PTAB

IPR2019-01033

Google LLC v. Realtime Adaptive Streaming LLC

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Data Compression/Decompression Based on Throughput
  • Brief Description: The ’046 patent discloses a system for adaptive data compression. The system monitors the throughput of a data processing system and dynamically selects a compression routine from a plurality of available routines to increase throughput and eliminate data transmission bottlenecks.

3. Grounds for Unpatentability

Ground 1: Obviousness over Kiel - Claims 1 and 23 are obvious over Kiel.

  • Prior Art Relied Upon: Kiel (Patent 5,276,898).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Kiel discloses all limitations of claims 1 and 23. Kiel teaches a data communications system that dynamically controls compression to reduce bottlenecks by monitoring system load. Specifically, Kiel discloses a method of tracking a "number of queued tasks" for a processor and comparing it to a predetermined "task threshold value." This, Petitioner asserted, corresponds directly to the ’046 patent's limitation of "tracking throughput" by "tracking a number of pending requests for data transmission" and comparing it to a threshold. Furthermore, Kiel discloses an embodiment that selects between two different compression algorithms: a faster one requiring less processor time and a slower one. The system selects the faster algorithm when processor utilization is high (analogous to when throughput is low) to improve performance. This maps to the claimed step of switching to a second, faster compression routine when tracked throughput falls below a predetermined threshold.
    • Motivation to Combine (N/A - Single Reference): Petitioner contended that Kiel itself teaches all claimed elements. For any minor variations, Petitioner argued a person of ordinary skill in the art (POSITA) would have been motivated to modify Kiel's on/off compression scheme to use its disclosed two-algorithm scheme in all scenarios. The motivation was to improve communications performance by reducing the amount of transmitted data, rather than turning compression off entirely when the system is busy, thereby reducing bottlenecks—a predictable result.
    • Expectation of Success: A POSITA would have had a high expectation of success as it involved applying known compression techniques taught within Kiel to achieve the predictable outcome of improved system performance.

Ground 2: Obviousness over Kobata in view of Kiel - Claims 1 and 23 are obvious over Kobata in view of Kiel.

  • Prior Art Relied Upon: Kobata (WO 99/26130) and Kiel (Patent 5,276,898).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Kobata teaches a system that adaptively selects from a plurality of compression algorithms to optimize data transmission. Kobata's system initially uses the algorithm with the highest degree of compression and then measures the ratio of transmission time to compression time. If this ratio (a proxy for throughput) falls below a threshold of 1, the system switches to an algorithm with a lower compression degree (i.e., a faster compression rate) to improve performance. While Kobata teaches the core concept of adaptively switching to a faster algorithm based on a measured performance metric, Petitioner asserted it does not explicitly disclose tracking a "number of pending requests for data transmission." Kiel, as detailed in Ground 1, supplies this missing element by teaching a system that monitors a queue of pending tasks to determine when to adjust compression.
    • Motivation to Combine: A POSITA would combine these references because they both address the same problem of optimizing data transmission via adaptive compression. Petitioner argued a POSITA would have been motivated to replace Kobata's indirect throughput metric (the ratio of send time to compression time) with Kiel's more direct and improved method of monitoring a queue of pending transmission tasks. A POSITA would recognize that a backlog of pending requests, as taught by Kiel, is a direct indicator of a system bottleneck, providing a superior trigger for switching compression algorithms.
    • Expectation of Success: A POSITA would have reasonably expected success in this combination, as it involved substituting one known technique for measuring system load (Kobata's ratio) with another known technique (Kiel's task queue) to achieve the predictable result of a more responsive and efficient adaptive compression system.

4. Key Claim Construction Positions

  • "compression rate": Petitioner proposed this term means "execution speed of a compression algorithm." This construction was argued to be consistent with the claims, which require using a second routine with a "greater" compression rate to "increase the throughput," implying faster execution.
  • "throughput": Petitioner proposed this term means "bandwidth, i.e. amount of data per unit time." This was supported by the patent specification, which uses "throughput" and "bandwidth" interchangeably, and by the claim language tying throughput to tracking pending requests for data transmission.

5. Arguments Regarding Discretionary Denial

  • Petitioner argued that discretionary denial under §314(a) or §325(d) would be inappropriate. The petition asserted this was Google's first challenge to the ’046 patent. While a parallel inter partes review (IPR) existed (IPR2019-00209 filed by Netflix), Petitioner argued the grounds in this petition were substantially different, notably including a single-reference obviousness ground. Furthermore, the prior art references relied upon (Kiel and Kobata) were never considered by the patent examiner during prosecution. Petitioner also noted that the parties in co-pending district court litigation had submitted a joint stipulation to stay the case pending the outcome of the IPR.

6. Relief Requested

  • Petitioner requests institution of IPR and cancellation of claims 1 and 23 of the ’046 patent as unpatentable.