US 7031870

I'll search for authoritative information on this specific patent.

Let me check specifically for any 2026 Federal Circuit activity involving this patent or its owner.

Summary of US 7,031,870 B2 — "Data center evaluation using an air re-circulation index"

Search coverage and caveat up front. My searches covered Google Patents (which mirrors the USPTO full-text record), USPTO-derived third-party databases (uspto.report, Justia, RPX, Unified Patents, PatSnap), CourtListener/EDTX dockets, and CAFC/2026 Federal Circuit materials. On the CAFC question specifically: I found no Federal Circuit docket, appeal number, opinion, or Rule 36 judgment in 2026 that names US 7,031,870. What I did find in the 2026 Federal Circuit output is a different Valtrus case — Google LLC v. Valtrus Innovations Ltd., No. 2024-1945, argued March 4, 2026 — which concerns Valtrus's search/CPU-utilization patents (e.g., the '704, '604, '809 family), not the '870. So the honest statement is: no 2026 CAFC activity on '870 is reflected in the sources I could reach; I cannot rule out a docket entry that isn't indexed in open search results. Treat the litigation portion below as "what is verifiable," not as a complete docket history.

---

Bibliographic data (per Google Patents / USPTO record)

Source: https://patents.google.com/patent/US7031870/en and https://uspto.report/patent/grant/[7,031,870](/patent/7031870)

---

Abstract (verbatim)

"In a method for evaluating one or more components in a data center, inlet and outlet temperatures of one or more heat dissipating devices are detected. In addition, the temperatures of air supplied by one or more computer room air conditioning (CRAC) units are also detected. Indices of air re-circulation for the one or more heat dissipating devices are calculated based upon the detected inlet temperatures, outlet temperatures and supplied air temperatures. The indices of air re-circulation are determined at various flow field settings of air delivered to the one or more heat dissipating devices and the one or more components are evaluated based upon changes in the indices of air re-circulation for the one or more heat dissipating devices at the various flow field settings."

---

Core technical content — the SHI

The patent's "index of air re-circulation" is the Supply Heat Index (SHI), a measure of hot-air infiltration into cold air:

• Equation (1): SHI_i = δQ / (Q + δQ), where Q is the rack's heat dissipation and δQ is the enthalpy rise of the cooling air before it enters the rack.
• The specification reduces this to SHI_i = (T_in − T_ref) / (T_out − T_ref) (Equation 4/5), where T_ref is the CRAC/vent-tile supply temperature.
• A linear SHI-vs-rack-inlet-temperature plot (across racks/flow settings) indicates re-circulation heat load is negligible and the data center is "substantially optimized"; non-linearity signals adverse airflow. The spec suggests R² > 0.9 as a "nearly linear" threshold.

Priority document cross-reference: the SHI concept is also described in Ser. No. 10/446,854 (filed May 29, 2003), incorporated by reference.

---

Plain-language overview of the independent claims

Claim 1 — Method (the flagship claim). A method for evaluating components in a data center:

1. Detect inlet AND outlet temperatures of heat dissipating devices (e.g., racks);
2. Detect supply air temperatures of one or more CRAC units;
3. Calculate air-re-circulation indices (SHI) from those three temperatures;
4. Vary a flow field setting (e.g., CRAC blower volume flow rate);
5. Determine whether the indices changed in response to the varied setting; and
6. Evaluate the components based on those changes.

Claim 13 — CFD tool. A computational fluid dynamics tool with a modeling program that obtains the inlet/outlet and CRAC supply temperatures (either from user input or by computing airflow characteristics), calculates the re-circulation indices at various flow field settings, and evaluates the components based on changes in those indices. In other words, the same evaluation logic implemented as a simulation/design tool rather than live instrumentation.

Claim 22 — System (apparatus). Heat dissipating devices with inlet and outlet temperature sensors + CRAC units with supply-air temperature sensors + a controller programmed to compute the re-circulation indices at various (volume) flow field settings and evaluate the components based on changes in the indices.

Claim 32 — System (means-plus-function). The same functional combination expressed as means for detecting inlet/outlet temperatures, means for detecting CRAC supply temperatures, means for calculating indices at various flow field settings, and means for evaluating components based on the calculated indices.

Representative dependent claims (illustrative of scope):

• Claim 24: control logic flags the data center infrastructure as substantially optimized when a near-linear relationship exists between SHI and inlet temperatures (and not optimized otherwise).
• Claim 26: recites the SHI = (T_in − T_ref)/(T_out − T_ref) equation expressly.
• Claims 28–29: if SHI is unchanged across CRAC flow rates, output that the rack is not influenced by that CRAC; if it changes, output that it is influenced, and quantify the level of influence.
• Claims 30–31: determine the magnitude of SHI change and correlate it to a rack airflow characteristic.

Claim text sources: https://insight.rpxcorp.com/patent/[US7031870B2](/patent/US7031870B2) ; https://patents.justia.com/patent/20050267639

---

Litigation and enforcement context

The Google Patents record flags this patent as part of a family with litigation, listing nine E.D. Tex. cases tied to the '870 family (2024), and Darts-IP noting a "first worldwide family litigation" event.

Verifiable enforcement activity:

• Valtrus Innovations Ltd. v. NTT Data Services, LLC, No. 2:24-cv-00361 (E.D. Tex.) — the '870 is Exhibit 4 to the complaint (U.S. Patent 7,031,870), alongside exhibits for the '179 patent. https://www.courtlistener.com/docket/68526786/1/4/valtrus-innovations-ltd-v-ntt-data-services-llc/
• Valtrus v. Digital Realty Trust and Valtrus v. CyrusOne — PatSnap reports the '870 as one of seven asserted patents (US 7,031,870; 6,862,179; 7,939,967; 6,854,287; 6,718,277; 7,339,490; 9,310,855) targeting data-center cooling/thermal-management systems at DFW11 and DFW26 facilities; the Digital Realty action is described as dismissed. https://www.patsnap.com/de/resources/blog/litigation/valtrus-v-digital-realty-trust-data-center-cooling-ip-dismissed-patsnap/
• In re: Valtrus Innovations Ltd. Patent Litigation, MDL No. 3190 — J.P.M.L. conditional transfer order (CTO-1, effective after Aug. 7, 2026), consolidating cases before Judge Gilstrap in E.D. Tex. https://www.govinfo.gov/content/pkg/USCOURTS-txed-2_26-cv-00762/pdf/USCOURTS-txed-2_26-cv-00762-0.pdf
• Parallel Valtrus campaigns continue against FedEx, NetApp, Starbucks, Google, and others (2025–2026 IPR filings reference these), but those petitions and the 2024-1945 appeal concern other Valtrus patents ('416, '832, '704, '604, '809) — not the '870. https://www.cafc.uscourts.gov/03-04-2026-2024-1945-google-llc-v-valtrus-innovations-ltd-audio-uploaded/

Bottom line on the CAFC question: No 2026 Federal Circuit proceeding naming US 7,031,870 is documented in the sources I can access. The '870's activity is at the district-court level (E.D. Tex., now under MDL 3190). I explicitly flag this as a negative finding limited to available sources, not a certified absence of any filing.

