PTAB

IPR2014-00901

VMware Inc v. Elec & Telecomm'N Research INST

Log In
Key Events
Petition
petition Intelligence

1. Case Identification

2. Patent Overview

  • Title: Apparatus for Redundant Interconnection Between Multiple Hosts and RAID
  • Brief Description: The ’346 patent discloses a fault-tolerant storage system featuring redundant interconnections between multiple host computers and a Redundant Array of Independent Disks (RAID). The system architecture includes two RAID controllers, each equipped with two network interface controllers (NICs), connected to hosts via two hub or switch devices, enabling communication between the RAID controllers for failover purposes.

3. Grounds for Unpatentability

Ground 1: Obviousness over Mylex and Hathorn - Claims 1-9 are obvious over Mylex in view of Hathorn (and vice versa).

  • Prior Art Relied Upon: Mylex (“Storage Area Networks: Unclogging LANs and Improving Data Accessibility,” a 1998 white paper) and Hathorn (Patent 5,574,950).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Mylex, a white paper describing commercial RAID controllers, discloses nearly every element of the challenged claims, including a redundant system with multiple hosts, dual RAID controllers, and hubs/switches. Mylex teaches fault tolerance where NICs on a functioning controller take over for those on a failed controller. However, Mylex achieves inter-controller communication for fault monitoring via a direct "heartbeat" path. Hathorn discloses a similar redundant storage architecture and explicitly teaches that dedicated communication paths are expensive.
    • Motivation to Combine: A person of ordinary skill in the art (POSITA) would combine these references for two primary reasons. First, IBM (assignee of Hathorn) acquired Mylex, creating a strong business and technical motivation to integrate their respective storage technologies. Second, a POSITA would be motivated to improve the cost-effectiveness of the Mylex system by replacing its expensive, dedicated heartbeat path with Hathorn's more efficient method of using the existing switch network for inter-controller communication, which was a known design choice to achieve a predictable result.
    • Expectation of Success: A POSITA would have a high expectation of success, as the combination involves replacing one known method of inter-controller communication (direct path) with another known, more economical method (switched network) within similar storage system architectures.

Ground 2: Obviousness over Deitz/Mylex and Griffith/DeKoning - Claims 1-9 are obvious over Deitz in view of Griffith.

  • Prior Art Relied Upon: Deitz (Patent 6,578,158) and Griffith (Patent 6,401,170). Petitioner also presented combinations with Mylex and DeKoning (Patent 6,073,218) as alternatives.

  • Core Argument for this Ground:

    • Prior Art Mapping: Petitioner asserted that Deitz, like Mylex, discloses a redundant RAID system with multiple hosts, hubs, and dual controllers providing transparent failover. Deitz teaches active and inactive (failover) ports on each controller and, critically, uses a direct inter-RAID-controller communication path for transmitting heartbeat signals to monitor for failures. The primary element allegedly missing from Deitz is the specific claimed feature of exchanging information between network controlling units of different RAID controllers through the host-side connection units (hubs/switches). Griffith expressly teaches this missing element, disclosing that fault tolerance information can be exchanged either through a direct path or by allowing the NICs to use the existing switch network.
    • Motivation to Combine: A POSITA designing a fault-tolerant RAID system would have looked to the solutions of major RAID providers like IBM (assignee of Deitz) and Digi-Data Corporation (assignee of Griffith). A POSITA would combine the robust failover architecture of Deitz with the flexible and cost-effective communication method taught by Griffith. Implementing Griffith's teaching of using the existing switch fabric for inter-controller communication into the Deitz system would be a simple substitution of one known communication technique for another to achieve the predictable benefits of reduced cost and complexity.
    • Expectation of Success: The combination was presented as highly predictable, as it involved applying a known communication option (using the existing network) from Griffith to a standard redundant RAID architecture as disclosed in Deitz.

  • Additional Grounds: Petitioner asserted alternative obviousness challenges based on combinations of Deitz or Mylex with DeKoning, but relied on similar theories of combining a base redundant RAID system with a known method for inter-controller communication.

4. Key Claim Construction Positions

  • "connection unit" (Claims 1-9): Petitioner proposed this term be construed as "a hub or switch." This construction is central to arguing that prior art references teaching communication through a switch or hub network meet the claim limitations for connecting hosts and RAID controllers.
  • "network interface controller," "network controlling unit," and "network interface controlling unit" (Claims 1-9): Petitioner proposed these related terms be construed as "the part of a RAID controller that allows the RAID controller to communicate with the 'connection units'." This broad construction allows Petitioner to map various ports and interfaces disclosed in the prior art, such as the active and inactive ports in Deitz, to these claimed elements.
  • "exchange/exchanges information" (Claims 1-9): Petitioner proposed this be construed to mean "to transmit and receive information reciprocally." This construction supports the argument that the prior art's teachings of sending heartbeat signals and failover commands meet the claimed information exchange limitations.

5. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 1-9 of the ’346 patent as unpatentable under 35 U.S.C. §103.