US Patent 6,122,664 (US6122664A) — Summary

Search note

I searched for patent number 6122664 (and the "6,122,664" form). I found no CAFC 2026 docket entries or any pending litigation involving this patent — it expired in 2017, so active 2026 litigation is implausible. If a CAFC case exists, it was not surfaced by my searches; treat that as unverified rather than confirmed absent.

Bibliographic data (verified against Google Patents / USPTO-facing sources)

Abstract (verbatim)

"The present invention relates to a process for monitoring a plurality of object types of a plurality of nodes (N1, N2, . . . , Nn) comprising a management node (MN) in an information system. Monitoring is configured and then distributed in a filtered way from the management node (MN) to autonomous agents (SAA), an autonomous agent being installed in each node to be monitored in order, by providing intertype correlation, either to locally process the different object types or all of the objects of a domain called a global object, defined generically, or to feed back information to be displayed to the graphical interface of the management node, each agent comprising a plurality of specific modules (SM1, SM2, . . . , SMn) specific to the different object types or to a particular domain, each specific module measuring static and dynamic parameters particular to the object type it monitors and collecting said measurements, testing conditions on said parameters relative to predefined thresholds and possibly triggering actions associated with said tested conditions, which parameters, conditions and actions can be modified by the user of the management node."

What the invention does (plain language)

The patent solves the classic SNMP manager/agent problem — heavy network traffic and slow reaction to events — by pushing autonomous, locally-executing monitoring agents out to each monitored node. Each agent contains a generic agent plus plug-in "specific modules" (one per object type: Oracle, Informix, Tuxedo, UNIX system, Distributed Print Facility, etc.). Each module measures static/dynamic parameters, evaluates conditions against thresholds (with hysteresis, "during n seconds," ranges), and triggers local actions (alerts, reconfiguration, correction) without round-tripping every decision to the manager. The management node retains a GUI, a configuration file, status files and trace files, and only receives display feedback. It also supports "global objects" (aggregates of basic objects), inter-type correlation of measurements, multiple conditions evaluated at the management node, log-file scanning (SCANLOG/error-control files), and failover via BACKUP_NODE. Configurations are "filtered" per node before distribution.

Independent claims — overview with a caveat

Caveat (important): the claims section was not included in the full patent text supplied to me, and my searches only returned the text of dependent claims 6–20 (via everypatent.com). The wording of claims 1–5 could not be verified verbatim. From the dependency chain visible in the search results (claim 6 → claim 5, claim 7 → claim 6, etc.), claims 1–5 are the independent claims. Based on the specification's summary passages, the independent claims are understood to cover, in substance:

1. Claim 1 — core monitoring process: Configuring monitoring at a management node and distributing it in a filtered way to autonomous agents installed in each node to be monitored, where each agent has multiple specific modules (per object type or domain), each module measures and collects static/dynamic parameters, tests conditions relative to predefined thresholds, and triggers associated actions — with the parameters, conditions and actions being modifiable by the management-node user, and intertype correlation provided (per the abstract/summary).
2. Claim 2 (likely) — management-node structure: The management node includes a graphical user interface (object icons, parameter curves), a configuration file of object descriptions and predefined parameters (analyzable/dynamically modifiable), node status files, parameter display files, and trace files for measured parameters.
3. Claim 3 (likely) — agent structure: The autonomous agent in each node is composed of a generic agent linked to a plurality of specific modules plus files of basic functions, with each node having its own parameter/condition/action files controlling its local monitoring.
4. Claim 4 (likely) — multiple-condition processing: A condition not linked to a single parameter (spanning one or more objects) is processed at the management node by a generic "management" agent (GA′) that evaluates conditions across several nodes and initiates the action there.
5. Claim 5 (likely) — cross-node value requests / configuration merge: Either (a) a predefined/external action requesting a parameter value from another node routes the request through the management node, or (b) the generic management agent's operation includes reading its configuration file, performing syntactic/semantic analysis, merging the management-node configuration with specific-module configurations, and sending the resulting filtered configuration to each agent while accounting for malfunctioning/backup nodes — claim 6 explicitly builds on claim 5 for this operation, so claim 5 is the base for the "generic management agent" configuration mechanism.

Verified dependent-claim content (for reference)

• Claim 6 (on claim 5): generic management agent operation — configuration read, syntactic/semantic analysis, merging of management-node and specific-module configuration files, filtered distribution to each agent, handling of malfunctioning nodes and backup nodes.
• Claim 7 (on claim 6): generic agent of a monitored node — processes only local objects; builds object, parameter-configuration, parameter-value and multiple-condition tables; supplies basic functions; starts specific modules; configures log-file scanning; evaluates multiple conditions and initiates internal/external actions; verifies module functioning.
• Claim 9 (on claim 7): specific-module operation — scans parameter config table for its own objects, sorts parameters by period, performs measurements (internal or external commands), stores values, traces/display, evaluates single conditions, initiates actions when true, handles real-time measurement requests.
• Claims 13–15: cross-node parameter requests routed via the management node; log-file collection (paramlog/actionlog) at the management node for independent analysis.

