US 7340597

I'll verify a few details against live sources before summarizing.

Let me pin down the independent claims (the provided text truncates at claim 15).

One more targeted check for the other independent claims.

Summary — US Patent 7,340,597 B1

Note on the searches: I searched for the literal identifier 7340597 (US 7,340,597 B1) and discarded results for superficially similar numbers (e.g., app. nos. 13/734,178 and 13/526,514 that surfaced in the results, and unrelated "securing a device" patents). I found no 2026 docket activity for this patent — the CAFC and PTAB activity associated with it is from 2015–2019. I flag that explicitly rather than inventing a 2026 docket.

Bibliographic data (from the authoritative patent text plus corroborating sources)

Google Patents (the authoritative text you supplied) gives filing 2003-09-19 and issue 2008-03-04; Unified Patents' portal lists 2003-09-18 / 2008-03-03 / expiration 2026-01-25, i.e., one day earlier throughout, most likely a time-zone/recordkeeping difference rather than a different patent. I did not find an authoritative source resolving which calendar date the USPTO treats as controlling.

Abstract (verbatim)

"A logging module is disclosed. A communications device can include, and so be made secure through the use of, the logging module. The logging module is configured to communicate information regarding a change to a configuration of a subsystem of the communications device."

Plain-language overview

The patent addresses a narrow problem: network devices are configured by software, so an attacker who compromises the software can silently reconfigure (or disable security on) the device. Cheriton's idea is a logging module — preferably implemented primarily in hardware — that sits inside the communications device, watches the configuration register/table state of a subsystem (e.g., a line-card/communications interface), and cannot be reconfigured by the device's own software. When the configuration changes, the module detects it and reports it off-box.

Key mechanism details from the specification:

• Log records describing each write to a hardware configuration register or table are packed into packet frames carrying a sequence number, sent to an assigned IP/Ethernet multicast address and protocol type. A reset resets the sequence number (e.g., to zero) and is itself logged as a configuration change.
• The device's software is blocked from transmitting packets using the logging module's address/protocol (preventing spoofing) and blocked from forwarding logging packets sourced from its own address.
• A periodic "heartbeat" is sent so monitors can detect a device that has gone silent.
• Rate-limiting of the logging traffic is fixed (e.g., ~3% of port bandwidth, or 3 Mbps on a 100 Mbps port) and changeable only by physical modification (e.g., a jumper on the supervisor card).
• Optional hardware-based digital signature / MAC (and optional encryption) with a key inaccessible to software.
• A security monitor subscribes to the multicast address, tracks sequence numbers, and flags missing numbers as a possible compromise — setting the device to "untrustworthy," notifying an administrator, and/or disconnecting it. Redundant monitors with majority voting are suggested to defeat false reports.
• The specification notes the logging could alternatively be done in software (accepting weaker guarantees), in which case the CLI should not permit disabling it.

Independent claims

The claim text you supplied is truncated at claim 15 (it labels the set "Claims (110)" but reproduces only claims 1–15), so I can quote only claim 1 verbatim from the authoritative text. Independent claim 1:

"1. An apparatus comprising: a communications device comprising: a subsystem; and a logging module, coupled to said subsystem, and configured to detect a change to a configuration of said subsystem of said communications device, and communicate information regarding said change to said configuration of said subsystem of said communications device."

Other independent claims — I can only partially verify these, and I flag the uncertainty:

• Claim 71 (appears verbatim in a Justia snippet): a communications device comprising a subsystem, a processor, a computer-readable medium, and computer code configured to cause the processor to detect a change in a configuration of the subsystem and communicate information regarding the change. (This is the software-oriented counterpart to claim 1.)
• Claim 84 (appears verbatim in a Justia snippet): a computer program product comprising a first set of instructions to detect a change in a configuration of a subsystem of a communications device, a second set of instructions to communicate information regarding the change, and computer-readable media encoding the product.
• Claim 29, claim 39, and claim 63 are treated by the PTAB and Federal Circuit as the independent broadcast/method claims (claim 64 depends from 63; claims 40–42 depend from 39; claims 73 and 86 depend from the 71 and 84 families respectively). I do not have the verbatim claim text for 29, 39, or 63, and I cannot rule out additional independent claims among claims 16–110. Do not treat my characterizations of those three as quoted language.

Litigation / validity history (all pre-2026; no 2026 CAFC docket found)

• N.D. Cal. 5:14-cv-05343, Cisco Systems v. Arista Networks (filed Dec. 2014); a related case is indexed as 5:16-cv-00923 (Google Patents lists this as "California Eastern District Court"; the Stanford NPE database lists it as N.D. Cal. — an unresolved venue discrepancy). Google Patents also links ITC investigation 337-TA-944.
• IPR2015-00978, Arista Networks v. Cisco Systems (petition filed April 1, 2015) on claims 1, 14, 15, 29, 39–42, 63, 64, 71–73, 84–86.
  • Final Written Decision, Sept. 28, 2016: claims 1, 14, 39–42, 71, 72, 84, 85 unpatentable; claims 29, 63, 64, 73, 86 (and claim 15) not proven unpatentable.
• CAFC Nos. 2017-1525 and 2017-1577, Arista Networks, Inc. v. Cisco Systems, Inc., 908 F.3d 792 (Fed. Cir. Nov. 9, 2018). Held: (1) assignor estoppel is not available in IPR (affirming Cisco's cross-appeal refusal); (2) the Board erred in construing "broadcast"/"broadcasting"; the court construed "broadcast" = "a transmission to one or more devices using a multicast address" and remanded.
• Final Written Decision on Remand, July 11, 2019: claims 29, 63, 64, 73, 86 not unpatentable — these claims survived; the previously invalidated claims (1, 14, 39–42, 71, 72, 84, 85) stand cancelled.
• Other CAFC appeals Google Patents associates with this patent number: 16-2539, 16-2563, 17-1046, 17-1281 (I did not independently verify these are directed to the '597 patent rather than sibling Cisco/Arista patents).

Uncertainty statements

1. The claim listing in the supplied authoritative text stops at claim 15 of 110 — so my treatment of independent claims 29, 39, and 63 is inferential.
2. The filing/issue/expiration dates differ by one day between Google Patents and Unified Patents; I have not resolved this.
3. The 5:16-cv-00923 venue is reported inconsistently (Northern vs. Eastern District of California).
4. No 2026 CAFC docket for this patent appeared in my searches; the latest activity I found is the 2019 PTAB remand decision.

