PTAB

IPR2017-00557

Teradata Operations Inc v. Realtime Data LLC

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Content Independent Data Compression Method and System
  • Brief Description: The ’867 patent describes a method for data compression on a block-by-block basis. The system determines whether to compress each data block using one of several available encoders, and if so, which encoder provides the best compression ratio. The output includes either the compressed data block with a corresponding "data compression type descriptor" or the original data block with a "null data compression type descriptor" if no compression is performed.

3. Grounds for Unpatentability

Ground 1: Anticipation by Hsu - Claims 16, 32, 34, and 35 are anticipated by Hsu

  • Prior Art Relied Upon: Hsu (W.H. Hsu, et al., Automatic Synthesis of Compression Techniques for Heterogeneous Files, Oct. 1995).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Hsu disclosed every limitation of independent claim 16. Hsu described a method for compressing "heterogeneous files" by treating a file as a collection of fixed-size data blocks. Hsu determined whether to compress each block by analyzing its data type and calculating "redundancy metrics." If a block’s metrics exceeded a threshold, it was deemed compressible, and an appropriate algorithm was selected. An "identifying tag" for the chosen algorithm was then recorded in a "compression plan." If the metrics were below the threshold, the block was deemed uncompressible, and a "skip" instruction was recorded. Petitioner asserted this "skip" instruction is a null data compression type descriptor, and the "identifying tag" is a data compression type descriptor, thus anticipating the core steps of claim 16.
    • Key Aspects: Petitioner argued Hsu also anticipated dependent claims by disclosing the use of variable-length blocks (claim 32), implementing compression algorithms as software modules in C on a Unix platform (claim 34), and using lossless algorithms like Lempel-Ziv and arithmetic coding (claim 35).

Ground 2: Anticipation by Franaszek - Claims 16-19, 32, 34, and 35 are anticipated by Franaszek

  • Prior Art Relied Upon: Franaszek (Patent 5,870,036).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner contended that Franaszek disclosed a system that tests samples of each data block with a plurality of compression mechanisms to select an optimal one. Based on a comparison to a threshold, the system decided whether to compress the block. If compressed, an identifier for the chosen method was stored in a "compression method description (CMD)" area associated with the block. Petitioner argued that Franaszek inherently disclosed a null descriptor, as its system stored some blocks in uncompressed format and its teaching that "each block of data includes a coding identifier" necessitated a way to identify these uncompressed blocks. Franaszek also disclosed transmitting data between a memory and a remote system, anticipating the "transmitting" and "receiving" limitations of claim 17.
    • Key Aspects: Franaszek was argued to anticipate dependent claims by expressly recognizing that input blocks may have a variable size (claim 32), disclosing encoders as "program code" executed by a microprocessor (claim 34), using lossless encoders like run-length and Lempel-Ziv (claims 18 and 35), and explicitly teaching that its compression tests could be performed in parallel (claim 19).

Ground 3: Obviousness over Hsu and Franaszek - Claims 17, 18, and 32 are obvious over Hsu in view of Franaszek

  • Prior Art Relied Upon: Hsu, Franaszek.

  • Core Argument for this Ground:

    • Prior Art Mapping: This ground was presented as an alternative, arguing that if Hsu was found not to disclose the "transmitting" and "receiving" limitations of claim 17, these elements were taught by Franaszek. Franaszek described a compressor and de-compressor that could be in remote systems, necessarily requiring the transmission and reception of compressed data and descriptors. Similarly, if Hsu was found not to teach variable-length blocks for claim 32, Franaszek explicitly disclosed that input block sizes could be variable.
    • Motivation to Combine: A POSITA would combine Hsu and Franaszek to solve the common problem of optimally compressing heterogeneous data blocks for network transmission. Both references operate in the same technical field. A POSITA would have been motivated to modify Hsu's system, which creates a compression history, with Franaszek's teachings on transmitting data between remote systems. This would be a predictable improvement to adapt Hsu's file-based system for efficient use over a network.
    • Expectation of Success: The combination would have yielded predictable results, as appending descriptors to data for transmission was a well-known technique. It would provide the decompressor with the necessary instructions for decompression in a straightforward manner.

  • Additional Grounds: Petitioner asserted additional obviousness challenges, including combining Hsu with Langdon, Jr. (Patent 5,097,261) to teach parallel encoder configurations (Ground 6) and various other combinations of Hsu, Franaszek, and Langdon, Jr. as alternative arguments.

4. Key Claim Construction Positions

  • "a data compression type descriptor": Petitioner adopted the patent's express definition, construing this term as "any recognizable data token or descriptor that indicates which data encoding technique has been applied to the data."
  • "a null data compression type descriptor": Petitioner also adopted the patent's definition for this term, construing it as "any recognizable data token or descriptor that indicates no data encoding has been applied to the input data block."

5. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 16-19, 32, 34, and 35 of the ’867 patent as unpatentable.