Uncertainty summary

• Claims 1–5 wording is reconstructed, not verbatim — verify against the USPTO full text (patents.google.com/patent/US6122664A or USPTO Patent Center) if exact claim language matters.
• Assignee naming varies across sources ("Bull S.A." vs. "Bull SAS"); Google Patents shows original assignee "Bull SAS" and current assignee EVIDAN.
• No CAFC 2026 docket activity was found; I cannot affirmatively rule out a case, only report that searches returned nothing.

Litigation Search Results: US Patent 6,122,664

Finding: No known litigation located

After searching for US patent 6,122,664 (the "6122664" identifier, read literally), I found no district court cases, ITC proceedings, or CAFC appeals involving this patent.

What I searched and what came back

1. Unified Patents portal (portal.unifiedpatents.com/patents/patent/US-6122664-A) — The patent page exists and shows bibliographic data, cited prior art, and 398 "referenced by" (forward-citation) documents, but no litigation tab or case list was surfaced for this patent.

2. General litigation queries — Searches combining the patent number with terms like "litigation," "lawsuit," "case," "court docket," "defendants," and "infringement" returned no results tied to 6,122,664.

3. Assignee-focused query — A search for "EVIDAN patent 6122664 litigation" (EVIDAN being the later assignee per the recorded nunc pro tunc assignment from Bull S.A. in 2006) returned nothing.

Watch out for a false positive

One search result did match on the "664" digits, but it is a different patent and must not be conflated:

• Lowe v. Shieldmark, Inc., No. 23-1786 (Fed. Cir. Mar. 24, 2025) concerns U.S. Patent No. 10,214,664 (industrial floor-marking tape, inventor Clifford A. Lowe). That is not 6,122,664. I am flagging it only because it is the kind of near-miss that can pollute a docket search.

Context supporting the "no litigation" conclusion

• The patent's priority date is 1996-06-27; it issued 2000-09-19 and expired (Lifetime) with anticipated expiration on 2017-06-27. A patent that expired in 2017 is an implausible subject for active 2026 litigation, and I found no evidence of any historical case either.
• The Forward Citations (398) and NPL references shown on Unified Patents are citation/citation-network data, not litigation data.

Caveat

My search tools cover public web sources (Unified Patents, Justia, FreePatentsOnline, Google Patents, legal-news aggregators). I did not independently query PACER's full-text docket database or the CAFC's ECF system directly in this session, so I cannot affirmatively certify the complete absence of a case that might exist only behind PACER's paywall. Based on all sources I could access, however, no known litigation involving US patent 6,122,664 was found. If a case exists, it was not surfaced by these searches and should be treated as unverified rather than confirmed.

Proceedings overview

Total AIA trial proceedings on US 6,122,664: zero. The USPTO Open Data Portal (canonical source for this analysis) reports no IPR, PGR, or CBM proceedings against this patent, and independent web searches for "6122664 IPR," "inter partes review," PTAB decisions, and Federal Circuit appeals returned nothing specific to this patent. There is no PTAB record of claims being invalidated, sustained, settled, or denied institution — the patent simply was never challenged in an AIA trial. For a defendant, the decisive fact is not the PTAB record but the patent's expiry on 2017-06-27 (status: "Expired – Lifetime") — any demand letter today can be answered with expiry, regardless of validity.

Proceedings

None on file. No IPR, PGR, CBM, or derivation proceeding exists for US 6,122,664 in the USPTO ODP data, and none surfaced in web searches. I will not invent proceeding numbers, judge panels, institution decisions, or Final Written Decisions — there are none to report.

What the record does show, for completeness:

• No PTAB activity: The ODP API ingest shows zero AIA trials for this patent. Searches of PTAB decision databases and news aggregators surfaced no petition, no institution decision, and no FWD citing this patent.
• No ex parte reexamination surfaced: Unified Patents' portal tracks US-6122664-A (assignee listed as EVIDAN) and lists 12 items of "Patent Art" (e.g., US-5,655,081-A BMC Software; US-5,491,796-A NetLabs; US-5,682,523-A IBM) and 2 NPL references (e.g., Hacker, "Operations Management," ICL Technical Journal, Nov. 1991), but my searches surfaced no reexamination control number or challenge outcome for the patent. Treat any reexamination history as unverified, not confirmed.
• No Federal Circuit appeal: No CAFC docket entry involving this patent or the EVIDAN/Bull chain was found (including for 2026).
• No district-court assertion campaign surfaced: No EVIDAN v. [defendant] litigation naming this patent appeared in searches. Absence of litigation is consistent with a patent that expired in 2017 — a post-expiry assertion would be limited to pre-2017-06-27 damages, which is a steep hill for any plaintiff.

Strategic summary

Claim-by-claim status — all UNTESTED, not CANCELED or SUSTAINED. No PTAB tribunal has ever evaluated any claim of 6,122,664. All twenty claims (as far as can be verified — note the earlier caveat that independent claims 1–5 could not be confirmed verbatim from the supplied text) remain in their original issued form in the record, but the patent is expired, which is the operative limitation. A plaintiff asserting this patent today faces a far more basic problem than validity: the patent's enforceable life ended 2017-06-27, and 35 U.S.C. § 286 caps damages at six years before filing suit in any event — a suit filed in 2026 could reach at most conduct from 2020 onward, all of which post-dates expiry. That is a complete defense independent of any PTAB proceeding.