I'll research litigation involving US patent 7340597 across patent litigation databases and court records.

Let me dig deeper into the specific cases and PTAB/ITC proceedings.

Litigation involving US Patent 7,340,597

Patent background (for context): US 7,340,597 B1, "Method and Apparatus for Securing a Communications Device Using a Logging Module," inventor David R. Cheriton, originally assigned to Cisco Technology, Inc., later transferred to Cisco Systems, Inc. Priority/filing date 2003-09-19; granted 2008-03-04. Google Patents lists the patent as expired (adjusted expiration 2026-01-26).

All known disputes involving the '597 patent arise from the Cisco–Arista Networks litigation campaign. There is no NPE / non-practicing-entity litigation on this patent — the Stanford NPE database lists only the two Cisco/Arista district-court cases.

1. Cisco Systems, Inc. v. Arista Networks, Inc. (district court)

2. Certain Network Devices, Related Software and Components Thereof (I) — ITC § 337 investigation

Appeal from the ITC: Cisco Systems, Inc. v. International Trade Commission, Appeal No. 2016-2563 (Fed. Cir.), reported at 873 F.3d 1354 (Fed. Cir. 2017). Cisco challenged, inter alia, the Commission's determination that Arista's accused products do not infringe the '597 patent. The Federal Circuit affirmed the ITC's non-infringement finding on the '597 patent.

3. Arista Networks, Inc. v. Cisco Systems, Inc. (district court)

4. Inter partes review — Arista Networks, Inc. v. Cisco Systems, Inc.

Additional Federal Circuit dockets linked to this patent on Google Patents

The Google Patents litigation listing also shows CAFC dockets 17-1046, 17-1281, 16-2539 and the IPR-remand docket. These appear to be part of the broader Cisco–Arista appellate cluster (various patents in the same family/campaign), which Google Patents aggregates under this patent's "family litigation." I did not independently verify that each of these appeals specifically concerns the '597 patent; treat them as campaign-level, not patent-specific, unless confirmed in the appellate dockets.

Summary of the '597 patent's litigation posture

• Cisco (owner) asserted the '597 patent against Arista Networks in the N.D. Cal. district-court case (5:14-cv-05343) and in ITC Inv. No. 337-TA-944.
• Arista (challenger) attacked it via IPR2015-00978, invalidating several claims, with the validity of the surviving claims left on remand after the Federal Circuit's 2018 decision.
• Cisco did not prevail on the '597 patent at the ITC (no violation found; affirmed on appeal in Cisco v. ITC, 873 F.3d 1354).
• The flagship legal holding associated with this patent is Arista Networks v. Cisco (Fed. Cir. 2018), establishing that assignor estoppel does not apply in IPR proceedings.
• The overall Cisco–Arista disputes were resolved by a global settlement announced in August 2018.

Caveats: Case-number renderings differ between the Stanford NPE database (4:14-cv-05343) and Docket Alarm (5:14-cv-05343), and Google Patents mislabels the venue for 5:16-cv-00923 as the Eastern District of California. I was unable to access PACER directly to confirm the precise '597 disposition in the district-court case or the exact 2016 filing date for 5:16-cv-00923; those items are flagged above as lower-confidence. I found no litigation involving the '597 patent by any party other than Cisco and Arista Networks.

I'll flag immediately: your structured block says no PTAB activity, but the patent record itself lists a PTAB case. Let me verify.

PTAB Activity on US 7,340,597 — Analyst Findings

Flag on the structured data first. The "PTAB proceedings on file" block says ODP returns no AIA trials. That is wrong for this patent. The patent record itself, in the same prompt, lists a PTAB case ("PTAB case IPR2015-00978 filed (Final Written Decision)"). The Board's own paper (Paper 57) confirms it. I treat IPR2015-00978 as the one and only AIA trial on the '597 patent, and I flag the ODP gap as a data-integrity problem you should not rely on.

Second flag. The Google Patents page's litigation block carries a "Unified Patents PTAB Data" attribution next to the PTAB entry. That is a data-source license notice, not the petitioner. The petitioner in IPR2015-00978 was Arista Networks, Inc. Unified Patents is not a party to this proceeding. Do not let a defense team argue "Unified Patents lost an IPR" — that framing is factually wrong.

Proceedings overview

There is one AIA trial proceeding on US 7,340,597 — IPR2015-00978 (Arista Networks, Inc. v. Cisco Systems, Inc.) — which was instituted on all challenged claims except claim 15, went to a Final Written Decision (2016-09-28) that invalidated claims 1, 14, 39–42, 71, 72, 84 and 85 and sustained claims 29, 63, 64, 73 and 86, was affirmed-in-part / reversed-in-part and remanded by the Federal Circuit (2018-11-09), and produced a Final Written Decision on Remand (2019-07-11) that sustained claims 29, 63, 64, 73 and 86. Net defensive posture: the patent is materially narrowed — every independent claim (1, 39, 71, 84) was canceled — but the five surviving dependent claims were twice sustained on substantially the same prior art, and the Board's sustaining rationale is grounded in a Federal Circuit claim construction, so a fresh § 103 attack on those five claims is not a walkover.

IPR2015-00978 — Arista Networks, Inc. v. Cisco Systems, Inc.

