PTAB

IPR2019-00209

Netflix Inc v. Realtime Adaptive Streaming LLC

Log In

1. Case Identification

2. Patent Overview

  • Title: Data Compression with Dynamic Algorithm Selection
  • Brief Description: The ’046 patent describes methods and systems for dynamically switching between different compression algorithms. The selection is based on tracking the throughput of a data processing system and switching to a routine with a faster compression rate when the system throughput falls below a predetermined threshold.

3. Grounds for Unpatentability

Ground 1: Obviousness over Imai and Couwenhoven

Claims 1, 4, 8, 10-11, 13-14, 16-17, 19, 21, 23, 26-27, 29-32, and 34-35 are obvious over Imai in view of Couwenhoven.

  • Prior Art Relied Upon: Imai (Japanese Patent Application Publication No. H11331305) and Couwenhoven (Patent 5,596,602).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Imai taught a streaming media system with a plurality of selectable encoders (compression routines), where selection could be based on factors like the detected network transmission rate. Couwenhoven, in turn, taught controlling a compression module's bit rate by monitoring the fill conditions of a rate buffer, which corresponds to tracking pending data transmission requests. Petitioner contended that Couwenhoven’s rate controller monitors these pending requests (buffer fullness) against a predetermined threshold (e.g., channel transmission rate) and sends a control signal to modify the compression rate. When throughput fell below the threshold (buffer underflow threat), Couwenhoven taught increasing the compression module's output bit rate, which corresponds to selecting a faster compression routine as claimed.
    • Motivation to Combine: Petitioner asserted a POSITA would combine these references because they addressed complementary aspects of the same technical problem in streaming media. Imai provided a system with multiple selectable encoders and suggested that compression rate should be controlled, while Couwenhoven provided a well-known and specific mechanism (buffer feedback control) for doing so. A POSITA would have been motivated to apply Couwenhoven's simpler, faster, and localized feedback technique to Imai's multi-encoder system to improve performance, especially in server environments with constrained CPU resources where dynamically selecting a less computationally intensive encoder would prevent underflow.
    • Expectation of Success: The combination involved applying known throughput-control techniques (Couwenhoven) to a known type of multi-algorithm compression system (Imai) to achieve the predictable result of matching the compression data rate to the available transmission rate. As both technologies were well-known in the art of media streaming, a POSITA would have had a reasonable expectation of success.

Ground 2: Obviousness over Imai, Couwenhoven, and Ishii

Claim 24 is obvious over Imai and Couwenhoven in view of Ishii.

  • Prior Art Relied Upon: Imai (Japanese Patent Application Publication No. H11331305), Couwenhoven (Patent 5,596,602), and Ishii (Patent 5,675,789).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground built upon the combination in Ground 1 and added the teachings of Ishii to address the limitations of claim 24, which uniquely recites "a plurality of access profiles" used to determine a compression routine. Petitioner argued that Ishii explicitly taught using "access profiles"—defined as information regarding the number or frequency of reads/writes and the file type—to select a "suitable compression method." The combination of Imai and Couwenhoven provided the base system for tracking throughput and switching encoders, while Ishii provided the claimed method of using access profiles as an additional criterion for that selection.
    • Motivation to Combine: A POSITA would have been motivated to incorporate Ishii's teachings into the Imai/Couwenhoven system to further optimize the encoder selection process. Ishii provided a complementary factor (data access frequency and type) that improves compression efficiency beyond just monitoring transmission throughput. For example, for frequently accessed data, a POSITA could use Ishii's teachings to select a more symmetric or less computationally intensive algorithm to reduce server load and decoding complexity, which was a known consideration taught by Imai. All three references were directed to the same field of data compression and taught improvements for similar systems.
    • Expectation of Success: The combination would have yielded the predictable result of a more optimized compression system. Adding another known selection criterion (access profiles) to an existing selection framework was a well-understood design choice in the art with a high expectation of success.

4. Key Claim Construction Positions

  • "asymmetric" or "asymmetrical algorithm": Petitioner proposed this term means "an algorithm in which the execution time for the compression and decompression routines differ significantly." This construction was based on an express definition in the ’046 patent's specification.
  • "access profile": Appearing only in claim 24, Petitioner proposed this term means "information regarding the number or frequency of reads or writes." This construction was based on the specification's description of how different data types (e.g., operating systems vs. user documents) correspond to different read/write patterns, which are then used to select an appropriate compression algorithm.

5. Relief Requested

  • Petitioner requested institution of an inter partes review and cancellation of claims 1, 4, 8, 10-11, 13-14, 16-17, 19, 21, 23-24, 26-27, 29-32, and 34-35 of the ’046 patent as unpatentable.