---

Confidence and uncertainty notes

• High confidence: title, application number (10/855,400), inventors, filing date (2004-05-28), issue date (2006-04-18), original assignee, abstract, formula, and the independent-claim content (corroborated across Google Patents, RPX, Justia, and unifiedpatents).
• Moderate confidence: the exact count of independent claims (41 total; I read four independents — 1, 13, 22, 32 — from third-party claim listings; I did not retrieve the official USPTO claim set image-by-image).
• Unresolved discrepancy: priority date listed as 2004-05-27 (Unified Patents) vs. 2004-05-28 (Google Patents/USPTO filing date).
• I did not verify the current legal-status field against USPTO PatentCenter directly; "Expired – Lifetime / adjusted expiration 2024-10-01" is taken from Google Patents' status metadata.
• No 2026 CAFC docket for '870 found, as discussed above.

I'll research the litigation history for US patent 7031870. Let me start with searches on the litigation databases.

Let me dig deeper into the specific cases and defendants listed.

I have substantial data. Let me identify the remaining cases listed on the patent record and confirm the '870 patent's role.

Litigation involving US 7,031,870

Important framing: US 7,031,870 ("Data center evaluation using an air re-circulation index," inventors Bash/Patel/Sharma; original assignee Hewlett-Packard Development Co., L.P.) is now owned by Valtrus Innovations Ltd. (acquired from HPE via OT Patent Escrow, LLC). The '870 patent was asserted as one of a seven-patent portfolio in a 2024 assertion campaign by Valtrus Innovations Ltd. and Key Patent Innovations Ltd. against data-center operators and in a supplier declaratory-judgment action. The seven patents asserted together were:

• US 7,031,870 (the patent at issue here)
• US 6,862,179, US 6,854,287, US 6,712,277, US 7,339,490, US 9,310,855, US 7,939,967

The '870 patent appears in the patent record's litigation list as being involved in ten Texas Eastern District Court cases (all filed in 2024). Below are the ones I could verify and their status.

Verified cases in which the '870 patent was at issue

| # | Plaintiffs | Defendant(s) | Jurisdiction / Case No. | Filed | Status / Outcome |
|---|---|---|---|---|
| 1 | Valtrus Innovations Ltd. (later with Key Patent Innovations Ltd.) | NTT Data Services, LLC, et al. (incl. NTT Global Data Centers Americas, Inc.) | E.D. Tex., Marshall Div. — 2:24-cv-00361-JRG (LEAD CASE, Judge Rodney Gilstrap) | ~May 14, 2024 | Open. Lead consolidated case. '870 listed among the "Asserted Patents" in the P.R. 4-3 Joint Claim Construction and Prehearing Statement (Aug. 11, 2025). Markman hearing held Oct. 14, 2025; docket control order sets pretrial conference Mar. 2, 2026 and jury selection Apr. 6, 2026. Vertiv intervened. |
| 2 | Valtrus Innovations Ltd. | Digital Realty Trust, Inc. and Digital Realty Trust, L.P. | E.D. Tex. — 2:24-cv-00139-JRG | 2024 (Feb. 27, 2024 per one docket source; a later exhibit lists June 7, 2024 — see caveat) | Dismissed with prejudice (joint motion) Jan. 22, 2025. |
| 3 | Valtrus Innovations Ltd. and Key Patent Innovations Ltd. | Digital Realty Trust, Inc. and Digital Realty Trust, L.P. | E.D. Tex. — 2:24-cv-00535-JRG | ~July 12, 2024 | Dismissed with prejudice (joint motion) Jan. 22/23, 2025. |
| 4 | Valtrus Innovations Ltd. | CyrusOne, LLC | E.D. Tex. — 2:24-cv-00259-JRG | 2024 | Dismissed with prejudice (joint motion) Nov. 22, 2024. |
| 5 | Valtrus Innovations Ltd. and Key Patent Innovations Ltd. | CyrusOne, LLC | E.D. Tex. — 2:24-cv-00534-JRG | ~July 12, 2024 | Dismissed with prejudice (joint motion) Nov. 22, 2024. |
| 6 | Valtrus Innovations Ltd. | Dawn Acquisitions LLC (d/b/a Evoque Data Center Solutions) | E.D. Tex. — 2:24-cv-00142 | 2024 | A dismissal order relating to this case is referenced in a PTAB exhibit list (EX1025), indicating it was resolved. I could not confirm the exact disposition date. |
| 7 | Vertiv Corporation (declaratory judgment plaintiff) | Valtrus Innovations Ltd. | E.D. Tex. — 2:24-cv-00907-JRG (Member Case, consolidated into Lead Case 2:24-cv-00361) | 2024 | Open/consolidated. Vertiv seeks declarations of non-infringement/invalidity; '870 is among the patents addressed in the joint claim construction statement. |

Cases listed on the '870 litigation record that I could NOT verify

The patent record (Google Patents / Unified Patents litigation data) also lists these E.D. Tex. cases for the '870 patent, but my searches did not return their parties or details:

• 2:24-cv-00776
• 2:24-cv-00777
• 3:24-cv-01152 (Tyler Division)

These are presumably further 2024 Valtrus data-center-operator suits in the same campaign, but I cannot confirm the defendants, dates, or status. I would check the PACER dockets directly.

Related multi-district litigation (note)

The later wave of Valtrus cases was centralized as In re: Valtrus Innovations Ltd. Patent Litigation, MDL No. 3190 (J.P.M.L.), transferred to E.D. Tex. before Judge Gilstrap. Per the JPML transfer order, that MDL concerns the '287, '682, and '277 patents — not the '870 patent — so the '870 patent is not part of the MDL's asserted set, even though it was asserted in the earlier 2024 cases and the Vertiv declaratory-judgment action.

Key caveats

• Patent term: The '870 patent's legal status is listed as "Expired – Lifetime," with an adjusted expiration of October 1, 2024. Any recovery would therefore be for past damages only.
• Filing-date discrepancies: Sources conflict on some filing dates (e.g., 2:24-cv-00139 is shown as filed 02/27/2024 by one docket aggregator but referenced as June 7, 2024 in a PTAB exhibit list). Treat these dates as approximate pending direct PACER confirmation.
• No appellate history found: I found no Court of Appeals for the Federal Circuit appeal or PTAB (IPR/PGR) challenge specific to the '870 patent. (The IPRs I located, e.g., Google v. Valtrus, IPR2025-01157, concern the '967 patent, not '870.)
• Dismissal character: The CyrusOne and Digital Realty dismissals were by joint motion with prejudice and are consistent with settlement; they are not merits rulings on the '870 patent.