• Type: Inter Partes Review (35 U.S.C. §§ 311–319)
• Filed: 2015-04-01 per the Federal Circuit's opinion and PTAB docket (the FWD on Remand at Paper 57 recites "May 1, 2015"; the discrepancy is unreconciled — treat 2015-04-01 as the docket date). Petition served within one year of Cisco's 2014-12-09 service of the district court complaint, satisfying § 315(b).
• Status (verbatim framing from the structured data): "PTAB case IPR2015-00978 filed (Final Written Decision)." Plain-English gloss: trial completed, judgment entered, appealed, remanded, and re-decided on remand; now final. No active trial stage remains.
• Judge panel:
  • Institution decision (Paper 7): Bryan F. Moore, Matthew R. Clements, Peter P. Chen (opinion by Chen).
  • Original Final Written Decision (Paper 32): Moore, Clements, Chen (authored by Chen).
  • Final Written Decision on Remand (Paper 57): Bryan F. Moore, Miriam L. Quinn, Patrick M. Boucher (authored by Quinn) — note the panel composition changed between the original FWD and the remand FWD.
• Petition grounds: 16 claims challenged — 1, 14, 15, 29, 39–42, 63, 64, 71–73, 84–86. Four grounds:
  • Ground 1 (§ 102 anticipation by Sheikh): claims 1, 14, 15, 29, 39–42, 63, 64, 71–73.
  • Ground 2 (§ 103 obvious over Sheikh): claims 71–73, 84–86.
  • Ground 3 (§ 103 obvious over Sheikh + Endicott): claims 84–86.
  • Ground 4 (§ 103 obvious over Sheikh + Iwayama): claims 29, 63, 64, 73, 86.
• Institution decision: Partially instituted, 2015-10-06. The Board instituted on claims 1, 14, 29, 39–42, 63, 64, 71–73, and 84–86, but denied institution as to claim 15, finding Sheikh's logging module does not restrict a change to its configuration by the communications interface — the limitation claim 15 requires. The Board also exercised § 42.108(b) / § 315(d) discretion to decline the additional obviousness ground (Ground 3, Sheikh + Endicott) as redundant, noting review "will proceed on all of the claims challenged by Petitioner in its other asserted ground."
• Claim construction at institution (this is the pivot of the whole case): The Board rejected both parties' constructions of "broadcast" and adopted the ordinary-and-customary meaning — "a transmission of a message simultaneously to all destinations in a network." Both parties objected that this excluded the patent's only disclosed broadcast embodiment (multicasting via a multicast address).
• Final Written Decision (Paper 32, 2016-09-28) — verdict at claim level:
  • Unpatentable / canceled: claims 1, 14, 39, 40, 41, 42, 71, 72, 84 and 85 ("invalidated claims 1, 14, 39–42, 71, 72, 84, and 85 as anticipated or obvious").
  • Not shown unpatentable / sustained: claims 29, 63, 64, 73 and 86.
  • Reasoning in brief: the five surviving claims all require communicating the change by broadcasting. The Board held Sheikh does not disclose broadcasting (an alert to a system administrator is not a broadcast) and — under its "all destinations in a network" construction — Arista failed to show Sheikh + Iwayama rendered the broadcast-dependent claims obvious.
  • Assignor estoppel: Cisco argued Arista (founded by Dr. David Cheriton, the named inventor, who assigned the '597 patent to Cisco, and who was Arista's Chief Scientist and one of its largest shareholders) was estopped from challenging validity. The Board declined to apply assignor estoppel in IPR (citing then-precedential Board authority), and this refusal was preserved for appeal.
  • Request for rehearing denied: 2017-01-13 (Paper 37); Arista's notice of appeal followed 2017-01-23 (Paper 38).
• Settlement / termination: No. The IPR ran the full course to FWD and then to a remand FWD; there was no adverse judgment, no settlement termination, and no pre-institution termination. (The underlying Cisco v. Arista district court and ITC matters did later settle in 2018, but that did not terminate or moot the IPR.)
• Appeal to the Federal Circuit: Yes.
  • Docket numbers: 2017-1525 (Arista's appeal) and 2017-1577 (Cisco's cross-appeal), decided together: Arista Networks, Inc. v. Cisco Systems, Inc., 908 F.3d 792 (Fed. Cir. Nov. 9, 2018). Panel: Prost (C.J.), Schall, Chen.
  • Issues: (1) Arista — error in the Board's construction of "broadcast"/"broadcasting," which drove the rejection of Arista's obviousness challenge to claims 29, 63, 64, 73, 86; (2) Cisco (cross-appeal) — reviewability of the Board's refusal to apply assignor estoppel, and whether the Board should have applied it on the merits.
  • Disposition: AFFIRMED-IN-PART, REVERSED-IN-PART, AND REMANDED. Affirmed — assignor estoppel has "no place" in IPR proceedings; § 311(a)'s text ("a person who is not the owner of a patent may file") is unambiguous, and the issue is reviewable and not barred by § 314(d). Reversed — the Board's construction excluded the only disclosed embodiment and could not stand; the Court construed "broadcast" = "a transmission to one or more devices using a multicast address" and "broadcasting" = "transmitting to one or more devices using a multicast address." Remanded for the Board to assess, in the first instance, the claims raised in Arista's appeal under the correct construction.
  • Links: CourtListener — https://www.courtlistener.com/opinion/[4552226](/patent/4552226)/arista-networks-inc-v-cisco-systems-inc/ ; FindLaw full text — https://caselaw.findlaw.com/court/us-federal-circuit/[1962503](/patent/1962503).html
  • Caveat for the file: one third-party docket aggregator lists the '597 appeal as "2017-1046." The Federal Circuit's own caption and the parties' briefs control: 2017-1525 / 2017-1577. The '597 patent page lists several other CAFC dockets (16-2539, 16-2563, 17-1046, 17-1281) that belong to the other Cisco patents in the same Arista/Cisco IPR campaign.
• Final Written Decision on Remand (Paper 57, 2019-07-11):
  • Panel: Moore, Quinn, Boucher (opinion by Quinn).
  • Arista renewed only its § 103 obviousness ground over Sheikh and Iwayama as to claims 29, 63, 64, 73 and 86.
  • Holding, verbatim: "Petitioner has not met its burden of showing, by a preponderance of the evidence, that claims 29, 63, 64, 73, and 86 of the '597 patent are unpatentable."
  • Net effect: even under the patentee-favorable (narrower) Federal Circuit construction that "broadcast" requires a multicast address, Arista could not prove Sheikh + Iwayama disclosed the broadcast limitations. The five surviving claims survived a second, post-remand merits pass.
  • Link: Paper 57 PDF via the case docket — https://www.docketalarm.com/cases/PTAB/IPR2015-00978/Inter_Partes_Review_of_U.S._Pat._7340597/docs/07-11-2019-Board/Order-57-Final_Written_Decision_on_Remand___35_USC_318_and_37_CFR_4273.pdf
• Further appeal of the remand FWD: Not confirmed. The public docket I can see terminates at Paper 57 (2019-07-11) with the CAFC mandate recorded at Paper 56 (2019-04-23). I found no evidence of a second appeal from the 2019 remand decision, and no subsequent PTAB paper. I state that as absence of evidence, not proof of absence.
• Defensive value: Mixed, but net-helpful. The four independent claims — 1, 39, 71 and 84 — are canceled, so any infringement theory built on them is dead on arrival and, if asserted, is sanction-exposed. But claims 29, 63, 64, 73 and 86 are twice-standing survivors, and they are the narrow, broadcast/multicast-specific claims. An IPR attack modeled on Arista's (Sheikh alone, or Sheikh + Iwayama) has already failed twice — including after the Federal Circuit gave the challenger a more favorable construction than the Board's. Re-running that play is not a strategy.