Estoppel landscape — a blank slate. Because no IPR was ever filed, there is no § 315(e)(2) estoppel binding anyone, and no prior-art ground has been "used up." But this cuts the wrong way for a challenger: the absence of a PTAB record means no party has established unpatentability, so a defendant cannot lean on an FWD — the defense must be built from scratch (expiry, § 286 damages cap, indefiniteness/§ 101, and prior art including the 12 references Unified Patents catalogs, if a plaintiff somehow asserts pre-2017 conduct). No petitioner, no privy, and no estoppel concerns arise.

Pattern signals — none. There is no serial petitioner, no defensive-aggregator challenge (Unified Patents merely tracks the patent in its portal; no challenge by Unified was surfaced), and no evidence of the patent owner fighting at the PTAB. The one notable ownership signal is the 2006 nunc pro tunc assignment from Bull S.A. to EVIDAN, but no enforcement activity tied to that assignment was found. The cleanest read: this is a legacy, expired Bull-era monitoring patent that never attracted a post-grant challenge — the type of patent that usually appears in the record because it is being asserted. Its absence from the PTAB docket, combined with expiry, suggests it is not being actively monetized.

Recommended next steps

1. Lead with expiry, not PTAB. There is no FWD to quote and no disposition to link. Your motion to dismiss / summary judgment should be built on the patent's status — "Expired – Lifetime," anticipated expiration 2017-06-27 (Google Patents / USPTO Patent Center records) — plus 35 U.S.C. § 286's six-year damages bar. Any 2026 complaint necessarily seeks damages only for post-2020 conduct, all of which is outside the enforceable term.
2. Verify the claims if the plaintiff actually pleads them. The independent claims (1–5) were not available verbatim in the materials supplied; pull the full text from USPTO Patent Center or Google Patents (patents.google.com/patent/US6122664A) before responding, so you can challenge claim construction and indefiniteness precisely rather than relying on the reconstructed summary.
3. If the plaintiff produces a previously unseen assignment or reissue, re-run the PTAB/reexamination check — but do not assume hidden proceedings exist. The ODP data and independent searches both come up empty, and that absence is itself meaningful: a patent this old, this expired, and this unchallenged is far more likely a paper tiger than a litigation asset.
4. Do not file an IPR. The statutory one-year bar in § 315(b) runs from service of a complaint, and more fundamentally, a challenge to an expired patent's claims is a wasted expenditure when expiry is a complete defense — and under current USPTO practice (2025–2026 director-level institution review), a petition against a decades-old, expired patent would face steep discretionary-denial headwinds on "settled expectations" grounds regardless.

Bottom line: no PTAB proceeding exists on this patent — not one. The defensive posture is not "claims 1–5 canceled" and not "patent owner prevailed at the PTAB"; it is "the patent expired in 2017 and was never challenged in any AIA trial." A defendant's strongest, simplest answer to any demand is the expiration date.

I have reached the tool-call limit, so I'll finalize based on the verified data I have: the full patent text supplied (which includes Google Patents legal events), the Unified Patents portal entry, and the Register article on Bull's 2000 spin-off. I could not retrieve reel/frame numbers or correspondent names from my searches — I flag that plainly below rather than fabricate them.

Ownership-chain reconstruction — US 6,122,664

Inventors

• Marcel Boukobza — France; employer at filing: Bull S.A. (he is the first-named assignor on the 1998 recorded assignment to Bull S.A., consistent with an employee-inventor assignment).
• Gerard Sitbon — France; employer at filing: Bull S.A. (second-named assignor on the same recorded assignment).
• Pattern note: Both inventors assigned their rights to Bull S.A. before issuance — the normal employee-inventor pattern. There is no evidence of either inventor departing Bull within 12 months of filing or of any inventor-side retention that would precede a portfolio fire-sale. Not a suspicious pattern.

Original assignee

• Bull S.A. (Google Patents normalizes the name to "Bull SAS"; the assignment record uses "BULL S.A."). French enterprise-computing manufacturer (Groupe Bull), headquartered in Les Clayes-sous-Bois, France.
• Product embodiment: Yes — the patent covers Bull's distributed systems/application monitoring architecture, and the commercial continuation of that software line (Bull's systems-management products, spun off with the BullSoft unit) shipped as Evidian's "OpenMaster" intelligent-monitoring product. The assignee therefore shipped products squarely in the field of the claims.
• Current status: Operating (as part of Atos). Bull S.A. was reorganized in 2000 (spin-off of the software business into Evidian), its services arm combined with Atos Origin, and Atos SE completed acquisition of Bull in 2014. The patent itself expired 2017-06-27 (Lifetime).

Assignment timeline

Only two conveyances are reflected in the record (per the Google Patents legal events embedded in the authoritative patent text and mirrored in Unified Patents' assignee field). Reel/frame numbers and correspondents of record were not retrievable from my searches — verify them directly at the USPTO Assignment Center (https://assignmentcenter.uspto.gov/) by searching patent number 6122664. I will not guess at reel/frames.

• ~1997 (executed) / 1998-01-07 (recorded) — Reel/frame: not retrieved (unverified); Google Patents event type: "ASSIGNMENT OF ASSIGNORS INTEREST"

  • Conveyance: Assignment (inventor → employer)
  • Assignor: Marcel Boukobza; Gerard Sitbon
  • Assignee: BULL S.A.
  • Correspondent: not retrieved
  • Context: Standard employee-inventor assignment recorded at issuance — original-owner acquisition, not a transfer of concern.