Sources

• US 7,031,870 patent page and litigation record: https://patents.google.com/patent/US7031870/en
• Lead case P.R. 4-3 Joint Claim Construction Statement (lists '870 among asserted patents): https://storage.courtlistener.com/recap/gov.uscourts.txed.[230090](/patent/230090)/gov.uscourts.txed.230090.117.0.pdf
• Dismissal order (CyrusOne, 2:24-cv-00259/00534): https://www.courtlistener.com/docket/68526786/83/valtrus-innovations-ltd-v-ntt-data-services-llc/
• Dismissal order (Digital Realty, 2:24-cv-00139/00535): https://www.courtlistener.com/docket/68526786/96/valtrus-innovations-ltd-v-ntt-data-services-llc/
• JPML MDL 3190 transfer order: https://www.jpml.uscourts.gov/sites/jpml/files/MDL-3190-Transfer_Order-7-26.pdf
• CyrusOne answer identifying '870 as the "Fourth Claim": https://www.courtlistener.com/docket/68526786/66/valtrus-innovations-ltd-v-ntt-data-services-llc/
• Stanford NPE litigation database entries: https://npe.law.stanford.edu/party/valtrus-innovations-ltd-0

Bottom line: There is known litigation involving US 7,031,870 — primarily the 2024 Valtrus Innovations v. data-center-operator suits (NTT Data, Digital Realty, CyrusOne, Dawn/Evoque) and the Vertiv declaratory-judgment action, all in the Eastern District of Texas before Judge Gilstrap. The Digital Realty and CyrusOne cases were dismissed with prejudice by joint motion; the NTT Data lead case remained open as of the latest records, with trial calendared for April 2026 and the '870 patent being part of the asserted set. Three additional case numbers on the patent's litigation record (2:24-cv-00776, 2:24-cv-00777, 3:24-cv-01152) remain unconfirmed.

I'll start by verifying the structured block's default against live sources, since the ODP ingest may lag recently-filed petitions.

The search surfaced Valtrus-related IPRs, but I need to confirm which patents they actually target — the exhibits reference US 7,057,509, a different patent. Let me pin down the patent numbers.

The Valtrus IPRs I've found all target different patents ('244, '575, '139, '409, '264, '509). Let me run targeted checks specifically for '870 and for a Unified Patents filing.

Proceedings overview

Zero AIA trial proceedings on file against US 7,031,870. The structured USPTO ODP block returns no IPR, PGR, or CBM for this patent, and my targeted web searches did not surface any petition, institution decision, FWD, or appeal naming the '870 patent — so the count is 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, and 0 institution denied. The defensive posture this produces for a defendant is neither "the patent is hardened" nor "the claims are dead": the claims are entirely untested at the PTAB, and the patent has already expired (adjusted expiration 2024-10-01, per the Google Patents record), so there is nothing left to prospectively enjoin. The practical consequence of the zero-count is that no § 315(e)(2) estoppel attaches to anyone, and all prior-art invalidity grounds remain fully available in the district court case.

No proceeding to report — and why that matters

Because the canonical list is empty, there is no proceeding number, petitioner, panel, petition ground, institution decision, FWD, settlement, or appeal to itemize. I am explicitly not populating the per-proceeding template with a zero-record placeholder, because inventing a proceeding number or a disposition would be fabrication. What I can report is the verification trail and the near-miss candidates.

Flagged but NOT on US 7,031,870 (verify before relying)

Valtrus Innovations Ltd. / Key Patent Innovations Ltd. is an active PTAB target, but every proceeding I could tie to a patent number involved a different patent in the portfolio. Do not cite any of these as '870 proceedings:

• IPR2025-00414, -00415, -00416, -00417, -00418, -00420 — SAP America, Inc. and SAP SE v. Valtrus Innovations Ltd.
  • Type: Inter Partes Review (six petitions)
  • Filed: 2025-01-15
  • Status: Institution denied (discretionary denial, 35 U.S.C. § 314(a)) — 2025-07-10
  • Patents challenged: US 7,152,182 ('414); US 6,691,139 ('415); US 7,313,575 ('416); US 6,823,409 ('417); US 6,871,264 ('418); US 6,889,244 ('420). None is '870.
  • Decision-maker: Acting Under Secretary and Acting Director Coke Morgan Stewart, who granted Patent Owner's discretionary-denial requests and denied institution in each case. Reasoning: the projected FWD due dates (August/September 2026) post-dated the parallel E.D. Tex. trial set for 2025-09-22, and the challenged patents had been in force over 17 years (≈20 years for '415/'417/'418), creating "strong settled expectations."
  • Decision: https://www.docketalarm.com/cases/PTAB/IPR2025-00418/SAP_America_Inc._v._Valtrus_Innovations_Ltd/07-10-2025-Board/Director_Discretionary_Decision_Deny-10-Director_Discretionary_Decision_Deny/ ; mirrored at https://bannerwitcoff.com/wp-content/uploads/2025/08/IPR2025-00417.pdf
• IPR2026-00262 — Google LLC v. Valtrus Innovations Ltd.
  • Filed: 2026-02-17; Status: Pending; Panel: not yet assigned. Board papers show a discretionary-denial fight, with Patent Owner exhibits referencing US 7,057,509 and Valtrus's search-engine patents — not '870. Docket: https://sandbox.docketalarm.com/cases/PTAB/IPR2026-00262/Google_LLC_v._Valtrus_Innovation_Limited/
• IPR2022-01408 — Google LLC v. Valtrus Innovation Ltd. (search-engine patents, e.g., '704/'764/'604 family) — also not '870.

Verification note / confidence level: the ODP block is authoritative for the proceedings it indexes, and my independent searches returned no '870 petition. I reached my per-search budget before I could run a final confirmation pass against PTAB E2E and the PTAB Decisions database directly, so I flag a residual risk that a very recently filed, not-yet-indexed petition exists. Confirm at PTAB E2E (https://ptacts.uspto.gov) and the PTAB API (https://developer.uspto.gov/ptab-api) before giving a client a clean "no IPR" opinion.

Strategic summary

Claim status: untested, not canceled, and now expired. Nothing in the record cancels, narrows, or confirms any claim of '870. The only claim text I can verify from the patent itself is independent claim 1, the method for evaluating one or more components in a data center (detect inlet temperatures, outlet temperatures, and CRAC supply temperatures; calculate an air re-circulation index, described as a Supply Heat Index, at various flow field settings; evaluate based on changes in the index). No FWD has ever construed it, and no certificate has ever canceled it. The corresponding file-history risk is real but latent: the printed patent record shows status "Expired – Lifetime," adjusted expiration 2024-10-01, which means past damages only (subject to the six-year lookback, i.e., roughly 2018-10-01 forward), no injunction, and — critically — no IPR-driven claim cancellation to anchor a validity defense. Anyone asserting '870 today is litigating an expired, PTAB-untested patent on backward-looking damages.

Estoppel landscape: essentially none, which cuts both ways. Because no IPR was ever instituted, no petitioner is subject to § 315(e)(2) estoppel, and no § 325(e) estoppel exists either. For a defendant now being asserted against, that means the entire prior-art universe is available in district court under §§ 102/103 and § 112 — with one important exception: the § 315(b) one-year bar has likely already closed for the 2024 wave of defendants. Valtrus's campaign was filed in E.D. Tex. beginning 2024-05-14 (lead case Valtrus Innovations Ltd. v. NTT Data Services, LLC, 2:24-cv-00361-JRG, where US 7,031,870 appears as Exhibit 4: https://www.courtlistener.com/docket/68526786/1/4/valtrus-innovations-ltd-v-ntt-data-services-llc/). Any defendant served on or before 2024-09-29 is now time-barred from filing an IPR on '870. Newly served defendants retain a one-year window from service. Practically, though, the expiry date makes an IPR a poor investment — the Board has discretionarily denied petitions on far stronger facts (see the SAP denials above) where the patent was old, the trial date preceded the FWD, and the only exposure was monetary.