Strategic summary

Canceled vs. sustained vs. untested. Claims 1, 14, 39, 40, 41, 42, 71, 72, 84 and 85 are canceled (held unpatentable in the 2016 FWD; Cisco's cross-appeal did not disturb that — the cross-appeal was directed only at assignor estoppel, and the claims Cisco lost were not resurrected at the Federal Circuit or on remand). Claims 29, 63, 64, 73 and 86 are sustained — found not unpatentable in 2016 and again on remand in 2019. Claim 15 was challenged but never instituted and is therefore untested by the PTAB. Everything else (claims 2–13, 16–28, 30–38, 43–62, 65–70, 74–83, 87–110) was never challenged and remains untested at the Board.

A structural wrinkle you must run down before advising a client. Every sustained claim is a dependent claim whose base claim was canceled: claim 29 depends from canceled claim 1; claim 63 depends from canceled claim 40; claim 73 from canceled claim 72; claim 86 from canceled claim 85 (claim 64 depends from claim 63). The Board sustained those dependents on the merits and never addressed the canceled-parent problem. I cannot confirm from the records available to me whether Cisco corrected, reissued or statutorily disclaimed anything to cure this, and I will not guess. It is, however, the single highest-value question for a defense team: if the surviving claims cannot be enforced as written, the practical exposure collapses even further. Note also that the patent record lists an adjusted expiration of 2026-01-26, which means the patent term has now run — leaving only past damages in play for any live dispute.

Estoppel landscape. § 315(e)(2) estoppel attaches to Arista Networks and its real parties-in-interest and privies, not to the world. Arista is barred in civil actions from asserting invalidity grounds it raised or reasonably could have raised in IPR2015-00978 — which covers Sheikh, Iendicott and Iwayama-based § 102/§ 103 grounds on the instituted claims. But: (a) claim 15 was not instituted, so estoppel does not reach grounds on that claim; (b) claims never challenged are outside estoppel entirely; and (c) a new defendant carries no estoppel at all — it inherits only the preclusive effect of the judgments, i.e. the cancellation of claims 1, 14, 39–42, 71, 72, 84 and 85, which is absolute against Cisco everywhere. For a fresh defendant, the available grounds are: any art not before the Board on the five sustained claims; any § 112 or § 101 theory (neither was litigated); and any § 102/§ 103 art on the untested claims. The constraint is practical, not statutory: § 315(b)'s one-year bar from service of a complaint, and the Board's § 325(d) / General Plastic discretion to deny follow-on petitions against a patent that already has a final written decision.

Pattern signals. This was a single-petitioner, single-proceeding IPR — Arista did not file multiple IPRs against the '597 (it did file a broad parallel campaign, IPR2015-00973 through -01049 and IPR2016-00244/-00301/-00309, against other Cisco patents). No defensive aggregator is in the chain; the "Unified Patents" label on the Google Patents page is source attribution only. The patent owner (Cisco) litigated hard and cross-appealed, and won on the cross-appeal — assignor estoppel is unavailable in IPR because of this case, which remains the controlling precedent (later narrowed in the district court context by Minerva Surgical v. Hologic, 141 S. Ct. 2298 (2021), which did not disturb the IPR holding). The surrounding dispute was Cisco v. Arista (N.D. Cal. 5:14-cv-05343) and ITC Inv. No. 337-TA-944, in which the ALJ/Commission found assignor estoppel did bar Arista — a forum split the Federal Circuit expressly called an intentional congressional choice.

Recommended next steps

1. If you are a defendant and the demand letter cites claims 1, 14, 39–42, 71, 72, 84 or 85 — say so immediately. Those claims are canceled. The cancellation is the judgment of the Board as sustained on appeal. Any infringement theory built on them is untenable; quote the disposition from the 2016 FWD (Paper 32 at 26) and the remand order (Paper 57). Link the FWD on remand: https://www.docketalarm.com/cases/PTAB/IPR2015-00978/Inter_Partes_Review_of_U.S._Pat._7340597/docs/07-11-2019-Board/Order-57-Final_Written_Decision_on_Remand___35_USC_318_and_37_CFR_4273.pdf
2. Run the canceled-parent issue down before assessing the exposure from claims 29, 63, 64, 73 and 86. I could not confirm how Cisco handled the fact that each surviving claim depends from a canceled base claim, and no correction/reissue/statutory disclaimer is visible in the records I reviewed. Treat this as an open, high-value item — not as a settled win.
3. Confirm the absence of a second appeal. I found no second CAFC appeal from the 2019-07-11 remand FWD and no post-2019 PTAB activity, but the ODP block already under-reported this case once. Pull the live docket at PTAB Center (https://ptacts.uspto.gov/) for IPR2015-00978 and the Federal Circuit docket for 2017-1525 / 2017-1577 before relying on "final and unappealable."
4. If you contemplate your own IPR: the § 315(b) one-year clock starts on service of the complaint, and General Plastic / § 325(d) discretion is a real risk on a patent with an existing FWD. Do not propose Sheikh alone or Sheikh + Iwayama on claims 29, 63, 64, 73 or 86 — that exact combination failed in 2016 and again in 2019, the second time under a construction the challenger won on appeal. Target the untested claims (2–13, 16–28, 30–38, 43–62, 65–70, 74–83, 87–110), claim 15, or § 101/§ 112 theories that were never adjudicated.
5. No pending trial-stage milestones exist. There is no institution-decision deadline, no oral hearing and no statutory 1-year FWD due date to track — the sole proceeding is closed. If your case is about past damages, the relevant dates are the 2016-09-28 original FWD and the 2019-07-11 remand FWD.

Confidence notes. I am confident about: the existence and number of the proceeding; the petitioner identity (Arista); the claim-level results in both FWDs; the institution date and the claim-15 denial; the Federal Circuit docket numbers and disposition; and the remand panel. I am not confident about: the precise petition filing date (04-01 vs. 05-01); whether any further appeal or PTAB paper exists after 2019-07-11; and how the canceled-parent defect in the surviving dependent claims was or was not cured. Do not state those three as fact without primary-source verification.

