PTAB

IPR2019-00591

NetApp Inc v. KOM Software Inc

Log In

1. Case Identification

2. Patent Overview

  • Title: Automated File Management in a Virtual Storage Medium
  • Brief Description: The ’642 patent relates to systems and methods for automated file management that store data across multiple physical storage media. The system presents these disparate media to a user as a single, coherent "virtual file-based non-volatile storage medium" and uses an indexing system to map virtual file locations to their corresponding physical storage locations.

3. Grounds for Unpatentability

Ground 1: Obviousness over Carter - Claims 1-6 and 16 are obvious over Carter

  • Prior Art Relied Upon: Carter (Patent 5,987,506).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Carter, which was not considered during prosecution, discloses all limitations of the independent claims. Carter teaches a "distributed shared memory system" that creates a "virtual storage space" spanning multiple physical, non-volatile storage devices (e.g., hard disks) across a network. This system functions as a "coherent, single file system" that is file-based, not block-based. Petitioner contended that Carter’s use of directory entries, Inodes, and stream descriptors constitutes an indexing system that maps virtual locations to physical storage pages, determines free space for allocation, stores data, and updates index information, thereby meeting the limitations of independent claims 1 and 16.
    • Key Aspects: Petitioner emphasized that Carter teaches the core concepts that the patentee relied on to secure allowance, namely a file-based virtualization system that stores a single file across multiple physical drives.

Ground 2: Obviousness over Carter and Mutalik - Claims 3-5 are obvious over Carter in view of Mutalik

  • Prior Art Relied Upon: Carter (Patent 5,987,506) and Mutalik (Patent 6,161,111).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground was presented as an alternative in case Carter's indexing system was viewed as mapping to virtual pages rather than directly to physical locations. Petitioner argued Mutalik teaches a file management system where an "extent descriptor" directly maps to physical storage by identifying the specific disk storage device, the starting block, and the number of successive blocks for a file portion. This provides a direct, single-layer mapping from a logical representation to a physical location.
    • Motivation to Combine: A POSITA would combine Mutalik's direct physical mapping with Carter's virtual file system to improve efficiency and reduce complexity. Incorporating Mutalik's teaching would eliminate an intermediate layer of virtual-to-physical mapping that may be present in Carter, thereby simplifying the indexing process, reducing overhead, and speeding up data retrieval by avoiding the need to consult a second, separate mapping structure.
    • Expectation of Success: A POSITA would have a high expectation of success because Carter already provides the necessary hardware and data structures (Inodes, etc.) for mapping. Modifying Carter’s software to implement Mutalik's more direct mapping approach would be a straightforward modification for a person of ordinary skill.

Ground 3: Obviousness over Carter and Cannon - Claims 7, 10, 17, and 20 are obvious over Carter in view of Cannon

  • Prior Art Relied Upon: Carter (Patent 5,987,506) and Cannon (Patent 5,983,239).

  • Core Argument for this Ground:

    • Prior Art Mapping: This ground addressed dependent claims requiring data migration and archiving based on usage monitoring. Petitioner asserted that Carter already discloses monitoring data access patterns and migrating data between nodes to balance loads. Cannon was cited for its explicit teaching of a hierarchical storage system that performs "automatic data migration" for archiving. In Cannon, files are moved between different storage tiers (e.g., from expensive, high-performance media to inexpensive, long-term archival media) based on criteria such as data age and frequency of use, with its index (a database) being updated to reflect the new file locations.
    • Motivation to Combine: A POSITA would be motivated to integrate Cannon's criteria-based archiving functionality into Carter's data migration framework to optimize both system performance and storage cost. This combination would allow less frequently used data in the Carter system to be automatically moved to slower, cheaper storage tiers, freeing up faster, more expensive media for frequently accessed data.
    • Expectation of Success: The combination was asserted to be predictable. Carter already provided the fundamental components: a system for monitoring data access, hardware with varied storage media, and a mechanism for updating its index. A POSITA could readily modify Carter’s existing migration routines to include Cannon's archiving triggers (e.g., "time since last access") as a routine software update.

  • Additional Grounds: Petitioner asserted additional obviousness challenges, including claim 5 over Carter in view of Frey, and claim 5 over Carter in view of Frey and Mutalik, based on combining known data striping and direct-mapping techniques to improve system performance.

4. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 1-7, 10, 16, 17, and 20 as unpatentable under 35 U.S.C. § 103.