PTAB
IPR2018-00656
RIVERBED TECHNOLOGY, INC. v. Realtime Data LLC
1. Case Identification
- Case #: IPR2017-01710, IPR2017-02129, and IPR2018-00612 are referenced as related matters before the Board.
- Patent #: Patent 8,717,204
- Filed: February 15, 2018
- Petitioner(s): Riverbed Technology, Inc.
- Patent Owner(s): Realtime Data LLC
- Challenged Claims: 1-30
2. Patent Overview
- Title: METHODS FOR ENCODING AND DECODING DATA
- Brief Description: The ’204 patent relates to systems and methods for data compression to accelerate data transmission. The core technology involves recognizing a characteristic of data residing in data fields, selecting a specific encoder based on that characteristic, and compressing the data.
3. Grounds for Unpatentability
Ground 1: Obviousness over XMill - Claims 12–18, 20, and 21 are obvious over XMill.
- Prior Art Relied Upon: XMill (Hartmut Liefke & Dan Suciu, XMill: an Efficient Compressor for XML Data, 2000).
- Core Argument for this Ground:
- Prior Art Mapping: Petitioner asserted that XMill, a system for compressing XML data, teaches the limitations of independent claim 12. XMill processes data in fields (from a web log file), recognizes a data characteristic (its XML path), selects an encoder ("semantic compressor") based on that path, and compresses the data using state machines ("deterministic automatons") to achieve compression ratios (e.g., 29:1) far exceeding the claimed "over 4:1" ratio. XMill also discloses point-to-point data transmission that is faster than transmitting the data in uncompressed form.
- Key Aspects: Petitioner noted that these arguments mirror those the Board adopted when cancelling a nearly identical claim in a related reexamination of a sibling patent.
Ground 2: Obviousness over Ferris in view of XMill - Claims 1–8, 10, 11, 16, 17, 22–28, and 30 are obvious over Ferris in view of XMill.
- Prior Art Relied Upon: Ferris (WO 02/13058) and XMill.
- Core Argument for this Ground:
- Prior Art Mapping: Petitioner argued that XMill teaches the core "recognizing," "selecting," and "compressing" limitations of independent claims 1 and 22. Ferris, which discloses broadcasting compressed XML data to a "multiplicity of receivers," supplies the "broadcasting the compressed data to a plurality of clients" limitation absent from XMill's point-to-point disclosure.
- Motivation to Combine: A POSITA would combine the general-purpose XML compression of XMill with the XML broadcasting system of Ferris to achieve the "efficient use of bandwidth" explicitly sought by Ferris.
- Expectation of Success: The combination was predictable and would have presented a high expectation of success, as both references utilize standard XML, making the application of XMill's known compression tool to Ferris's broadcasting system a straightforward implementation.
Ground 3: Obviousness over XMill in view of Network Protocols - Claims 14, 15, and 19 are obvious over XMill in view of RFC768 and/or RFC1180.
Prior Art Relied Upon: XMill, RFC768 (User Datagram Protocol), and RFC1180 (A TCP/IP Tutorial).
Core Argument for this Ground:
- Prior Art Mapping: Petitioner argued that while XMill alone likely renders the claims obvious, the RFCs explicitly teach the networking details. RFC1180 describes the structure of an IP packet (used by XMill's "rcp" application) as containing multiple messages, satisfying limitations in claims 14 and 15. For claim 19's User Datagram Protocol (UDP) requirement, Petitioner asserted it would have been obvious to substitute XMill's TCP-based transfer with a UDP-based one like TFTP (taught in RFC768), as both were well-known file transfer mechanisms.
- Motivation to Combine: A POSITA would be motivated to combine XMill with the protocols in the RFCs because XMill explicitly uses a network application ("rcp") that operates according to these fundamental internet protocols. The choice between TCP and UDP was a simple design choice between a finite number of well-known options.
Additional Grounds: Petitioner asserted that claims 9 and 29 are obvious over XMill and Ferris in further view of Comer (a textbook on TCP/IP client-server programming). Petitioner also presented a contingent anticipation ground, arguing that if the Board finds the '204 patent's priority chain was broken, then an intervening 2003 CIP application becomes prior art that anticipates claims 1-30.
4. Key Claim Construction Positions
- "data packet" (claims 4, 14, 22): Petitioner argued for the broadest reasonable interpretation, asserting a "data packet" is simply a segregation of data and does not require a specific internal structure (e.g., a header or control information). This position is based on the patent's use of conditional language ("If there is internal structure to packets...") and is directly supported by the Board's rejection of a narrower construction in a reexamination of a sibling patent.
- "recognizing" / "analyzing" (claims 1, 11, 12, 21): Petitioner contended these terms require no special definition beyond their ordinary meaning. This opposes the Patent Owner's prior attempts in a related case to narrow the terms to mean "directly analyzing" or an "active step"—arguments the Board previously rejected.
5. Key Technical Contentions (Beyond Claim Construction)
- Invalid Priority Claim due to Lack of Enablement: Petitioner contended claims 1-30 are not entitled to any priority date before the patent's own filing date (September 24, 2013). The argument is that the unbounded compression ratios (e.g., "over 10:1") are not enabled by the parent specifications, which only disclose achieving specific, lower ratios. Citing MagSil Corp., Petitioner argued that an open-ended range must be enabled across its full scope, which the specification fails to provide. This alleged break in the priority chain makes several references available as prior art.
- Invalid Priority Claim due to New Matter: As an alternative, Petitioner argued that the claims cannot have a priority date before May 7, 2003, because key concepts like a "data field specific compiler" constituted new matter introduced in the 2003 CIP application, severing the priority chain to any earlier applications.
6. Arguments Regarding Discretionary Denial
- Petitioner argued that this inter partes review (IPR) is not redundant of the original prosecution and should not be denied under §325(d). Although XMill was cited by the Patent Owner during prosecution, the examiner never applied it in a rejection; instead, the patent received a first-action allowance. This petition presents new arguments, different prior art combinations, and significant new evidence not before the examiner, including the Board's FWD in a related reexamination and a new expert declaration.
7. Relief Requested
- Petitioner requested institution of an IPR for claims 1–30 of the ’204 patent and a final judgment cancelling the claims as unpatentable.