I'll verify the assignment reel/frame details and litigation posture before writing.

Let me narrow down the specific reel/frame numbers and the inventor's employment context.

Let me try once more to pin the inventor-assignment reel/frame and check for any post-2014 transfers.

Inventors

Dr. David R. Cheriton — sole named inventor.

• Employer at time of filing: Cisco Technology, Inc. (the original assignee). Both the ITC's Certain Network Devices opinion and the Federal Circuit record state Cheriton "was employed by Cisco" at the time of invention and assigned the invention to Cisco "while he was employed there."
• Background: Cheriton is a Stanford University computer-science professor who worked with Cisco. He is also the co-founder (with Andreas Bechtolsheim) of Granite Systems (acquired by Cisco, 1996) and later of Arista Networks.
• Unusual pattern — flagged: This patent does show the "inventor departs the original assignee" pattern the task asks about, and it is unusually consequential here. Cheriton assigned the '597 invention to Cisco and then left Cisco to co-found Arista Networks — an eventual Cisco competitor in Ethernet switching — reportedly alongside roughly a dozen other Cisco employees (The Register, 2018-11-14). Cheriton served as Arista's Chief Scientist until resigning in March 2014. Unlike the classic fire-sale precursor, this departure did not lead to a transfer of the patent; it led to the inventor's new employer challenging his own assigned patent (Arista v. Cisco, IPR2015-00978). The inventor's exit is therefore a litigation fact here, not an ownership-transfer fact. I did not find a source giving a precise Cisco departure date, so I cannot state whether it fell within exactly 12 months of the 2003-09-19 filing — I am flagging the pattern, not asserting the 12-month window.

No other inventors appear in the authoritative text, Google Patents, Unified Patents, or the litigation record.

Original assignee

Cisco Technology, Inc. (a California corporation; principal place of business 170 West Tasman Drive, San Jose, CA 95134 per Cisco's standard assignment form).

• Product embodying the claims: Yes, within Cisco's line. The specification itself points to the architecture of the Catalyst 4K Supervisor III as the class of device the invention is designed around ("similar to that provided in the Catalyst 4K Supervisor III"), and the ITC/N.D. Cal. litigation concerned Arista's competing switch products. Cisco asserted the patent against an actual commercial competitor, not a random defendant.
• Primary line of business: Network equipment — routers, switches, and enterprise data-center networking.
• Current status: Operating. Cisco Systems, Inc. is a large, publicly traded NYSE company (SEC registrant). The patent is now expired (Google Patents adjusted expiration 2026-01-26; Unified Patents lists 2026-01-25), so it is no longer an enforceable asset, though it remained relevant to the litigated back-damages history through 2019.
• Note on the two Cisco entities: "Cisco Technology, Inc." is the invention-capture subsidiary that takes inventor assignments; "Cisco Systems, Inc." is the selling/commercial entity and was the named complainant in the ITC investigation and the patent owner in the IPR. The 2014 recording moved title from the former to the latter (see below). This is a routine intra-Cisco title-cleanup, not a sale.

Assignment timeline

The authoritative patent text (Google Patents legal events) exposes four assignment-relevant recorded events. Two caveats up front:

• Google Patents does not render the reel/frame for the 2003 inventor assignment in the text I have, and I could not retrieve the recorded cover sheet for it. The ITC refers to this document only as the "'597 Assignment Reel" (exhibit JX-0016, at page 002). I will not invent a reel/frame number for it.
• The only reel/frame number I can state with confidence is 034523/0654, which appears by reference inside the text of the 2014-12-18 corrective assignment ("PREVIOUSLY RECORDED AT REEL: 034523 FRAME: 0654"). That is the reel/frame of the 2014-12-03 recording.

• 2003-09-19 (executed) / recorded on or about 2003-09-23 — Reel/frame not stated in available sources (ITC "JX-0016, '597 Assignment Reel")

  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: David R. Cheriton (sole inventor)
  • Assignee: Cisco Technology, Inc.
  • Correspondent: Not determinable from the sources I could reach (Google Patents legal events and Unified Patents do not surface the correspondent for this record). I did not retrieve the recorded cover sheet. I am not attributing a correspondent based on Cisco's other patents.
  • Context: Original employment/invention assignment to Cisco; notarized, and per the ITC recorded in the PTO four days after signing (i.e., ~2003-09-23). The ITC found the assignment conveyed "the entire right, title and interest," satisfying the assignor-estoppel analysis.

• 2014-12-03 (recorded) — Reel 034523 / 0654

  • Conveyance: Assignment — "Confirmation of Assignment"
  • Assignor: Cisco Technology, Inc.
  • Assignee: Cisco Systems, Inc.
  • Correspondent: Not determinable from available sources; the reel/frame is confirmed only indirectly (it is the record the December 18 corrective assignment purports to correct).
  • Context: Internal corporate reorg / title consolidation within the Cisco family — Cisco Technology, Inc. (invention holder) confirming transfer to Cisco Systems, Inc. (operating and litigating entity). Not a sale to a third party.

• 2014-12-18 (recorded) — Reel/frame not stated in available sources

  • Conveyance: Corrective Assignment — "Corrective Assignment to correct the signature page inside assignment previously recorded at Reel 034523, Frame 0654"
  • Assignor: Cisco Technology, Inc.
  • Assignee: Cisco Systems, Inc.
  • Correspondent: Not determinable from available sources.
  • Context: Change-of-record/clerk correction only — fixing the signature page of the 2014-12-03 recording. No new consideration, no change in beneficial ownership.

Also appearing on the Google Patents legal-events list (not an assignment of this patent): a 2003-09-30 "Priority to US 10/675,506," which issued as US 7,900,251 B1. That is a related family member filed 11 days after the '597 application, not a transfer of '597 title. Do not read it as an ownership event.

No post-2014 assignment exists. There is no recorded transfer to any NPE, aggregator, or other third party. The chain terminates at Cisco Systems, Inc. Google Patents' legal-events list, Unified Patents' patent page (Original Assignee: Cisco Technology Inc; Current Assignee: Cisco Systems Inc; Parent Company: Cisco), and the ITC record are all consistent on this point.