Pattern signals. Valtrus has been a serial assertor in E.D. Tex. (Judge Rodney Gilstrap), running a coordinated campaign with parent Key Patent Innovations against data-center and hyperscaler defendants, with Vertiv Corporation intervening in multiple member cases. Defendants have answered with clustered IPR filings (SAP's six-petition January 2025 volley; Google's 2026 petition), and the Patent Owner has litigated those petitions hard, winning discretionary denials across the board on settled-expectations and trial-timing grounds. Notably, none of those petitioners chose '870 — plausibly because the patent expired in 2024 and the sibling patents carried more remaining life. Several 2024 member cases (Digital Realty 2:24-cv-00139, CyrusOne 2:24-cv-00259, and others) were dismissed with prejudice by joint motion in early 2025, on terms that are facially consensual with no merits finding on infringement or validity; the lead case 2:24-cv-00361-JRG was expressly maintained as open.

Recommended next steps

• If you are a defendant facing an '870 assertion: you have no FWD to point to — you cannot say "claim 1 is cancelled." Your leverage is elsewhere: (a) the expiration of the patent on 2024-10-01, which limits recovery to pre-expiration damages and eliminates injunctive relief; (b) the complete absence of § 315(e)(2) estoppel, leaving the full prior-art field open for §§ 102/103 and § 112 defenses in the district court; and (c) the six-year damages lookback, which is the real battleground. If you were served after 2025-09-29, you still have a live IPR window, but weigh it against the near-certainty of a § 314(a) discretionary denial given the SAP outcome and the patent's age.
• If you were served in 2024: assume the § 315(b) bar has run. An IPR on '870 is almost certainly unavailable to you personally, and a joinder strategy under § 315(c) requires an existing instituted IPR to join — of which there is none on this patent.
• Do not confuse the Valtrus IPRs with '870 IPRs. If opposing counsel or a demand letter implies the '870 claims have been "upheld" or "invalidated" at the PTAB, that is unsupported — the record is silent. Cite the empty ODP result and the expired status instead.
• Confirm the negative before relying on it. Re-run PTAB E2E (https://ptacts.uspto.gov) and the PTAB Open Data API (https://developer.uspto.gov/ptab-api) for patent number 7031870, since my search pass completed without a final direct-database confirmation. The absence of PTAB activity on a patent this widely asserted in 2024 is itself a signal worth double-checking rather than assuming.

I'll reconstruct the ownership record using the patent's own legal-events data plus targeted searches for the recorded-assignment details (reel/frame, correspondent) and the associated litigation.

I have the legal-events chain and the litigation context. Let me pin down the recorded-assignment reel/frame numbers, the correspondent of record, and confirm which suits name this patent.

Below is the reconstructed ownership record for US 7,031,870 B2 ("Data center evaluation using an air re-circulation index"), application 10/855,400, filed 2004‑05‑28, granted 2006‑04‑18, now Expired – Lifetime (adjusted expiration 2024‑10‑01).

Data-source caveat up front: I was able to pull the full conveyance chain from Google Patents legal events and from litigation exhibits, but the Assignment Center UI itself (https://assignmentcenter.uspto.gov/) did not return direct record pages in my searches, so reel/frame numbers and the correspondent-of-record below are partially reconstructed — where a reel/frame is stated I flag whether it is confirmed for this patent or inferred from a sibling patent recorded under the same instrument. Treat the marked-inferred items as leads to verify, not as confirmed record.

---

Inventors

All three executed an assignment to Hewlett‑Packard Development Company, L.P. on the filing date, 2004‑05‑28 (Google Patents legal event "Assigned to HEWLETT‑PACKARD DEVELOPMENT COMPANY, L.P.; Assignors: BASH, CULLEN E., PATEL, CHANDRAKANT D., SHARMA, RATNESH K."). This is a standard employee‑invention assignment, and the trio are HP Labs' data‑center thermal‑management group.

Unusual-pattern check: No evidence of any inventor departing HP within 12 months of filing, and no inventor appears anywhere later in the chain. This is not a founder-departure story — it is an ordinary corporate invention conveyed immediately to the employer. Nothing here precedes a fire‑sale.

---

Original assignee

Hewlett‑Packard Development Company, L.P. (HP), then Hewlett Packard Enterprise Development LP after the 2015 spinoff.

• Line of business: HP / HPE — enterprise servers, storage, networking, and data‑center infrastructure. This patent (supply heat index / air re‑circulation index for data centers) sits squarely in HPE's core data‑center cooling technology, developed by HP Labs.
• Did they ship a product embodying the claims? Not a standalone product — the SHI metric was an evaluation/design methodology for data‑center airflow, used internally by HP/HPE and reflected in its data‑center design services. There is no evidence of a commercial product sold under the claims. (The patent's own disclosure teaches using the index in CFD modeling and CRAC‑control algorithms.)
• Current status: Operating. HPE is a large, listed public company. The 2021 transfer was a portfolio divestiture, not a bankruptcy. The patent family is now held downstream by Valtrus Innovations Limited.

---

Assignment timeline

Chronological, post-filing:

• 2004‑05‑28 (executed) / recorded 2004‑05‑28 — Reel not retrieved (inventor‑to‑HP assignment; original-assignee record)

  • Conveyance: Assignment
  • Assignor: Cullen E. Bash; Chandrakant D. Patel; Ratnesh K. Sharma
  • Assignee: Hewlett‑Packard Development Company, L.P.
  • Correspondent: not retrieved (typical HP internal recording; HP's patent recordings route through hpe.jp.mail@hpe.com, 3404 E. Harmony Road, MS 79, Fort Collins, CO 80528 — inferred from sibling HPE records, not confirmed for this patent).
  • Context: original corporate assignment of employee inventions (internal).

• 2015‑11‑09 (effective) / recorded 2015‑11‑09 — Reel not confirmed for this patent; the HP Development Co LP → HP Enterprise Development LP spinoff instrument is recorded at Reel 037079/0001 for sibling HP patents.

  • Conveyance: Assignment of Assignor's Interest
  • Assignor: Hewlett‑Packard Development Company, L.P.
  • Assignee: Hewlett Packard Enterprise Development LP
  • Correspondent: not retrieved.
  • Context: internal reorg — the HP→HPE corporate separation; a mass transfer of the HP patent estate to the enterprise spin‑off.

• 2021‑01‑15 (agreement effective date; HPE signature 2021‑01‑20, escrow signature 2021‑01‑21) / recorded 2021‑01‑26 — Reel 055269/0001 (⚠️ confirmed from a sibling HPE patent recorded under the identical instrument; frame for 7,031,870 not individually verified)

  • Conveyance: Patent Assignment, Security Interest, and Lien Agreement (a two‑step escrow assignment, not a clean sale)
  • Assignor: Hewlett Packard Enterprise Company and Hewlett Packard Enterprise Development LP (both)
  • Assignee: OT Patent Escrow, LLC, a Delaware LLC, principal place of business 200 West Madison, 37th Floor, Chicago, IL 60606
  • Correspondent: not retrieved from Assignment Center. The escrow‑agent instrument was signed by Joel Lutzker (for OT Patent Escrow, LLC); HPE signed via Brett Alten (SVP, DGC, Chief IP Counsel). Flag: single appearance in this chain — not enough to establish recurrence.
  • Context: transfer‑to‑asserter, structured as a securitization/escrow — the patent was conveyed to an escrow LLC to be held for later assignment to Valtrus Innovations Limited, with HPE taking back a security interest and lien. Exhibit A‑2 of the same agreement lists the escrowed patents in tranches (the "HPE‑OT_Valtrus" tranche).