• ~2000 (effective) / 2006-09-10 (recorded) — Reel/frame: not retrieved (unverified); Google Patents event type: "NUNC PRO TUNC ASSIGNMENT"

  • Conveyance: Assignment (nunc pro tunc, i.e., back-dated to an earlier effective date)
  • Assignor: BULL S.A.
  • Assignee: EVIDAN (the operating company "Evidian," a Bull Group subsidiary; note the record's spelling "EVIDAN")
  • Correspondent: not retrieved
  • Context: Internal reorg/spin-off — Bull created Evidian in July 2000 out of its BullSoft security and network/systems-management software business (per The Register, 2000-12-01); the 2006 nunc pro tunc record formalized that spin-off for this patent. Transfer to the natural product-line owner, not a sale to an outside entity.

No further assignments, security agreements, mergers, licenses, or releases were found. The record therefore appears to be closed at EVIDAN (Evidian).

Timeline diagram

timeline
    title Ownership of US 6122664
    1996 : Priority date filed in France
    1997 : US application filed by Bull
    1998 : Assigned to Bull SA
    2000 : Patent issued
         : Bull spins off Evidian software unit
    2006 : Assigned to Evidan nunc pro tunc
    2017 : Patent expires

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The 2006 transfer went to EVIDAN, which is not a licensing-only LLC: it is a Bull Group operating subsidiary founded in 2000, with employees, R&D, customers, and shipping products (Evidian OpenMaster monitoring; later identity-and-access-management lines). No registered-agent address, no single-purpose Delaware/Texas LLC, no "IP/Patents/Holdings" name pattern.

2. Known asserter in the chain — not present. Neither Bull S.A. nor Evidian appears on the Acacia/Marathon/IV/Wi-LAN/Mosaid/Vringo/Innovatio/MPHJ-type NPE lists, and Unified Patents' portal (portal.unifiedpatents.com, patent US-6122664-A) lists Evidan as assignee in the ordinary course — not as a tracked high-frequency plaintiff. No district-court or CAFC docket naming this patent was surfaced by any search.

3. Repeat correspondent across the chain — unclear (unverified). I could not retrieve the correspondent names or reel/frame numbers for either recording from the sources available to me. The 1998 and 2006 filings are 8+ years apart and belong to different eras (Bull in-house vs. Evidian), so a repeat-correspondent signal is not evidenced either way. This item should be checked against the Assignment Center images directly; absent that, no finding.

4. Cascading transfers — not present. Only two recorded conveyances total, separated by ~8 years, both involving the same corporate family (Bull → Bull-owned Evidian). No chained LLCs, no <24-month cascade.

5. Pre-litigation transfer — not present. No infringement suit naming US 6,122,664 was found (the patent expired 2017-06-27 and searches returned no CAFC or district-court activity). The 2006 nunc pro tunc record is ~17 years post-issue and tied to a corporate reorg, not to enabling litigation.

6. Bankruptcy fire-sale — not present. Bull's 2000 reorganization and 2014 Atos acquisition were not bankruptcy proceedings; the patent was not sold in any Chapter 7/11 estate (contrast Kodak/Nortel).

7. Privateering — not present. The transfer ran within Groupe Bull to the subsidiary that owned the corresponding product line (OpenMaster). There is no evidence of an operating company funneling the patent to an outside NPE to assert against competitors.

8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN.

Verdict

Operating-company assertion

Justification: The chain is short and entirely intra-Groupe-Bull — inventors → Bull S.A. (recorded 1998-01-07) → EVIDAN, the operating Evidian subsidiary (recorded 2006-09-10, nunc pro tunc) that ships the OpenMaster monitoring product line descended from the patent's subject matter. No shell LLC, no known asserter, no cascading transfers, no litigation, and no bankruptcy sale appear anywhere in the record, so none of the NPE signals are present. Caveat: because the patent expired in 2017 and no infringement suit was found, the "assertion" label reflects the owner profile (a product-shipping operating company, not an NPE) rather than confirmed litigation, and reel/frame + correspondent details remain unverified pending a direct Assignment Center lookup.

Verify the reel/frame and correspondent fields at: https://assignmentcenter.uspto.gov/ (search "6122664") and cross-reference Google Patents legal events at https://patents.google.com/patent/[US6122664A](/patent/US6122664A).

The Google Patents text pages aren't surfacing the citation list. Let me try targeted searches on citation-listing aggregators and the USPTO's own records.

I have reached the search-step limit, so I will now compile the prior-art analysis from the verified data I did obtain (primarily the Unified Patents portal's citation listing for US-6122664-A, cross-checked against Google Patents and EPO GPI search-report hits). I flag every date or fact that comes from my training data rather than the live search results.