Verification: https://assignmentcenter.uspto.gov/ (search patent number 7340597) and the indexed mirror https://assignment.uspto.gov/patent/index.html

Timeline diagram

timeline
    title Ownership of US 7340597
    2003 : Cheriton assigns to Cisco Technology
         : Application filed Sep 19
         : Recorded about Sep 23
    2008 : Patent issued Mar 4
    2014 : Cisco Technology confirms assignment to Cisco Systems
         : Dec 3 recorded as reel 034523 frame 0654
         : Dec 18 corrective assignment recorded
         : Cisco sues Arista in Dec
    2015 : Arista files IPR against the claims
    2016 : PTAB final written decision
    2018 : CAFC construes broadcast and remands
    2019 : Remand decision upheld claims 29 63 64 73 86
    2026 : Patent expired Jan 26

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The only post-inventor transferee is Cisco Systems, Inc. (reel 034523/0654, recorded 2014-12-03), an operating NYSE company and the patent's own litigating entity. No LLC with an "IP / Patents / Licensing / Holdings / Ventures" suffix appears; there is no registered-agent-service address and no single-purpose entity anywhere in the chain.

2. Known asserter in the chain — not present. Neither assignee (Cisco Technology, Inc.; Cisco Systems, Inc.) matches any public NPE list. Cisco is the target of IPR2015-00978, not the filer, and it asserted the patent offensively against a market competitor (Arista) — the definition of an operating-company campaign.

3. Repeat correspondent across the chain — not determinable (no finding). The Assignment Center exposes a correspondent per record, but neither Google Patents' legal events nor Unified Patents surface it for this patent, and I could not retrieve the reel 034523/0654 cover sheet or the 2003 cover sheet. I will not infer a correspondent from Cisco's other assignment filings. This is an information gap, not a negative finding — verify directly on Assignment Center.

4. Cascading transfers — not present. There is exactly one inter-entity transfer in the patent's 23-year life (Cisco Technology → Cisco Systems, 2014), plus a clerical correction four days later. Nothing resembles chained LLC-to-LLC transfers, and there are no shared-correspondent or common-principal patterns to evaluate.

5. Pre-litigation transfer — not present, with a note. The 2014-12-03 Cisco Technology → Cisco Systems recording (reel 034523/0654) lands in the same month as the December 2014 N.D. Cal. complaint, which superficially resembles a pre-suit venue/standing move. It is not the signal, however: it is styled a "Confirmation of Assignment" between two entities in the same corporate family, confirming a pre-existing transfer rather than creating a new one — Cisco Systems was already the complainant of record. Worth noting only as a plausible reason the paperwork was updated when it was.

6. Bankruptcy fire-sale — not present. Cisco has not filed Chapter 7/11, and no proceeding appears anywhere in the record.

7. Privateering — not present. Cisco retained and itself asserted the patent. There is no transfer to a third-party asserter, and no SEC disclosure (10-K/8-K) or Patent Progress/EFF coverage of a Cisco-to-NPE transfer exists for this patent. The IPR was filed by Arista, a Cisco competitor defending against Cisco's own suit.

8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. (Unified Patents appears in this record only as an IPR petitioner against other patents and as a data source — it is not an assignee here.)

Verdict

Operating-company assertion.

Cisco owns the patent from invention (David R. Cheriton → Cisco Technology, Inc., recorded ~2003-09-23) through the present, with the sole subsequent change being an intra-family Confirmation of Assignment to Cisco Systems, Inc. at reel 034523/0654 (recorded 2014-12-03, corrected 2014-12-18). Cisco is a large operating manufacturer whose own specification builds on its Catalyst 4K Supervisor III architecture, and it asserted the '597 patent against a direct competitor, Arista Networks, in both the N.D. Cal. action and ITC Investigation 337-TA-944. Zero NPE signals are present; the only anomalous feature in this chain — the inventor leaving to co-found the defendant — runs against an NPE reading, since it produced a validity challenge rather than an ownership transfer.

Caveats: (i) the reel/frame for the 2003 inventor assignment is not exposed in the sources I could reach, and I have not fabricated one; (ii) correspondent data was unavailable to me for every record, so signal 3 is unresolved rather than negative; (iii) the patent expired 2026-01-26, so any ownership interest is now residual.

I'll search for the patent's cited references (prior art) before analyzing them.

Now let me find the references cited specifically on the face of the patent (the "References Cited" / Citations section).

Let me try FreePatentsOnline/Justia which reproduce the "References Cited" table.

Note on scope and sources. I searched the literal identifier 7340597 and only used results for US 7,340,597 B1 ("Method and apparatus for securing a communications device using a logging module," Cheriton, Cisco). I did not substitute similar numbers. I also checked for the two different things that "patent citation for 7340597" can mean:

1. the front-page "References Cited" list of record in the patent itself, and
2. the prior art actually relied on against this patent in the PTAB/courts.

I retrieved solid evidence for (2) and only partial evidence for (1). I say so explicitly below rather than filling gaps.

1. What I could and could not verify about the face-of-patent citations

Because I can't ground (1), I will not invent it. The remainder of this memo addresses (2), which is the prior art that was actually tested against the claims and is therefore the materially relevant prior art for this patent. If you need the strict front-page reference list, that would require pulling the patent PDF's cover page or the USPTO PatentCenter "References Cited" tab.

Cross-reference / consistency check with the prior sections: my prior-generated section characterized the independent claims (1, 29, 39, 63, 71, 84) and the IPR outcome. The IPR institution decision now in hand corroborates it: "Claims 1, 39, 71, and 84 of the '597 patent are independent" (quoting the Board's institution decision) — i.e., no contradiction, but note the Board named only 1, 39, 71, 84 as independent in that sentence while claim 29 was nonetheless challenged. That is a nuance, not a conflict.

2. The prior art of record against US 7,340,597 (IPR2015-00978)

Source: Arista's exhibit list and the Board's institution decision in IPR2015-00978 (Finnegan's posted institution decision; Unified Patents/docketalarm exhibit lists). Petitioner = Arista Networks, Inc.; Patent Owner = Cisco Systems, Inc.