• 2022‑02‑09 (executed) / recorded 2022‑02‑09 — Reel not retrieved

  • Conveyance: Assignment of Assignor's Interest
  • Assignor: OT Patent Escrow, LLC
  • Assignee: Valtrus Innovations Limited, Irish company no. 669289, registered office The Glasshouses GH2, 92 Georges Street Lower, Dun Laoghaire, Dublin A96 VR66, Ireland
  • Correspondent: not retrieved. Note: Valtrus's licensing outreach for this portfolio is handled by Fabricant LLP, working with Valtrus and "Patent Platform Services"; Valtrus's E.D. Tex. litigation counsel of record is Findlay Craft, P.C. (Eric Hugh Findlay). These are known counsel, not confirmed recording correspondents.
  • Context: transfer from escrow vehicle to the operating NPE/licensing entity — completes the two‑step fire‑sale structure begun in Jan 2021.

• 2024‑10‑01 — Adjusted expiration recorded; patent status = Expired – Lifetime. (No asset ever returned up‑chain; Valtrus held to term.)

No bankruptcy, merger, change‑of‑name, or release conveyance is recorded in this chain.

---

Timeline diagram

timeline
    title Ownership of US 7031870
    2004 : Filed by Hewlett Packard Dev Co
         : Inventors assign to HP
    2006 : Issued as US7031870B2
    2015 : Moved to HP Enterprise Dev LP
    2021 : Moved to OT Patent Escrow LLC
    2022 : Moved to Valtrus Innovations Ltd
    2024 : Patent term expires
         : Valtrus data center suits

---

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT.
Patent moved from operating assignee HPE to a two‑tier holding structure: OT Patent Escrow, LLC, a Delaware LLC at a Chicago office‑tower address (200 W. Madison, 37th Fl.), with no products in commerce and existing solely to hold patents in escrow (Reel 055269/0001, 2021‑01‑15/26), which then passed the asset to Valtrus Innovations Limited, an Irish entity with a registered office but no US operations and no products (recorded 2022‑02‑09). Valtrus's own demand letters describe it as "focused on licensing these patents," and its corporate parent is New PP Licensing LLC per its Federal Rule 7.1 disclosures. This is a textbook licensing‑only vehicle.

2. Known asserter in the chain — PRESENT.
Valtrus Innovations Ltd. is a high‑frequency patent plaintiff tracked by RPX and Unified Patents. Chain evidence: Valtrus's first suit was Valtrus Innovations Ltd. v. Google LLC (E.D. Tex., 3:22‑cv‑00066 / IPR2022‑01399, Jan 2022); it has since sued Lumen, Digital Realty, CyrusOne, NTT Data, Cologix, CoreSite, Iron Mountain, Netrality, H5, Prime Data Centers, EvoDC, NetApp, FifteenFortySeven, Continental Broadband, and others, with Key Patent Innovations Ltd. (KPI) as co‑plaintiff. In June 2026 the JPML created MDL No. 3190 consolidating the Valtrus data‑center cases in E.D. Tex. (see CTO‑2 transfers, e.g. 2:26‑cv‑00820, 2:26‑cv‑00793). KPI is also the entity behind the BlackBerry portfolio monetization per RPX ("Bulk of BlackBerry's Patent Portfolio Finds a New Home," March 2023) and is intertwined with monetization firm Patent Platform Services (Frisco, TX — co‑founder Kapu Kumar). This is a well‑documented assertion operation, not a name‑only suspicion.

3. Repeat correspondent across the chain — UNCLEAR.
The only named individual appearing on the transfer instruments is Joel Lutzker, signatory for OT Patent Escrow, LLC on the 2021 escrow agreement (Reel 055269/0001 tranche). He appears once in this chain; I have no Assignment‑Center correspondent data showing him on the 2022 OT‑Escrow→Valtrus recording or on other link(s). A single appearance is not the signal — recurrence is — so I mark this unclear. Verify the recording correspondent for the 2022‑02‑09 link; if Lutzker (or a single firm) appears on both the 2021 and 2022 recordings, this flips to present.

4. Cascading transfers — PRESENT (structural).
A deliberate two‑step chain within ~13 months: HPE → OT Patent Escrow, LLC (2021‑01‑15) → Valtrus Innovations Limited (2022‑02‑09). The escrow step exists only to bridge HPE to Valtrus, and both steps are documented as one integrated "Patent Escrow Agreement" transaction (the assignment itself recites that the patents "are to be assigned to Assignee to hold in escrow for later assignment to Purchaser" = Valtrus). Add the 2015 internal reorg and this is a three‑hop chain, the last two hops being vendor→escrow→NPE with an HPE security interest retained.

5. Pre‑litigation transfer — NOT PRESENT (for this patent).
The transfer to Valtrus was executed/recorded 2022‑02‑09, whereas every suit naming US 7,031,870 was filed in 2024 (E.D. Tex. 2:24‑cv‑00139, ‑00534, ‑00535, ‑00776, ‑00777, ‑00907, 3:24‑cv‑01152, etc.). That is ~2 years, outside the 6‑month window. (The Jan 2022 Google suit predates the Feb 2022 recording but asserted other HPE/Verity patents, not the '870.) So no arrangement‑for‑venue timing signal here — mark not present.

6. Bankruptcy fire‑sale — NOT PRESENT.
HPE is a solvent, operating public company; the 2021 transfer was a negotiated portfolio divestiture (~1,500 patents, ~half US), not a Chapter 7/11 sale. No bankruptcy docket or SEC 8‑K distress disclosure is implicated. The structure resembles a distressed divestiture, but the bankruptcy tell is absent.

7. Privateering — UNCLEAR.
The structure has privateering hallmarks: an operating company (HPE) sheds ~1,500 patents to a licensing entity (Valtrus) that now asserts against HPE's own competitors and against data‑center operators/suppliers in HPE's market, and HPE retained a security interest and lien over the patents in the escrow instrument (Assignee OT Patent Escrow granted Assignor HPE the security interest) — an atypical retention for a clean sale. That said, I found no SEC filing, Patent Progress, or EFF coverage showing HPE directs or funds Valtrus's suits. Retention of a security interest ≠ control. Mark unclear pending direct evidence of HPE direction.

8. Defensive aggregator (anti‑NPE) — NOT PRESENT.
The chain terminates at Valtrus Innovations Limited, an active asserter, not at RPX / AST / LOT / Unified / OIN. The inverse signal applies: the patent was weaponized, not neutralized. (Unified Patents did file IPRs against related Valtrus patents, but that is Unified acting as a challenger, not as the chain's terminal owner.)

