PTAB
IPR2017-00179
Dell Inc v. Realtime Data LLC
Key Events
Petition
1. Case Identification
- Case #: IPR2017-00179
- Patent #: 9,054,728
- Filed: November 14, 2016
- Petitioner(s): Dell Inc., EMC Corporation, Hewlett-Packard Enterprise Co., and HP Enterprise Services, LLC
- Patent Owner(s): Realtime Data LLC d/b/a IXO
- Challenged Claims: 1-10, 15, 20, and 24
2. Patent Overview
- Title: Data Compression Systems and Methods
- Brief Description: The ’728 patent discloses a data compression system that uses a combination of content-dependent and content-independent compression methods. The system analyzes an incoming data block to identify its parameters or attributes and, if identified, selects a corresponding content-dependent encoder; otherwise, it applies a content-independent encoder.
3. Grounds for Unpatentability
Ground 1: Obviousness over Franaszek, Hsu, and Sebastian - Claims 1-3, 9, 10, 15, 20, and 24 are obvious over Franaszek in view of Hsu, with Sebastian providing an alternative teaching.
- Prior Art Relied Upon: Franaszek (Patent 5,870,036), Hsu (a 1995 journal article on compression synthesis), and Sebastian (Patent 6,253,264).
- Core Argument for this Ground:
- Prior Art Mapping: Petitioner argued that Franaszek taught the fundamental architecture of the claimed system: a compression system that selects from a list of content-specific encoders if a data type is known (e.g., from a type field) and selects from a "default list" of encoders if the data type is unknown. However, the key limitation of claim 1—analyzing data within a block to identify its attributes while excluding analysis based solely on a descriptor—was taught by Hsu. Hsu disclosed a system that determines a data block’s type and compressibility by analyzing samples from the beginning, middle, and end of the block itself, as well as calculating "redundancy metrics" based on the data's content. Petitioner contended that implementing Hsu's content-based analysis within Franaszek's framework would render claim 1 obvious. For the "single data compression encoder" limitation, Petitioner argued that Franaszek’s process of selecting the single best-performing encoder from its default list met the limitation. Alternatively, Sebastian explicitly taught using a single "generic" filter (encoder) when a data type is not matched by a format-specific filter.
- Motivation to Combine: A POSITA would combine Franaszek and Hsu to improve the robustness of Franaszek's system. While Franaszek relied on a simple descriptor field, Hsu provided a more sophisticated method for handling heterogeneous files, where data types might vary within a single block. A POSITA would have been motivated to incorporate Hsu’s content analysis to make more accurate and efficient encoder selections, a predictable improvement. The motivation to substitute Sebastian's single generic encoder would be to increase speed and simplify the process compared to Franaszek's more complex method of testing multiple default encoders.
- Expectation of Success: A POSITA would have reasonably expected success in combining these references, as it involved applying Hsu’s known data analysis techniques to improve the functionality of a known compression architecture like Franaszek's. The outcome—a more accurate compression system—was predictable.
Ground 2: Obviousness over Franaszek, Hsu, and Aakre - Claims 4-8 are obvious over Franaszek in view of Hsu and further in view of Aakre.
- Prior Art Relied Upon: Franaszek (Patent 5,870,036), Hsu (a 1995 journal article), and Aakre (Patent 4,956,808). The alternative combination with Sebastian is also asserted.
- Core Argument for this Ground:
- Prior Art Mapping: This ground incorporated all arguments from Ground 1 and added Aakre to address the "real-time" compression limitations recited in dependent claims 4-8. Petitioner asserted that Franaszek was silent on real-time operation. Aakre, however, explicitly disclosed a "real time data transformation and transmission apparatus" that compresses data "as fast as the tape accepts the data." Aakre's system used a controller and buffer to overlap the compression of one data block with the writing of the previously compressed block, enabling a continuous "streaming mode" to a storage medium like a tape drive.
- Motivation to Combine: A POSITA would combine Aakre's teachings with the Franaszek/Hsu system to solve a well-known problem in data storage systems: latency caused by starting and stopping a storage medium (like a tape drive) between data blocks. By incorporating Aakre's real-time controller and buffering solution, the combined system could stream compressed data efficiently to a second memory, improving performance. This was merely applying a known solution to a known problem for its intended purpose.
- Expectation of Success: A POSITA would have had a high expectation of success, as integrating a buffering and control system for real-time data streaming was a well-understood technique for improving throughput in storage systems. The result would be a predictably more efficient system.
4. Key Claim Construction Positions
- "the data block being included in one or more data blocks" (claims 2 and 3): Petitioner argued this phrase should be construed to mean the data block is "included among a group of one or more data blocks." Petitioner contended that a literal reading—where one block is contained within another block—is inconsistent with the patent's description of a "data stream comprising one or more data blocks" being processed sequentially. The proposed construction was argued to be necessary to align the claim language with the specification.
5. Relief Requested
- Petitioner requested the institution of an inter partes review and the cancellation of claims 1-10, 15, 20, and 24 of the ’728 patent as unpatentable.