2(a) The § 102 anticipation reference — the single most relevant prior art

Bottom line: Sheikh is the only § 102 reference asserted against the '597 patent; it is therefore the answer to "which reference potentially anticipates." Its anticipation of claim 1 and the 39/71/84 families was ultimately sustained; its reach to the "broadcast"-reciting claims (29, 63, 64, 73, 86) failed, and the failure is entangled with claim construction, not merely with the reference's disclosure.

2(b) § 103 combination references (not § 102 anticipators)

I flag these as obviousness art. Neither is alleged to anticipate any claim; each was combined with Sheikh.

The Iwayama combination is directly tied to the Federal Circuit's reversal: the court held the Board's construction of "broadcast"/"broadcasting" was wrong and construed "broadcast" = "a transmission to one or more devices using a multicast address" (Arista Networks, Inc. v. Cisco Systems, Inc., 908 F.3d 792 (Fed. Cir. Nov. 9, 2018)), then remanded — which is why claims 29/63/64/73/86 survived in 2019.

2(c) References appearing on Arista's exhibit list but not shown as an instituted ground

These were made of record as exhibits. Based on the grounds table in the institution decision (which lists only Sheikh, Endicott, and Iwayama), I could not confirm that these were the substantive grounds; treat the following as supporting/background or redundant art unless a fuller record shows otherwise:

2(d) Non-patent literature cited as exhibits

• Newton's Telecom Dictionary (8th ed. 2002) — used for claim-construction support (e.g., "broadcast").
• The Authoritative Dictionary of the IEEE Standards Terms (7th ed. 2000) — same purpose.
• John P. van Gigch, System Design Modeling and Metamodeling (1991) — background/theory.

These are dictionary/technical references, not § 102 art.

3. Claim-by-claim anticipation map (the actionable answer)

(Note the asymmetry: the Board's institution text names 1, 39, 71, 84 as the independent claims, and the "84–86" family is challenged only under § 103, not § 102.)

4. Explicit uncertainty statements

1. Face-of-patent citations not retrieved. I could not obtain the patent's own "References Cited" table. Unified Patents indicates 3 cited references, but I cannot name them. Everything in §2 is the IPR/litigation prior art, which is the operative prior art tested against the claims.
2. Exhibit-list vs. grounds distinction. DiFalco, Lingafelt, Apacible, Reger, and Civanlar appear on Arista's exhibit list, but the institution decision's grounds table reflects only Sheikh, Endicott, and Iwayama. I cannot confirm the former five were asserted as anticipation/obviousness grounds.
3. Sheikh full description. I did not retrieve the Sheikh publication's title/abstract; my characterization rests on the parties' and Board's framing. Confirm the title before quoting it.
4. No 2026 activity. Consistent with my earlier section, the most recent substantive outcome I found is the July 11, 2019 PTAB remand decision; the litigation/PTAB history is 2015–2019.

I need to ground this analysis in the actual prior-art record. Let me search for the references that were actually at issue in the IPR on this patent, since no "Prior Art" section was included in the material provided to me.

I now have the actual IPR record. Let me pull details on the primary reference (Sheikh) and the remand decision that is pivotal to the obviousness question.

Let me confirm the remand reasoning — why the broadcast claims survived even under the Federal Circuit's "multicast address" construction.

Preliminary note: the "Prior Art section of this page" was not provided

Strictly applied, the task instruction says to use the results from the Prior Art section of the page. No such section was included in the material supplied to me — the material I received contains only the full patent text (Google Patents), a prior-generation patent summary, and a prior-generation litigation summary. Rather than invent a prior-art list, I reconstructed the actual prior-art record from the live PTAB/CAFC filings in IPR2015-00978, which is the only adjudicated § 102/§ 103 record on this patent. Every reference below is grounded in those primary documents, with URLs. Where I could not verify a reference's disclosure, I say so.

1. Corrections to the earlier generated sections

Two items in the previously generated "Patent summary" conflict with the primary record and should be corrected:

Also note: the current task date ("April 26, 2026") differs from the session date (2026-09-29). Immaterial here, but flagged. The patent's adjusted expiration is 2026-01-26, i.e., the patent is expired — which matters for the § 103 posture (Phillips construction applies in any new proceeding, and only historic damages inside the 6-year § 286 window remain live).

2. The actual prior-art record (IPR2015-00978)

Instituited grounds (Institution Decision, Oct. 6, 2015) and outcomes (FWD, Sept. 28, 2016):

Sources: Institution Decision; FWD; Remand FWD.

Appellate overlay. Arista Networks v. Cisco Systems, 908 F.3d 792 (Fed. Cir. Nov. 9, 2018) construed "broadcast"/"broadcasting" as "transmission to one or more devices using a multicast address" and remanded; it also held assignor estoppel unavailable in IPR. On remand (July 11, 2019) the Board held Petitioner had not carried its burden on claims 29, 63, 64, 73, 86 even under the new construction.

3. Graham factors

(a) Level of ordinary skill (proposed). The Board's stated level is not in the excerpt I retrieved, so this is my reasoned proposal, not a quote: a POSITA at the 2003-09-19 priority date would hold a B.S. in EE/CS (or equivalent) plus ~3–5 years in network-device architecture and network security, or an M.S. with ~2 years. I flag this as an estimate.

(b) Scope/content of the prior art. Sheikh is squarely in the field: a security software system that "monitors, in great detail, any configuration changes made to information systems within a network… web servers, firewalls, proxy servers, log servers, intrusion detection software systems, routers and any other device or application" (Sheikh, p. 6). Its architecture — "agent transports" carrying "sensors" on monitored systems, reporting to a "master transport" on a central server, with a baseline "template" compared against monitored machines and an alert to the administrator on mismatch — discloses each element of independent apparatus claim 1 as the Board found. Iwayama discloses monitoring a status change in an external database and "transmit[ting] the notification … in a broadcasting manner" to a group of information terminals, and "distribut[ing] mail in a broadcasting manner to a predetermined group of email addresses."

(c) Differences. The only differences the Board credited were (i) the broadcast/multicast delivery limitation (claims 29, 63, 64, 73, 86) and (ii) whatever claim 15 adds (its text is truncated in the supplied copy at "the logging module is further configure…" — I cannot state what claim 15 adds). Everything else in the independent claims was held anticipated/obvious over Sheikh.

(d) Secondary considerations. I found no evidence of record on commercial success, long-felt need, copying, or industry praise tied to the claims, and no nexus evidence. Arista did not run a secondary-considerations case; the survival of the broadcast claims rested purely on the claim-construction/disclosure gap, not on objective indicia. Net effect: no meaningful secondary-considerations counterweight to the § 103 case.