---

Verdict

NPE — high confidence.

Three strong signals land: (1) a two‑tier licensing‑only holding structure — HPE → OT Patent Escrow, LLC (Delaware, Chicago registered‑office address) → Valtrus Innovations Limited (Ireland, no US products) — recorded at Reel 055269/0001 (2021‑01‑15/26) and on 2022‑02‑09; (2) Valtrus / Key Patent Innovations is a documented high‑frequency plaintiff now consolidated into MDL No. 3190 (JPML, June 2026); and (3) the deliberate escrow‑bridge cascade (2021 → 2022) created solely to move the asset from the operating company to the asserting vehicle. This patent was asserted against data‑center operators in E.D. Tex. 2:24‑cv‑00139 and related 2024 filings, and expired 2024‑10‑01 — a classic "run‑to‑expiry" assertion of a divested operating‑company patent. Correspondent recurrence (Signal 3) is the one open thread; verify the recording correspondent on the 2022‑02‑09 link.

Verify at:

• USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search patent 7031870)
• Legacy/indexed mirror — https://assignment.uspto.gov/patent/index.html
• Google Patents legal events — https://patents.google.com/patent/US7031870/en

Reel/frame items flagged "inferred" (2015 Reel 037079/0001 series; 2021 Reel 055269/0001 series) were corroborated from sibling HPE patents recorded under the identical instruments and should be confirmed against 7,031,870's own Assignment Abstract of Title before external citation.

I'll research the prior art citations for US 7,031,870 specifically. Let me run targeted searches.

Let me pull the actual "References Cited" front page and the PatentCenter record.

Let me fetch the patent's front-page references directly from the PDF and the Justia/PatentCenter records.

Important limitation up front

I was unable to retrieve the complete "(56) References Cited" front page for US 7,031,870 within my search budget. My searches reached Google Patents, uspto.report, Justia, RPX/PatSnap, and IPR/office-action PDFs, but the raw examiner-cited list (the "U.S. PATENT DOCUMENTS / FOREIGN PATENT DOCUMENTS / OTHER PUBLICATIONS" block that appears on the printed front page) did not surface in the indexed results. I will not fabricate patent numbers or dates to fill that gap. Below I give (a) the references I can verify directly from the '870 record, (b) the closest prior art named in the patent's own text and field, and (c) the forward citations I confirmed — with explicit confidence labels for each.

Also flagging a metadata conflict for the record: the task header says "April 26, 2026" while my operating header says 2026-09-29. I note it rather than resolve it, consistent with the "don't auto-correct" rule.

---

A. References cited inside the '870 specification (incorporated by reference) — VERIFIED

These are not "examiner citations," but they are the prior-art documents the '870 expressly relies on. They appear in the granted text you supplied, so confidence is high.

---

B. Closest prior art from the inventors' own published work (named in the field, not necessarily on the '870 front page)

• Sharma, R., Bash, C., Patel, C., "Dimensionless parameters for evaluation of thermal design and performance of large-scale data centers," 8th AIAA/ASME Joint Thermophysics and Heat Transfer Conference, 2002 (paper 3091). This is the seminal publication that introduced the dimensionless SHI parameter. As a printed publication dated 2002 (>1 year before the May 2004 filing), it is §102(b) prior art. It could anticipate claims drawn to calculating an air-re-circulation/SHI index from inlet, outlet, and supply temperatures — e.g., the index-calculation portions of claims 1, 22, 26, 32 — but it does not appear to disclose the claimed "determine indices at various flow field settings and evaluate based on changes" step, which is the '870's point of novelty over this paper.
  • Source: Google Scholar profile for R.K. Sharma (citation list showing the 2002 AIAA paper, 354 cites).
• Bash, C.E., Patel, C.D., Sharma, R.K., "Efficient Thermal Management of Data Centers — Immediate and Long-Term Research Needs," HVAC&R Research 9(2):137–152, 2003. Survey/printed publication; background art, weak §102 relevance.

---

C. Forward citations — patents/publications citing the '870 (VERIFIED from Google Patents "Cited By")

Sources: https://patents.google.com/patent/US20100029193A1/en ; https://patents.google.com/patent/US8712735 ; R.K. Sharma Google Scholar patent list.

---

D. §102 anticipation assessment — the honest bottom line

1. The most dangerous §102 reference is Ser. No. 10/446,854, because it discloses SHI. However, it is commonly owned by HP and shares the same inventors (Sharma, Bash, Patel). Under 35 U.S.C. §103(c) (pre-AIA), commonly owned/same-inventor subject matter cannot be used in an obviousness combination, and as a §102 reference it is largely self-anticipating only of the SHI-computation limitations — not of the "vary flow field setting → observe change in indices → evaluate" limitation that distinguishes independent claim 1.

2. The claim 1 novelty anchor is the perturbation/evaluation step, not the SHI formula. The formula is expressly attributed in the patent to the earlier '854 application and to the 2002 AIAA paper. So any anticipation attack on claim 1 has to find a single reference teaching changing a flow field setting and evaluating the component based on the resulting change in SHI — none of the references I could verify does that.

3. Claim 13 (CFD tool) is the most exposed, because it recites the same logic implemented in simulation; a printed publication (2002 AIAA paper) plus a CFD-modeling disclosure could come closest, but I did not verify a single anticipatory CFD reference.

4. I could not verify any examiner-cited U.S. patent documents on the '870 front page, so I am not asserting that the references above are the complete §102 set the examiner relied on. That is a gap, stated plainly.

---

E. What I would need to close the gap

To give you the exact examiner citation list with claim-by-claim §102 mapping, the authoritative sources are:

1. USPTO PatentCenter for application 10/855,400 — the "References Cited" / "PTO-892" forms; and
2. The granted front page (PDF) at patentimages.storage.googleapis.com for US 7,031,870.

I was not able to open those specific documents before hitting my step limit. If you can paste the front-page "(56) References Cited" block or the PatentCenter PTO-892, I will produce the full per-reference table (citation, date, description, §102 claim mapping) without speculation.

Confidence: High on section A (taken directly from the granted text), High on section C (Google Patents "Cited By"), Moderate on section B (publication dates confirmed but full-text §102 mapping inferred), and explicitly incomplete on the examiner-cited prior-art list itself.

I'll gather the prior-art references cited on this patent's face and in its specification before running the §103 analysis.

I found the PTAB petition containing the '870-specific prior-art disclosures. Let me pull the full prior-art list and the key references.

Obviousness Analysis — U.S. Patent No. 7,031,870 B2 under 35 U.S.C. § 103

A note on the corpus used and a date discrepancy. The task instruction's "Current Date: April 26, 2026" conflicts with the operating context date (2026‑09‑29, per the Google Patents fetch). The discrepancy is immaterial to a § 103 analysis but I flag it per the "flag contradictions" rule. On prior art: the "Prior Art section" available to me is (i) the reference list and incorporated‑by‑reference documents named in the '870 specification itself, and (ii) a set of P.R. 3‑3(C) invalidity disclosures specifically charted against the '870 patent, which I located inside a PTAB petition record. That gives me an unusually well‑grounded prior‑art corpus for this patent, and I build the analysis on it.

