that provides a stream of incoming data directed to at least one data destination;

a pattern classifier for processing the stream of incoming data into a single stream of messages;

a pattern event generator for generating at least one pattern event;

a message encoder that changes an internal state in response to the at least one pattern event, the message encoder further comprising an encoder internal data structure comprising;

at least one string dictionary, and

a store for most recent messages comprising at least one most recent message; and

a message decoder that changes an internal state in response to the at least one pattern event.