PTAB

IPR2017-00806

Teradata Operations Inc v. Realtime Data LLC

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Data Compression System and Method
  • Brief Description: The ’506 patent discloses a method for compressing data by analyzing an incoming data block to identify its data type. The system then performs content-dependent compression using a tailored encoder if a data type is identified, or performs content-independent compression using a single or default encoder if the data type is not identified.

3. Grounds for Unpatentability

Ground 1: Obviousness over Franaszek, Hsu, and Sebastian - Claims 104 and 105 are obvious over Franaszek in view of Hsu, or in the alternative, Franaszek in view of Hsu and Sebastian.

  • Prior Art Relied Upon: Franaszek (Patent 5,870,036), Hsu (a 1995 journal article titled Automatic Synthesis of Compression Techniques for Heterogeneous Files), and Sebastian (Patent 6,253,264).
  • Core Argument for this Ground:

    • Prior Art Mapping: Petitioner argued that the combination of references taught every limitation of the challenged claims. Franaszek was asserted to be the primary reference, teaching a method for compressing data blocks by first determining if a data type is known. If the type is known (e.g., from a "type field"), Franaszek applied a tailored encoder selected from a pre-defined "Compression Method List." If the data type was unavailable, Franaszek used a "default list of compression methods" to test samples of the data block and select the single best-performing encoder to compress the entire block. This base system disclosed the core claimed concept of selecting between a content-dependent encoder and a default/single encoder based on data type identification.

      Petitioner contended that the key limitation missing from Franaszek—"analyzing of the data within the data block... excludes analyzing based only on a descriptor"—was taught by Hsu. While Franaszek relied on a descriptor field, Hsu taught a more robust method for identifying data types in heterogeneous files by analyzing the actual content of the data block. Specifically, Hsu’s "new-file" procedure examined samples from the beginning, middle, and end of a data block and compared them to known patterns to determine the most applicable data type. This direct analysis of the data payload, rather than reliance on a simple descriptor, directly corresponded to the claim limitation.

      As an alternative for the "single data compression encoder" limitation, Petitioner cited Sebastian. Sebastian disclosed a system using data type-specific "filters" (encoders) but employed a single "generic" filter when an installed filter did not match the data format. Petitioner argued that if Franaszek’s process of selecting the best encoder from a default list was not considered a "single" encoder, Sebastian explicitly taught using one generic, content-independent encoder for unidentified data types.

    • Motivation to Combine: Petitioner asserted that a person of ordinary skill in the art (POSITA) would have been motivated to combine the teachings of these references for predictable results. A POSITA would combine Hsu's content-based analysis with Franaszek's framework to improve its robustness. The known problem of handling heterogeneous files, where data types might change within a file and descriptors could be unreliable, provided a strong reason to incorporate Hsu's superior data type identification method into Franaszek's system. This would lead to more accurate encoder selection and better overall compression.

      Furthermore, a POSITA would have been motivated to substitute Sebastian's single "generic" encoder for Franaszek's more complex default method (sampling and testing) to improve speed and efficiency. In scenarios where processing speed was more critical than achieving the absolute best compression ratio, replacing a time-consuming sampling process with a single, known-good generic encoder was a well-understood design trade-off.

    • Expectation of Success: Petitioner argued that a POSITA would have had a reasonable expectation of success in making these combinations. The references addressed similar problems in the same field of data compression. Integrating Hsu’s data analysis module into Franaszek’s system or substituting Sebastian’s generic encoder were straightforward applications of known techniques to achieve the predictable benefits of improved accuracy and speed, respectively.

4. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 104 and 105 of the ’506 patent as unpatentable.