PTAB

IPR2017-01660

NetApp Inc v. Realtime Data LLC

Log In
Key Events
Petition
petition Intelligence

1. Case Identification

2. Patent Overview

  • Title: Systems and methods for fast and efficient data compression.
  • Brief Description: The ’506 patent discloses a method for compressing data by first analyzing an incoming data block to recognize its type or content. Based on this recognition, the system selects either a specific, content-dependent compression encoder or, if no specific encoder is associated with the identified data type, a content-independent default encoder.

3. Grounds for Unpatentability

Ground 1: Obviousness over Hsu and Franaszek - Claim 105 is obvious over Hsu in view of Franaszek.

  • Prior Art Relied Upon: Hsu ("Automatic Synthesis of Compression Techniques for Heterogeneous Files," Oct. 1995) and Franaszek (Patent 5,870,036).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Hsu, a 1995 journal article, teaches a system that satisfies most limitations of claim 105. Hsu's system analyzes data blocks using statistical methods (e.g., redundancy metrics for run-length and string repetition) to determine a data type, a process that inherently excludes relying only on a descriptor (limitation 105[F]). It then selects a corresponding content-dependent compression algorithm from a database (limitation 105[D]). However, Petitioner contended that for "non-useful" data types where no algorithm is assigned, Hsu’s system performs no compression at all. Franaszek allegedly supplied the missing element of claim 105[E] by expressly teaching the use of a "default list of compression methods" (a default Compression Method List, or CML) when a specific data type for a block cannot be identified.
    • Motivation to Combine: A POSITA would be motivated to combine Franaszek’s default compression with Hsu’s analysis system to improve overall compression efficiency and achieve better space savings, a primary goal of any compression system. While Hsu’s choice to forego compression on unrecognized data types was an attempt to conserve computing resources, Franaszek taught that applying a default encoder in such situations was a beneficial and known alternative. This modification would address the shortcoming in Hsu where a data block that is actually compressible might be left uncompressed due to misclassification or being a new, unrecognized type.
    • Expectation of Success: Petitioner argued the combination was a predictable substitution of one known option (no compression) with another known, functionally equivalent option (default compression) to solve a known problem. A POSITA would have reasonably expected that implementing a default encoder from Franaszek into Hsu's system—for example, by using a simple "if-else" programming structure—would successfully compress additional data blocks and predictably improve the system's overall compression ratio without undue experimentation.

Ground 2: Obviousness over Franaszek and Hsu - Claim 105 is obvious over Franaszek in view of Hsu.

  • Prior Art Relied Upon: Franaszek (Patent 5,870,036) and Hsu ("Automatic Synthesis of Compression Techniques for Heterogeneous Files," Oct. 1995).

  • Core Argument for this Ground:

    • Prior Art Mapping: Petitioner asserted that Franaszek teaches the claim's basic framework: receiving a data block, selecting a compression method based on a "type field" descriptor, and using a default method if the type field is not available. This combination meets most claim limitations. However, Petitioner argued that Franaszek is silent on how to determine the data type if the "type field" descriptor is missing or unavailable. Hsu allegedly supplied the method for satisfying limitation 105[F]—analyzing the data within the block itself to identify a data type. Hsu specifically discloses analyzing 512-byte segments and calculating statistical metrics from the data content, a method that is not based only on a descriptor.
    • Motivation to Combine: A POSITA implementing Franaszek's system would be motivated to enhance its capabilities by incorporating a method to handle data blocks that lack a pre-defined type descriptor. Hsu provided a well-documented, known technique for identifying data types by directly analyzing data content. A POSITA would therefore be motivated to integrate Hsu's content-analysis techniques into Franaszek's framework to make the system more robust and capable of handling a wider variety of data streams, particularly those where metadata is absent.
    • Expectation of Success: Implementing Hsu's analysis method to populate or supplement Franaszek's "type field" was described as a straightforward application of a known data analysis technique to improve a known compression system. This would lead to the predictable and beneficial result of broader applicability for the Franaszek system, with a high expectation of success.

  • Additional Grounds: Petitioner asserted additional obviousness challenges based on similar reasoning. One ground argued claim 105 is obvious over Hsu in view of Sebastian (Patent 6,253,264), where Sebastian, like Franaszek, teaches using a "generic" default encoder for unsupported data formats. Another ground argued claim 105 is obvious over Franaszek in view of Chu (Patent 5,467,087), where Chu, like Hsu, teaches analyzing data bytes directly to identify the data type when a descriptor is missing.

4. Key Claim Construction Positions

  • Petitioner stated that for the purposes of the petition, it adopted constructions from a related district court case involving the ’506 patent. These constructions were asserted to be consistent with the prior art's teachings.
    • "analyzing"/"analyze": Construed as "means directly examining / directly examine." This construction supports the argument that references like Hsu and Chu teach the claimed analysis.
    • "default data compression encoder": Construed as "an encoder used automatically in the absence of a designated alternative." This construction was argued to be met by the teachings of Franaszek and Sebastian.

5. Key Technical Contentions (Beyond Claim Construction)

  • Priority Date: Petitioner contended that challenged claim 105 is not entitled to the asserted December 11, 1998, priority date of its earliest parent, the '491 application. Petitioner argued that the '491 application lacks written description support for the claimed method of analyzing a data block to determine whether to apply content-dependent versus content-independent compression. Petitioner asserted the earliest possible priority date is October 29, 2001, when this subject matter was first introduced into the application chain. While Petitioner noted that all prior art references relied upon predate even this later date, the argument challenges a key aspect of the patent's history.

6. Relief Requested

  • Petitioner requested that the Board institute an inter partes review of claim 105 of the ’506 patent and issue a final written decision cancelling the claim as unpatentable under 35 U.S.C. §103.