Prior Art Analysis — US Patent 6,122,664 (US6122664A)

Methodological caveats (read first)

1. Claims 1–5 are reconstructed. As documented in the earlier summary, the verbatim text of claims 1–5 was not in the source text and could not be verified in this session. My claim-mapping below is therefore provisional, keyed to the reconstructed claim scope:
   • Claim 1 (core method): configure monitoring at a management node; distribute it in a filtered way to autonomous agents installed in each monitored node; each agent has multiple specific modules (per object type/domain); each module measures/collects static and dynamic parameters, tests conditions against predefined thresholds, triggers associated actions; parameters/conditions/actions modifiable by the management-node user; intertype correlation provided.
   • Claim 2 (management-node structure): GUI (icons, curves), configuration file, node status files, parameter display files, trace files.
   • Claim 3 (agent structure): generic agent + specific modules + basic-function files; per-node parameter/condition/action files.
   • Claim 4 (multiple conditions): conditions spanning >1 object evaluated at the management node by a generic management agent.
   • Claim 5 (base for claim 6): generic management agent operation incl. configuration read/analysis, merge of management-node and specific-module configs, filtered distribution, backup/malfunctioning-node handling (and/or cross-node value-request routing).
   • Claims 6–20: verified in substance from the earlier summary (claims 6, 7, 9, 13–15 content confirmed).
2. Statutory frame. The patent issued 2000-09-19 and claims priority to 1996-06-27, so it was examined under pre-AIA 35 U.S.C. § 102. A reference is § 102(a)/(b) art if published before 1996-06-27; a US patent whose underlying US application was filed before 1996-06-27 is § 102(e) art even if it issued later.
3. Dates. The filing/priority dates below come from the live Unified Patents citation listing. The publication dates come from my training data and are marked "(training data — verify)" where I am not fully certain.

Cited patent references (12)

1. US 5,491,796 A — "Apparatus for Remotely Managing Diverse Information Network Resources"

• Filing date (verified): 1992-10-22 · Publication (training data): ~1996-02-27
• § 102 basis: § 102(a)/(e) (filed and published before 1996-06-27)
• Description: Remotely manages diverse information-network resources from a central point; a manager console distributes management capability toward heterogeneous networked resources — close in spirit to the '664 management-node → agents model.
• Potential anticipation: Claim 1 (remote management of diverse object/resource types via distributed managers/agents) and possibly Claim 2 (management-node console/GUI functions), depending on how broadly the examiner construed "autonomous agents" and "specific modules." Weak on intertype correlation and filtered configuration, so full anticipation of claim 1 is doubtful on those limitations.

2. US 5,682,523 A — "System and Method for Collecting and Retrieving Network Problem Determination Data with a Generic Collection Subsystem Reporting to an Agent on Demand" (IBM)

• Filing date (verified): 1993-12-14 · Publication (training data): ~1997-10-28
• § 102 basis: § 102(e) (US filing before 1996-06-27)
• Description: A generic collection subsystem gathers network problem-determination data locally and reports to an agent on demand — a distributed, agent-facing data-collection architecture.
• Potential anticipation: Claim 1 (distributed collection of parameters by an agent-side component) and Claim 3 (agent structure with generic + module components). Does not appear to disclose threshold-condition evaluation with associated actions or user-modifiable parameter sets, so full § 102 anticipation of claim 1 is weak; it is stronger as § 103 combination art.

3. US 5,802,146 A — "Maintenance Operations Console for an Advanced Intelligent Network" (Bell Atlantic/NYNEX → Verizon)

• Filing date (verified): 1995-11-21 · Publication (training data): ~1998-09-01
• § 102 basis: § 102(e) (US filing before 1996-06-27)
• Description: A maintenance/operations console for an advanced intelligent network (AIN) — a centralized operator console presenting status of distributed network elements.
• Potential anticipation: Claim 2 (management node with GUI, status display, operator console) most plausibly; possibly Claim 4 (management-node-side processing). Not relevant to the autonomous-agent core of claim 1.

4. WO 1995/27249 A1 — "Method and Apparatus for Monitoring and Controlling Programs in a Network"

• Priority date (verified): 1994-04-04 · Publication (training data): ~1995-10-12
• § 102 basis: § 102(a)/(b) (published before 1996-06-27)
• Description: Distributed method/apparatus for monitoring and controlling application programs across a network — one of the closest prior-art teachings of pushing monitoring/control intelligence out to network nodes.
• Potential anticipation: Claim 1 (distributed monitoring of applications in multiple nodes) is its strongest target; possibly Claim 3. Whether it discloses the specific-module-per-object-type structure and threshold-condition/action triggering is unclear from the title alone — flag for full-text review.

5. US 5,655,081 A — "System for Monitoring and Managing Computer Resources and Applications Across a Distributed Computing Environment Using an Intelligent Autonomous Agent Architecture" (BMC Software)

• Filing date (verified): 1995-03-07 · Publication (training data): ~1997-08-05
• § 102 basis: § 102(e) (US filing before 1996-06-27)
• Description: The BMC PATROL-architecture patent: intelligent autonomous agents distributed across a computing environment monitor/manage computer resources and applications; agents are extensible via per-application/per-resource knowledge modules — structurally very similar to the '664 "generic agent + specific modules" design.
• Potential anticipation: The single most relevant reference. Likely the examiner's closest § 102 candidate against Claim 1 (autonomous agents, per-object-type modules, distributed monitoring) and Claim 3 (agent = generic framework + type-specific modules). Whether it discloses (i) filtered configuration distribution from the management node, (ii) condition/action triggering with user-modifiable parameters, and (iii) intertype correlation will determine whether anticipation of claim 1 holds or reduces to § 103. This is the reference I would obtain in full text first.

