EMC Corp v. Clouding Corp

1. Case Identification

• Case #: IPR2015-01922
• Patent #: 5,944,839
• Filed: September 16, 2015
• Petitioner(s): EMC Corporation, EMC International U.S. Holdings, Inc., and VMware Corp.
• Patent Owner(s): Clouding Corp.
• Challenged Claims: 1, 15, and 17

2. Patent Overview

• Title: System and Method for Automatically Maintaining a Computer System
• Brief Description: The ’839 patent describes a maintenance tool for computer systems. The tool uses a plurality of software "sensors" to gather data, which is analyzed by an Artificial Intelligence (AI) engine against a knowledge database of known problems ("cases") to diagnose and solve issues.

3. Grounds for Unpatentability

Ground 1: Anticipation by Douik - Claims 1, 15, and 17 are anticipated by Douik under 35 U.S.C. § 102(e).

• Prior Art Relied Upon: Douik (Patent 6,012,152).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Douik, which was not considered during the original prosecution of the ’839 patent, discloses every element of the challenged claims. Douik describes a "Software Fault Management" (SFM) system that uses an intelligent multi-agent system to automatically maintain a telecommunications network.
    • Claim 1 (Tool): Petitioner asserted that Douik’s SFM system is a tool for automatically maintaining a computer system. The system comprises a "Knowledge Base" (KB) that includes a "Trouble Report/Known Faults (TR/KF) case base," satisfying the claimed "knowledge database" holding a "plurality of cases" with problems and solutions.
    • Claim 1 (Sensors): Douik’s system employs software agents for "network monitoring" that gather and store data about the system, which Petitioner equated to the claimed "plurality of sensors." These agents detect problems by identifying alarms or trouble reports from the network.
    • Claim 1 (AI Engine): Douik’s system includes an AI engine that uses "Case-Based Reasoning" (CBR) to analyze fault data against its case base to determine if a suspected fault is a "known fault" with a likely solution, thus meeting the limitation of an AI engine utilizing cases to find a solution.
    • Claim 1 (Gathering More Data): Petitioner contended that if a fault is not readily identified, Douik’s system performs "further fault analysis," which may involve interfacing with "various test tools" and gathering "additional network monitoring data." This was mapped to the claim limitation where the AI engine activates a sensor to gather necessary data when the knowledge base is insufficient.
    • Claim 1 (Saving New Cases): A key element added during prosecution was saving gathered data as a new case when no solution is found. Petitioner argued Douik teaches this by describing that for "unknown faults," the system assists an engineer by providing relevant information and tools, and "ensures the logging of the fault specification and the undertaken corrective actions for future utilization." Petitioner asserted that a POSITA would understand this process of learning from experience and building an "episodic memory" as creating a new case.
    • Claims 15 and 17: Petitioner argued that independent claim 15, directed to a program storage device, is anticipated for the same reasons as claim 1. Dependent claim 17 adds the step of "periodically activating" a sensor, which Petitioner contended is taught by Douik's disclosure of a "proactive mode" where the system "continually monitors, through dynamic polling, the state and behavior of critical resources."

4. Key Claim Construction Positions

• Petitioner adopted several constructions from the Board's decision in a prior review (IPR2013-00095) involving the same patent.
• "solution to the computer problem": Construed to require "specific information on problem resolution or elimination and is not satisfied by mere identification of the problem." This construction is central to mapping Douik’s teachings of implementing "corrective actions."
• "sensors": Defined as "software programs that gather information from the computer system." This allows Petitioner to map Douik's "network monitoring agents" and "event report handler" to the claimed sensors.
• "Artificial Intelligence (AI) Engine": Construed as "a computer program... used for decision making or problem solving, based on reasoning or inferences." This broad construction supports identifying Douik's intelligent multi-agent system, which performs case-based reasoning, as the claimed AI engine.

5. Arguments Regarding Discretionary Denial

• Petitioner acknowledged that the Board had previously denied institution for claims 1, 15, and 17 in a prior petition (IPR2014-01309).
• However, Petitioner argued that denial of this petition would be improper because the prior art asserted here, Douik, is "entirely different" from the art cited in the prior petition and was never considered during the original examination of the ’839 patent. Petitioner contended that Douik specifically teaches the elements that were found to be missing in the prior art of record during prosecution, making this a distinct and meritorious challenge.

6. Relief Requested

• Petitioner requested the institution of an inter partes review (IPR) and the cancellation of claims 1, 15, and 17 of Patent 5,944,839 as unpatentable.