NetApp Inc v. Proven Networks LLC

1. Case Identification

• Case #: IPR2020-01437
• Patent #: 8,018,852
• Filed: August 13, 2020
• Petitioner(s): NetApp, Inc.
• Patent Owner(s): Proven Networks, LLC
• Challenged Claims: 1-18

2. Patent Overview

• Title: Port Selection for Network Switching
• Brief Description: The ’852 patent discloses a port selection method in a network with multiple equal-cost paths. The method uses "source learning" to associate an inbound data flow's network address with the specific port on which it was received, and then directs the corresponding outbound flow through that same port to ensure symmetric communication paths.

3. Grounds for Unpatentability

Ground 1: Claims 1-18 are obvious over Bardzil, Moy, and Rosen under 35 U.S.C. §103.

• Prior Art Relied Upon: Bardzil (Patent 7,426,577), Moy (an Internet Engineering Task Force RFC 2328 published in April 1998), and Rosen (an Internet Engineering Task Force RFC 3031 published in January 2001).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that the combination of these three references teaches every element of the challenged claims.
    • Moy provides the foundational routing protocol, Open Shortest Path First (OSPF). Moy taught that a router using OSPF identifies all minimal equal-cost paths to a destination and can distribute traffic among the corresponding outgoing ports or interfaces. This established the claimed environment of a switching device identifying a plurality of ports associated with minimal equal-cost paths.
    • Bardzil addressed a known problem with OSPF: its standard traffic distribution can create asymmetric paths, where inbound and outbound traffic between two nodes traverse different routes. Bardzil taught that such asymmetry is undesirable for real-time communications (e.g., IP telephony) and proposed a solution: forcing outbound traffic to use the same "next hop" (and thus the same path) that was used by the corresponding inbound traffic, thereby creating symmetric paths.
    • Rosen described Multiprotocol Label Switching (MPLS) and its Forwarding Equivalence Class (FEC) mechanism. Rosen taught that an FEC can group packets to be routed collectively and that the assignment to an FEC can be based on information like the source network address and the port on which an inbound packet arrived. Crucially, Rosen explained that an FEC could be used to force packets to follow a specific, explicitly chosen route, rather than the path that would otherwise be selected by a dynamic routing algorithm like OSPF. This provides the specific implementation mechanism for Bardzil’s symmetric routing concept.
  • Motivation to Combine: Petitioner contended that a person of ordinary skill in the art (POSITA) would have been motivated to combine the references to improve the performance of real-time network communications. A POSITA starting with the OSPF routing of Moy would recognize the asymmetry problem identified by Bardzil as a known issue to be solved. Bardzil’s proposal to enforce symmetric routing provides a clear objective. To implement this solution, a POSITA would have looked to known routing tools and found it obvious to use the FEC mechanism from Rosen. Rosen expressly taught using FECs to override standard OSPF routing to force traffic along a desired path, making it a perfect tool to implement Bardzil’s symmetric path solution within the OSPF framework disclosed by Moy. The motivation was to achieve the known benefits of symmetric routing—such as improved quality for IP telephony—using a known and suitable implementation mechanism.
  • Expectation of Success: A POSITA would have had a reasonable expectation of success in making this combination. Bardzil’s teachings are presented in the context of an OSPF network, indicating the concepts are compatible. Similarly, Rosen’s FEC mechanism was explicitly described as a tool to modify the behavior of dynamic routing protocols like OSPF. The references are from the same field of packet-switched networks and address interrelated problems with compatible solutions, leading to a predictable outcome.

4. Arguments Regarding Discretionary Denial

• Petitioner argued that discretionary denial under §314(a) based on the Fintiv factors would be improper. The petition was filed within three months of service in the parallel district court litigation (Proven Networks, LLC v. NetApp, Inc., W.D. Tex.). Key arguments included that no trial date had been set, the court’s and parties’ investment in the litigation was minimal, and the invalidity contentions had not yet been served, meaning there was no record of overlapping arguments. Petitioner also stipulated that should the IPR be instituted, it would not pursue the same invalidity ground in the district court, mitigating any concerns of duplicative efforts. Finally, Petitioner noted that the asserted prior art combination was not considered during the original prosecution of the ’852 patent.

5. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 1-18 of Patent 8,018,852 as unpatentable.