6. US 5,774,667 A — "Method and Apparatus for Managing Parameter Settings for Multiple Network Devices" (now RPX Clearinghouse)

• Filing date (verified): 1996-03-26 · Publication (training data): ~1998-06-30
• § 102 basis: § 102(e) (US filing 1996-03-26, before priority date 1996-06-27)
• Description: Centralized management of parameter settings across multiple network devices — i.e., configuration/parameter data authored centrally and pushed to distributed devices.
• Potential anticipation: Claim 5 (management-node configuration handling / filtered distribution of per-node parameter configuration) and the configuration aspects of Claim 1; weaker on the measurement/condition/action loop.

7. EP 0 609 990 A2 — (Fujitsu Services Ltd / ICL family; exact title not surfaced)

• Filing date (verified): 1993-02-04 · Publication (training data): ~1994-08-10
• § 102 basis: § 102(b) (published >1 year before 1996-06-27)
• Description: Fujitsu/ICL network-management patent from the same era as the Hacker "Operations Management" NPL (below) — likely covering centralized management of distributed systems. Exact title and content unverified in this session.
• Potential anticipation: Possibly Claim 1/Claim 2 (centralized management of distributed objects). Because the title/content is unverified, I cannot responsibly assign a specific claim; treat as a review item.

8. US 5,659,736 A — "Management Information Base and Method in an OSI Management System" (Mitsubishi Electric)

• Filing date (verified): 1993-11-28 · Publication (training data): ~1997-08-19
• § 102 basis: § 102(e) (US filing before 1996-06-27)
• Description: MIB construction/use in an OSI management system — object-class/instance modeling of managed resources.
• Potential anticipation: The object-type/MIB preamble concepts of Claim 1 and the object-description portions of Claim 2. Does not reach the autonomous-agent threshold/action mechanics; likely § 103 combination art for those limitations.

9. US 5,367,635 A — "Network Management Agent with User Created Objects Providing Additional Functionality"

• Filing date (verified): 1991-08-28 · Publication (training data): ~1994-11-22
• § 102 basis: § 102(b) (published before 1996-06-27)
• Description: A network-management agent (SNMP-era) that can be extended with user-created objects providing additional functionality — i.e., user-extensible agent behavior.
• Potential anticipation: Claim 1's "parameters, conditions and actions … modifiable by the user" limitation and Claim 3 (extensible agent structure). Strong § 102 candidate for the user-extensibility limitation; weak on autonomous multi-node distribution and intertype correlation.

10. WO 1992/05485 A2 — "Network Management System Using Model-based Intelligence" (Cabletron Systems)

• Priority date (verified): 1990-09-16 · Publication (training data): ~1992-04-02
• § 102 basis: § 102(b) (published before 1996-06-27)
• Description: The foundational Cabletron Spectrum patent: model-based network management using a virtual network machine, object models of managed resources, and intelligent correlation of events.
• Potential anticipation: Claim 1's object-model and correlation concepts (particularly "intertype correlation") and Claim 2 (management-node object display). It is a classic § 102(a)/(b) reference for model-based, object-oriented management, but its agent model (polling/event correlation at the manager) is the opposite of '664's autonomous local agents — which is precisely the problem the '664 specification says it solves.

11. US 5,684,967 A — "System and Method for Generalized Network Topology Representation" (Cabletron/Spectrum family)

• Filing date (verified): 1995-09-12 · Publication (training data): ~1997-11-04
• § 102 basis: § 102(e) (US filing before 1996-06-27)
• Description: Generalized representation of network topology for management display — hierarchical/object-based topology modeling.
• Potential anticipation: Claim 2 (graphical display of objects) and possibly the global-object/hierarchical display aspects (zoom on composite objects). Not a candidate for the autonomous-agent process claims.

12. US 5,751,964 A — "System and Method for Automatic Determination of Thresholds in Network Management" (Cabletron/Spectrum family)

• Filing date (verified): 1995-09-11 · Publication (training data): ~1998-05-12
• § 102 basis: § 102(e) (US filing before 1996-06-27)
• Description: Automatic computation/determination of thresholds used in network management condition evaluation.
• Potential anticipation: The threshold-condition aspects of Claim 1 (conditions "relative to predefined thresholds") and any dependent claim reciting threshold determination. This is the strongest reference for the threshold limitation specifically; combined with US 5,655,081 or WO 95/27249 it is potent § 103 art.

Non-patent literature (2 listed; only 1 fully surfaced)

NPL-1 — D. Hacker, "Operations Management," ICL Technical Journal, vol. 7, no. 4 (Nov. 1, 1991), pp. 741–750 (XP000268114)

• Date (verified from citation listing): 1991-11-01 · § 102 basis: § 102(b)
• Description: ICL's operations-management framework for distributed systems — background art on centralized management of distributed IT operations.
• Potential anticipation: Background for Claim 1's distributed-management premise; more likely § 103 context than a § 102 anticipation reference. (A second NPL is listed on the face of the patent but was not surfaced in my searches — unverified.)

