PTAB
IPR2017-00108
Oracle America Inc v. Realtime Data LLC
Key Events
Petition
1. Case Identification
- Case #: IPR2017-00108
- Patent #: 9,054,728
- Filed: October 24, 2016
- Petitioner(s): Oracle America, Inc.
- Patent Owner(s): Realtime Data LLC
- Challenged Claims: 1 and 17
2. Patent Overview
- Title: Data Compression Systems and Methods
- Brief Description: The ’728 patent relates to systems for data compression that analyze data blocks to identify their content or other attributes. Based on the analysis, the system routes a data block to either a content-specific "content dependent" encoder or, if the content is not identified, to a generic "content independent" single encoder.
3. Grounds for Unpatentability
Ground 1: Obviousness over Hsu and Sebastian - Claims 1 and 17 are obvious over Hsu in view of Sebastian.
- Prior Art Relied Upon: 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 Hsu discloses a system that meets nearly every limitation of the challenged claims. Hsu’s system for compressing "heterogeneous files" analyzes data within fixed-size blocks to identify parameters such as data type and redundancy metrics. Based on this analysis, the system selects an optimal, content-dependent compression algorithm from a database of encoders. Petitioner contended that Hsu's analysis of the first, middle, and last 512 bytes of data within a block—rather than relying on an appended descriptor—satisfies the claim limitation requiring analysis of the data itself. However, Hsu teaches that if a block is deemed "uncompressible," a 'skip' instruction is recorded, rather than applying a default compression. Petitioner asserted that Sebastian remedies this deficiency by expressly teaching the use of a "single data compression encoder" for unrecognized data types. Sebastian describes a system with data type-specific "filters" (encoders) alongside a single "generic" filter (such as a Lempel-Ziv variant) that is used when no specific filter matches the data format.
- Motivation to Combine: Petitioner argued a person of ordinary skill in the art (POSITA) would combine Hsu and Sebastian to solve the common problem of optimally compressing a collection of data blocks with varied content. Hsu's system was designed to be extensible, and a POSITA would have recognized the clear benefit of improving it. Instead of merely skipping unrecognized or uncompressible blocks as Hsu teaches, a POSITA would have integrated Sebastian's "generic" filter to handle such data. This would be a predictable modification to better achieve Hsu’s stated goal of providing optimal compression for heterogeneous files.
- Expectation of Success: A POSITA would have a reasonable expectation of success because the combination involved applying a well-known, universal lossless compression algorithm (as taught by Sebastian) to unclassified data, a common and predictable technique in the art of data compression.
Ground 2: Obviousness over Hsu and Wang - Claims 1 and 17 are obvious over Hsu in view of Wang.
- Prior Art Relied Upon: Hsu (as described above) and Wang (International Publication No. WO 00/46688).
- Core Argument for this Ground:
- Prior Art Mapping: The core argument is similar to Ground 1, with Wang serving as the secondary reference. Petitioner asserted that Hsu provides the primary system for analyzing data blocks and selecting from multiple content-dependent encoders. Wang was presented as supplying the teaching of a "single data compression encoder" for data whose format cannot be identified. Wang discloses an "intelligent method" that first attempts to identify a file's format. If the system fails to identify the format, it automatically compresses the file using a single lossless encoder, such as LZW (Lempel-Ziv-Welch). This default application of a single encoder to unidentified data directly corresponds to the functionality claimed in the ’728 patent.
- Motivation to Combine: Petitioner asserted that the motivation to combine Hsu and Wang was strong, as both references are in the same technical field and address the same problem. A POSITA would have found it obvious to enhance Hsu's sophisticated, extensible analysis system with Wang’s simple and effective method for handling unrecognized data. Modifying Hsu’s system to apply Wang’s default lossless LZW encoder to unidentifiable data blocks would be a straightforward improvement to ensure all data is compressed, rather than skipped, thereby increasing overall compression efficiency.
- Expectation of Success: Success would be predictable and expected. The proposed modification merely integrates a standard, universal lossless compression technique (Wang's LZW encoder) into a system designed for adaptive compression (Hsu), a well-understood combination of known elements to achieve a predictable result.
4. Key Claim Construction Positions
- "single data compression encoder": Petitioner proposed this term be construed as "a data compression encoder associated with one compression algorithm to compress input data without regard to the encoder's ability to effectively encode the data type or content of the data block." This construction was argued to be supported by the prosecution history of an ancestor patent, where the term was added to overcome a double patenting rejection and distinguish the invention from systems using a "list of compression methods." The term was used interchangeably with "content independent data compression."
- "wherein the analyzing of the data within the data block ... excludes analyzing based solely on a descriptor ...": Petitioner proposed this negative limitation means that "the analyzing ... includes analyzing data other than data that is appended to the data block and indicative of the ... parameters or attributes of the data." This construction was based on the examiner's reasoning during prosecution of an ancestor patent, which distinguished the claims from prior art that determined data type only by reviewing descriptive information (e.g., a "type field") appended to a data block.
5. Relief Requested
- Petitioner requested institution of an inter partes review and cancellation of claims 1 and 17 of Patent 9,054,728 as unpatentable.