---

1. Governing law and the analytical frame

• Statute: The '870 has a filing/priority date of May 28, 2004 — pre‑AIA. Obviousness is therefore governed by pre‑AIA 35 U.S.C. § 103(a) and Graham v. John Deere Co., 383 U.S. 1 (1966), as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007).
• Prior-art window: References published before May 28, 2004 are § 102(a)/(b) art; U.S. patents/applications are § 102(e) art as of their effective filing date. A publication by the inventors themselves dated more than one year before filing is § 102(b) art with no "by another" requirement — this matters enormously here.
• Claim interpretation matters: Per the '870 specification, "index of air re-circulation" = the Supply Heat Index (SHI), defined as (T_in − T_ref)/(T_out − T_ref); "heat dissipating devices" = racks/electronics cabinets; "flow field settings" = CRAC volume flow rates, vent‑tile characteristics, adjustable panel/louver positions, and rack‑mounted fan speeds. I treat these as the controlling constructions for the analysis below.

Level of ordinary skill (POSITA): a bachelor's degree in mechanical or electrical engineering (or equivalent), plus roughly 2–5 years of experience in data‑center thermal management, CFD modeling of electronics cooling, or HVAC control, with working familiarity with raised‑floor/plenum cooling, CRAC units, variable‑frequency drives (VFDs), and rack inlet/outlet temperature instrumentation. This is the level reflected in the references below, which were authored by the '870 inventors' own HP Labs colleagues.

---

2. The prior-art corpus

2.1 The prior art expressly charted against the '870 in the PTAB/litigation record

I located a document captioned "D. P.R. 3‑3(C) Disclosures for the '870 Patent" within a PTAB petition filing (petition 1557498). It lists the following non‑patent references as prior art to the '870 and as "relevant to the state of the art, obviousness and the knowledge of a POSITA as of the '870 Patent priority date":

Source: https://ptacts.uspto.gov/ptacts/public-informations/petitions/1557498/download-documents?artifactId=XNyb0Q9RhDQ57KpzvEQXnQWI6mTySbXgYJumBRx_it852nauAinMpNc

2.2 Patent references of record

2.3 The applicants' own incorporated‑by‑reference and admitted prior art

The '870 specification expressly incorporates: Ser. No. 10/262,879 (lowered‑ceiling returns); Ser. No. 10/425,624 (adjustable panels) and Ser. No. 10/425,621 (adjustable louvers), both filed Apr. 30, 2003; US 6,574,104 (computer‑controlled vent tiles); and Ser. No. 10/446,854 (filed May 29, 2003, "A more detailed description of SHI"). The Background section additionally admits as known: CRAC units with variable capacity compressors and chilled‑water systems, blowers with variable frequency drives to vary volume flow rate, and the whole menu of "flow field settings" recited in the claims. These admissions are usable as prior‑art evidence of the state of the art.

2.4 An important caveat on Ser. No. 10/446,854 / US 7,155,318

Ser. No. 10/446,854 and US 7,155,318 appear to share the same inventive entity as the '870 (Sharma, Bash, Patel) and were commonly owned by HP at the relevant time. Under pre‑AIA § 102(e)/103(c), a U.S. application is not § 102(e) prior art if it is "by another," and commonly owned § 102(e) art is disqualified for obviousness. I therefore do not rely on 10/446,854 / US 7,155,318 as primary § 103 art and treat it as corroborating context only. The same‑inventor analysis does not rescue the Sharma I paper (June 24, 2002), however: a printed publication by the inventors themselves more than one year before the May 28, 2004 filing is § 102(b) prior art, full stop.

---

3. Scope of the independent claims (recap)

From the previously generated sections (treated as authoritative): claim 1 (method), claim 13 (CFD tool), claim 22 (system), and claim 32 (means‑plus‑function system). Claim 1's core steps: detect rack inlet and outlet temperatures; detect CRAC supply‑air temperatures; calculate SHI indices from those three temperatures; determine the indices at various flow field settings; evaluate the components based on changes in the indices. Claim 13 is the same logic implemented in a CFD/design tool; claims 22/32 recast it as an apparatus with sensors + a programmed controller.

---

4. The differences between the prior art and the claims

The single most important reference is Sharma I (D‑7), which is the '870 inventors' own June 24, 2002 paper. It discloses the SHI and RHI dimensionless parameters, the SHI numerator/denominator decomposition, and the near‑linear SHI‑vs‑rack‑inlet‑temperature relationship across racks/flow conditions — i.e., essentially the entire metric and the linearity criterion that the '870 recites in claims 24 and 26, and that the '870 specification reproduces as Equations (1), (4) and (5). Sharma I is therefore § 102(b) art against the very formula the '870 claims.

What Sharma I (and Sharma II, D‑8) does not, standing alone, squarely reduce to practice in a live data center, is the closed‑loop use of the metric: perturb a flow field setting, measure the resulting change in SHI, and use that change to evaluate (a) whether the layout is optimized, (b) which CRAC unit influences which rack, and (c) rack airflow characteristics. That "perturb‑and‑observe‑change" limitation is the only meaningful gap the '870 claims fill.

Every element of that gap is supplied by the secondary references:

• Patel (D‑5, 2002) and US 6,574,104 supply the actuators and the CFD framework: variable‑output fans/compressors, independently controllable vent tiles, and CFD numerical modeling of the data center whose output drives cooling‑system manipulation (claim 21/22 of '104; see also https://insight.rpxcorp.com/patent/[US6574104B2](/patent/US6574104B2)). They explicitly teach varying flow and predicting/observing the thermal consequence.
• Bash (Ex. 1005) and US 6,862,179 supply feedback control of cooling fluid flow based on detected rack temperatures — the "evaluate and adjust" loop into which the SHI metric plugs.
• Guggari (D‑4) and Rambo (D‑6) supply CFD‑based optimization of data‑center layout/flow — directly corresponding to independent claim 13's "CFD tool" and to the layout‑design use case in the '870 specification.
• The Background admissions supply the "flow field settings" menu (VFD‑driven CRAC blowers, computer‑controlled vent tiles, adjustable louvers/panels, rack fans), all of which were known before the '870 filing and several of which the '870 incorporates by reference from the applicants' own earlier applications.

So the differences reduce to: applying a known data‑center recirculation metric (Sharma I), using known variable‑flow actuators and known CFD tools (Patel/'104, Guggari, Rambo), in a known feedback‑control architecture (Bash/'179), and observing how the metric changes.

---

5. Grounds of rejection

Ground 1 — Claim 1 (and claims 24, 26, 28–31) obvious over Sharma I in view of US 6,574,104 (and the applicants' admitted prior art)

• Sharma I discloses calculating SHI from rack inlet/outlet and reference (supply) temperatures and evaluating thermal performance via a linearity criterion — meeting the "calculating indices" and "evaluating … based on" limitations and, for dependent claims 24/26, the near‑linear/equation limitations verbatim.
• US 6,574,104 discloses the variable‑flow field settings (variable‑output fan, variable‑capacity compressor, independently controllable vents) and CFD‑driven manipulation, meeting "determin[ing] the indices at various flow field settings."
• Combining them to yield "evaluate based on changes in the indices" is at most an obvious, predictable application of the '104 actuator set to the Sharma I metric: perturb supply volume flow, read the metric's response. The '870 specification itself concedes the purpose (Background: non‑optimized layouts, wasted energy) that supplies the motivation.
• Dependent claims 28–29 (SHI unchanged ⇒ rack not influenced; SHI changed ⇒ rack influenced; quantify level of influence) and 30–31 (correlate magnitude of SHI change to rack airflow) are the straightforward differential application of a known metric to known perturbations; nothing more is required than routine experimentation of the type KSR holds obvious.