Ranking of most relevant prior art

Bottom line

• Strongest § 102 anticipation candidates for independent Claim 1: US 5,655,081 (BMC) and WO 95/27249 — both disclose distributed, agent-side monitoring/control with per-resource-type intelligence; whether either alone discloses filtered configuration distribution, threshold-condition/action triggering, user-modifiable parameters, and intertype correlation must be confirmed against the full texts.
• Strongest § 102 candidates for Claims 2 and 4 (management-node side): US 5,802,146, US 5,684,967, and WO 92/05485.
• Strongest § 102 candidate for Claim 5 (configuration distribution): US 5,774,667.
• Strongest for the user-modifiability limitation: US 5,367,635.
• Most potent § 103 combinations would pair BMC (US 5,655,081) or WO 95/27249 with Cabletron's threshold art (US 5,751,964) and/or the MIB modeling of US 5,659,736.

Explicit uncertainty flags

• Publication dates marked "(training data)" should be verified against the USPTO/EPO before relying on a § 102(b) bar date.
• Exact titles/contents of EP 0 609 990 A2 and the second NPL were not verified.
• Claim mapping is provisional because the verbatim text of claims 1–5 was not available; if exact claim language is needed, pull the full text from the USPTO Patent Center or the Google Patents "Claims" tab for US6122664A.
• No litigation was found in the prior analysis, consistent with the patent's 2017 expiry; nothing in this citation analysis changes that.

I located the examiner-cited prior art (the "Patent Art" list) for US-6,122,664. Let me verify details on the most relevant references before building the obviousness analysis.

Obviousness Analysis — U.S. Patent 6,122,664 (35 U.S.C. § 103)

0. Scope, evidence base, and caveats

This analysis is built on the prior-art record actually cited on the face of US-6,122,664, retrieved from the Unified Patents "Patent Art" list for US-6122664-A (12 U.S./PCT/EP patent references plus non-patent literature). Two caveats that shape the analysis:

1. Claims 1–5 are reconstructed. As documented in the prior sections of this file, the verbatim text of independent claims 1–5 was not available in the source text; claims 6–20 were verified. The element mapping below therefore targets the claim scope as understood from the abstract, specification, and verified dependent claims (claims 6, 7, 9, 13–15). If exact claim language differs, the mapping must be re-checked against the USPTO full text.
2. All 12 references were cited by the examiner and the claims were allowed over them. That does not defeat a § 103 challenge — KSR combinations can rest on rationales the examiner never considered — but it means the strongest challenge is a combination theory, not any single reference, and it raises the burden of showing a concrete motivation to combine.

1. The prior-art record (all pre-1996, i.e., prior to the 1996-06-27 priority date)

2. Person of ordinary skill in the art (POSITA)

A POSITA circa mid-1996 would be a system/network-management software engineer or architect with 2–4 years' experience in distributed systems management, familiar with SNMP/CMIP manager-agent models, MIBs, client-server architectures, UNIX administration, event/threshold processing, and GUI console design. The art was crowded and mature: manager-agent monitoring with thresholding and GUI consoles was textbook material (as the patent's own Background concedes, SNMP get/set/trap managers and agents were the status quo the invention set out to improve).

3. Primary combination: BMC '081 + '7635 + '964 (+ '667) → claim 1 and the "autonomous agent" core

The heart of claim 1 (reconstructed) is: configure monitoring at the management node, distribute it to autonomous agents in each node, each agent having multiple type-specific modules that measure/collect parameters, test conditions against thresholds, trigger associated actions, all user-modifiable, with intertype correlation.

Element-by-element mapping:

Why this combination is obvious. BMC '081 alone nearly anticipates the autonomous-agent architecture; it lacks (i) the explicit "user-modifiable per-object parameter/condition/action" configurability, and (ii) explicit multi-condition/action threshold semantics. '7635 supplies (i) (it exists precisely to let users extend an agent's object/functionality set), and '964 supplies (ii) (thresholds are the standard trigger mechanism in the same field). Each reference addresses the same problem — efficient distributed monitoring with reduced manager involvement — and the combination is a simple aggregation of known, complementary components with predictable results (KSR v. Teleflex, 550 U.S. 398 (2007)). The patent itself admits the SNMP manager/agent model was prior art and that the field's known problem was traffic load and slow reaction; pushing decision logic to the agent (BMC '081) plus user-configurable thresholds/actions ('964, '7635) is the textbook solution a POSITA would arrive at.

4. Secondary combination for the management-node features (claim 2 and claim 5/6)

Claim 2's management node (GUI with icons/curves, configuration file, status files, trace/display files) maps almost verbatim onto BMC '081's console: object-icon GUI (module 50), object/state database 49, event log 52, and the data repository 208 ("trace" files). The configuration-file-plus-dynamic-modification concept maps to '667 (parameter settings management) and '7635 (user-created objects). No inventive step.

Claim 5/6's "generic management agent" — reading the configuration, syntactic/semantic analysis, merging the management-node configuration with specific-module configurations, and distributing a filtered configuration per node, handling malfunctioning and backup nodes — is the combination of:

• '667 (centralized parameter management/distribution to multiple devices),
• BMC '081 (per-agent knowledge modules loaded selectively),
• '684,967 (generalized/hierarchical network topology — needed to know which node hosts which object, and to re-route on failover),
• '802,146 or '491,796 (backup/failover consoles and remote maintenance operations).