4. § 103 combinations

Combination A — Sheikh alone (obviousness variant of Ground 2)

Even where Sheikh was used as § 102 art, the Board also accepted an alternative § 103 theory for claims 71, 72, 84, 85, crediting Dr. Mohapatra that "Sheikh also would have taught or suggested the recited processor, computer readable medium, and computer code limitations" and the "computer program product comprising first and second sets of instructions, and computer readable media" limitations to a POSITA. Rationale: Sheikh is a software system; recasting its monitoring functions as program instructions on computer-readable media is a simple substitution of a known element for a known element yielding a predictable result (KSR rationale (b)). This is the most solidly established conclusion in the record.

Combination B — Sheikh + Iwayama (Ground 4)

Motivation to combine, as articulated in the Petition and credited at institution:

• Both references are in the same field (network/system status monitoring and notification) and address the same problem — efficiently getting status/configuration-change information to the parties responsible for acting on it. In re Inland Steel, 265 F.3d 1354, 1362 (Fed. Cir. 2001).
• Sheikh's own stated goal supplied the reason: a POSITA would recognize that Sheikh's notification system, "which communicates to one or more master transports," would benefit from "sharing the information simultaneously" and "conduct[ing] communication based on an assumption that all the [devices] share the same information."
• Iwayama teaches the enabling technique — disseminating a detected status change by "broadcast[ing]" a notification to a predetermined group (chat channel participants; mailing-list addresses).
• KSR rationales: (a) combining known elements by known methods to yield predictable results; (c)/(d) applying a known distribution technique to a known monitoring device; (e) a finite number of predictable solutions (unicast polling vs. group broadcast) with a reasonable expectation of success.

Why it failed: after the Federal Circuit added "using a multicast address," the Board held Arista had not proven that Sheikh + Iwayama disclosed that element. Iwayama's group-email/chat-channel distribution is application-layer group communication; mapping it onto a network-layer multicast address is arguable but was rejected. Arista's remand brief argued precisely the opposite ("Iwayama's Group Email System Teaches Transmissions to One or More Devices Using a Multicast Address") and lost. Recommendation: treat Ground 4 as a closed, adverse precedent, not as a live § 103 theory.

Combination C — Sheikh + IP-multicast art (the stronger, unrun theory)

The reason claim 29/63/64/73/86 survived is that Arista tied the "multicast address" element to Iwayama, a chat/email reference. But the record itself contains a reference that supplies the element directly: IETF RFC 1112, "Host Extensions for IP Multicasting" (Aug. 1989), which was filed as Patent Owner's own Exhibit 2001 in the IPR. RFC 1112 defines IP host-group (multicast) addressing as of 1989 and was unquestionably a § 102(b) printed publication.

A POSITA seeking to distribute configuration-change alerts to one or more security monitors (the '597 specification's own stated purpose) would naturally select IP multicast rather than replicate per-monitor unicast alerts — it is the established mechanism for exactly that. Motivation is strong under KSR rationales (c)/(d) (known technique to improve similar devices) and (f) (design incentives/market forces), with a full reasonable expectation of success. This combination avoids the "application-layer broadcast is not multicast" defect that defeated Ground 4, because the reference supplies network-layer multicast addressing itself. Caveat: I am not representing that RFC 1112 was mapped to the claims by anyone; it was submitted for a claim-construction argument. The theory is my reconstruction, and its viability depends on mapping RFC 1112's host-group/multicast-address teachings to the logging-module broadcast at the appropriate level of ordinary skill. A multicast-capable US patent from the 2000–2002 window (e.g., a US 6,xxx,xxx in the IPR exhibit list such as Civanlar, US 5,995,606) would likely be a cleaner vehicle than an RFC, and I have not verified Civanlar's disclosure.

Combination D — the unadjudicated dependent claims

Ninety-four of the 110 claims were never challenged (claims 2–13, 16–28, 30–38, 43–62, 65–70, 74–83, 87–110). A reasoned § 103 exposure assessment:

5. Headline conclusion

The § 103 picture for US 7,340,597 collapses almost entirely to one limitation:

• Independent claims 1, 39, 71, 84 and dependents 14, 40–42, 72, 84–85(†) were held unpatentable over Sheikh (anticipation and/or obviousness). Those determinations were not disturbed on appeal and are final.
• Every surviving claim (29, 63, 64, 73, 86) adds only the "broadcasting… using a multicast address" limitation over a parent claim that was itself invalidated. The patent's entire post-IPR life therefore rests on a single, narrow, construction-driven element — and the Board twice held (2016 and 2019) that the art of record did not supply it.
• The best untested § 103 theory is not Sheikh + Iwayama (which lost twice) but Sheikh + network-layer IP-multicast art (RFC 1112 / IGMP / a multicast-alert patent), which meets the Federal Circuit's "multicast address" construction head-on. Whether that theory can be raised is, however, constrained:
  • 35 U.S.C. § 315(e) estops Arista (and its real parties in interest/privies) from re-asserting before the Office or in the district court/ITC any ground it raised or reasonably could have raised in IPR2015-00978 — which plausibly captures RFC 1112, since it was already in that record as an exhibit.
  • A different petitioner is not so barred. But the patent expired 2026-01-26, so the practical exposure is limited to historic damages within the § 286 six-year lookback, and any new proceeding would construe the claims under Phillips rather than the broadest reasonable interpretation used in IPR.
• Claims 3 and 8 (anti-tamper restrictions on the logging module) are the genuinely defensible subject matter, and they were never tested. Any validity analysis that stops at the broadcast claims overstates the strength of the § 103 case against this patent.

(†) Claim 85's cancellation follows from Ground 2; the FWD's invalidated list is 1, 14, 39–42, 71, 72, 84, 85.

Residual uncertainties I will not paper over: (i) I do not have verbatim text for independent claims 39, 71, or 84, or for claim 15 — my characterization of their scope is inferential from the Board's and the Federal Circuit's paraphrases; (ii) I could not verify what Endicott (US 5,404,525) discloses; (iii) the petition filing date appears as both April 1 and May 1, 2015 in the official documents; (iv) I found no secondary-considerations evidence in the record, but absence of evidence in the sources I retrieved is not proof it does not exist.