Motivation. Distributing to each node only the configuration subset relevant to that node's objects ("filtered way") is an obvious bandwidth- and processing-saving design choice — the patent's own stated goal is reducing line traffic. Failover monitoring via a BACKUP_NODE is inherent in any high-availability management scheme ('802,146, '491,796). A POSITA merging a config manager ('667) with a topology model ('684,967) to produce per-node filtered configs is doing routine engineering, not inventing.

5. Combinations for the remaining verified dependent claims

• Claim 7 (generic agent of a monitored node: local-object-only processing, object/parameter/value/condition tables, basic functions, starting specific modules, configuring log scanning, evaluating multiple conditions, verifying module health): BMC '081 (agent tables, local-only operation, module/knowledge loading) + IBM '523 (generic collection subsystem for problem-determination data = log scanning) + WO-95/27249 (multiple-condition evaluation at agent level). Health-checking of daemons/modules was routine (e.g., UNIX inittab/mkitab resurrection, which the patent itself describes as ordinary).

• Claim 9 (specific-module operation: scan parameter config, sort by period, measure, store, trace/display, evaluate single conditions, initiate actions, handle real-time measurement requests): BMC '081's data repository + '964's threshold logic + '491,796's remote measurement commands. Period-sorted measurement scheduling is a standard timer/scheduler design.

• Claims 13–15 (cross-node parameter requests routed through the management node; paramlog/actionlog collection at the management node): '491,796 (remote management of diverse resources), BMC '081 (manager-side event cache 212 mirroring agent repositories), IBM '523 (collection subsystem). A POSITA routing a value request through the node that already holds the object/state database (BMC '081's manager) needs no invention.

6. The genuinely arguable features — and why they still fall under § 103

Four features are the best footholds for patentability; each still yields to combination:

1. "Filtered" per-node configuration distribution. Argued as the core advance. Response: '667 distributes parameter settings per-device; per-node filtering is the obvious application of a topology model ('684,967) to a config distributor ('667) to save traffic — the patent's own objective.

2. Global objects (aggregates of heterogeneous basic objects with inherited severity). Argued as novel "intertype" aggregation. Response: hierarchical/aggregate resource modeling was known ('684,967 "generalized network topology representation"; '491,796 "diverse information network resources"; OSI MIB scoping per '659,736). Inheriting a "worst-child" severity color to a parent icon is a standard alarm-correlation/summarization technique.

3. Backup-node quick-start monitoring. Argued as novel failover. Response: failover/backup consoles ('802,146) and configuration re-pointing are routine high-availability practice; the patent's own description ("a simple way to do this is to modify the configuration file") concedes it is a configuration substitution.

4. Multiple conditions evaluated at the management node across objects, and cross-node value requests. Argued as novel. Response: central evaluation of Boolean conditions over data gathered from multiple sources is the essence of the model-based/console systems ('WO-92/05485, BMC '081 event cache, WO-95/27249). The limitation that a condition "cannot contain more than ten single conditions" is an arbitrary, non-inventive design choice.

7. Counterarguments a patentee would raise (and the rebuttal)

• "The examiner considered all 12 references and allowed the claims." Rebuttal: Examiners frequently allow claims over the cited art where the specific combination theory was never developed; KSR expressly permits new combination rationales on the same record. The burden is simply to articulate the motivation (done above).
• "Secondary considerations" (commercial success, long-felt need, unexpected results). The record contains no evidence of commercial success (e.g., Bull's Ism/OpenMaster line), licensing, or unexpected results. Without such evidence, this factor is neutral, and the obviousness case stands.
• "The filtered distribution + autonomous local decisioning gives a surprising traffic/reaction-time improvement." Rebuttal: The improvement is the predictable consequence of moving decisions closer to the data — the exact benefit BMC '081 already claimed for autonomous agents. Magnitude of improvement, without unexpectedness of mechanism, is not a patentable distinction.

8. Bottom line

• Claim 1 (reconstructed): Very strong § 103 case — BMC '081 + '964 + '7635, with '667 as support. Every element maps to known references in the same field, combined for their known, predictable purposes.
• Claims 2–5 (reconstructed), 6–7, 9, 13–15 (verified): Strong case using the primary combination plus '667, '684,967, IBM '523, '491,796, '802,146, WO-95/27249, WO-92/05485, '659,736, EP-0,609,990, and the Hacker NPL.
• Most vulnerable claim features: "filtered distribution," "global object," "backup node," and "management-node-evaluated multiple conditions" — each is a known technique (per-device config distribution, hierarchical aggregation, failover, central Boolean evaluation) applied to a known architecture.
• Honest caveat: Because claims 1–5 could not be verified verbatim, this analysis is provisional on claim construction. If the issued claims contain a limitation not captured in the specification excerpts (e.g., a specific filtering algorithm or a specific correlation data structure), the mapping must be revisited.

Primary combination to assert: US-5,655,081 in view of US-5,367,635 and US-5,751,964 (with US-5,774,667 for the configuration-distribution element) — a POSITA combining the BMC autonomous-agent architecture with user-extensible agent objects and threshold-based condition/action processing would arrive at the claimed monitoring process as a matter of ordinary design choice, with a clear motivation: reducing manager-agent traffic and reaction time, the very problems the patent identifies as solved by prior SNMP-style systems.