Ground 2 — Claim 13 (CFD tool) obvious over Guggari or Rambo in view of Sharma I (optionally + US 6,574,104)

• Guggari (D‑4) and Rambo (D‑6) teach CFD modeling/optimization of data‑center layouts and airflow, and US 6,574,104 (claim 21/22) teaches using a CFD/numerical model of the data center to drive cooling decisions.
• Sharma I supplies the SHI metric and the linearity criterion to be computed within that CFD model. Programming a CFD tool to output SHI at various simulated flow settings and to flag non‑linear/optimized layouts is the routine automation of an analytical metric — a classic KSR "known technique to improve similar devices in the same way" scenario.
• The '870 specification's own description of the CFD embodiment ("the CFD tool may be programmed to substantially continuously re‑configure the data center infrastructure until a nearly linear relationship is established") is essentially Guggari/Rambo's methodology with Sharma I's metric substituted in.

Ground 3 — Claim 22 (system) and claim 32 (means‑plus‑function system) obvious over Sharma I in view of US 6,574,104 and Bash

• The structural elements — racks with inlet/outlet temperature sensors, CRAC units with supply‑temperature sensors, and a controller — are all disclosed across Sharma I (metric + instrumentation), US 6,574,104 (sensors, vents, variable‑output cooling device, controller), and Bash (controller adjusting cooling based on detected rack temperatures).
• Claim 32's "means for …" limitations map onto the same disclosed structures (temperature sensors, controller/processor with memory), so the means‑plus‑function recitation adds no distinct subject matter. A POSITA would have been motivated to consolidate these known elements into a single controller executing the Sharma I metric over the '104/Bash actuator set, with a reasonable expectation of success.

A note on the possibility of anticipation

If Sharma I (or US 6,574,104's CFD disclosure) is shown to plot SHI across multiple CRAC flow rates or configurations, independent claim 1 could be argued to be anticipated rather than merely obvious. I flag this because the '870's own Figures 4A/4B and the specification's "various flow field settings" language track Sharma I's methodology closely. I cannot verify the exact figures in Sharma I from the available text, so I state this as a hypothesis, not a conclusion.

---

6. Motivation to combine (explicit KSR-aligned rationales)

1. Common field and common problem. All references are directed to cooling racks/data centers — the classic "familiar elements according to known methods" combination KSR endorses. Sharma I and the '870 share authorship with HP Labs colleagues; US 6,574,104/'179 are HP patents; Guggari/Rambo are data‑center CFD papers.
2. The specification itself supplies the motivation. The '870 Background states that conventional data centers are designed "based upon intuition," produce "infrastructure that [is] not efficiently designed," and waste energy — the exact deficiency Sharma I's metric was created to quantify and Patel/'104's actuators were created to correct. That is a strong, record‑based reason to combine.
3. Predictable result / reasonable expectation of success. Applying a flow perturbation and reading the resulting SHI change is a direct, predictable measurement of recirculation response; the '870 specification presents it as such (no unexpected results claimed).
4. Design incentive and known techniques. Once SHI was known (Sharma I, 2002), using the already‑available VFD/vent actuators (Patel/'104, Background admissions) and the already‑available CFD tools (Guggari/Rambo/'104) to drive the metric was the natural next design step — a KSR "obvious to try" with a finite number of identified, predictable solutions.
5. Simultaneous invention / same‑family activity. The existence of US 7,155,318 ("Air re‑circulation index") and the numerous HP follow‑on patents in this family (cited in the search results: e.g., "Correlation of Vent Tile Settings and Rack Temperatures," "Energy Efficient CRAC Unit Operation," US 6,862,179) evidences that applying the SHI index in control and design workflows was the then‑prevailing engineering direction — strong objective evidence of obviousness of the integration, not of non‑obviousness.

---

7. Objective indicia (secondary considerations)

I found no evidence in the available record of nexus‑bearing secondary considerations for the '870 (no unexpected‑results data, no licensing‑driven praise tied to the claimed perturbation technique, no long‑felt‑need evidence attributed to this specific claim). The patent's commercial relevance (evidenced by the Valtrus assertion campaign across E.D. Tex. data‑center defendants) shows value, but value from a broadly practiced industry technique is not, without more, proof of non‑obviousness. I note the absence of evidence explicitly rather than asserting there is none.

---

8. Weaknesses, defenses, and uncertainties

• Assignor‑estoppel / same-inventor arguments do not defeat the Sharma I ground (it is § 102(b) art), but they do neutralize any reliance on Ser. No. 10/446,854 / US 7,155,318 under § 102(e)/103(c). I have therefore not used those as primary art.
• The "various flow field settings + evaluate based on changes" limitation is the '870's only real foothold. A patent owner would argue the references teach static SHI computation and independent flow control, but not the differential/perturbation evaluation. That argument is weakened by (a) the '870 specification's own statement that the perturbation is done "in relatively small amounts… not to create a different flow pattern but to alter the extent and magnitude of existing flow patterns" (i.e., the claim reads on conventional flow setting changes), and (b) the presence of VFD/vent control in the very references the applicants incorporated.
• Claim 13's software/CFD framing raises § 101‑type questions that are outside this § 103 task; I do not address eligibility here.
• Provenance limitation: the D‑4…D‑8 list comes from a P.R. 3‑3(C) invalidity‑contentions exhibit embedded in a PTAB petition file (petition 1557498), not from the '870's own USPTO face citations. That is a legitimate evidentiary source but it is a party's characterization, not a tribunal's finding. I flagged this so the grounds above are not mistaken for adjudicated results.
• Unverified items: the exact patent number of the "Bash" exhibit; the precise numerical dates of Sharma I/II beyond the month/day shown; and the claim‑by‑claim content of Sharma I's figures. These are stated with the uncertainty they carry rather than resolved.

---

9. Bottom line

On the corpus above, US 7,031,870 is vulnerable to § 103 rejection across every independent claim, with the strongest ground being:

Sharma I (D‑7, June 24, 2002) in view of US 6,574,104 (Patel, issued June 3, 2003), further in view of Bash, and optionally Guggari (D‑4)/Rambo (D‑6) for the CFD claim.

Sharma I is the '870 inventors' own § 102(b) publication teaching the SHI metric and its linearity criterion; US 6,574,104 and Bash teach the variable‑flow actuators, sensing, and feedback‑control architecture; and Guggari/Rambo teach CFD‑based layout optimization. The only nominally novel step — perturbing a flow field setting and evaluating the resulting change in SHI — is the predictable, motivated application of a known metric to known controls, with a reasonable expectation of success, and is squarely within KSR. Claims 24 and 26 (near‑linear criterion and the SHI equation) are essentially reproduced in Sharma I, making them the most exposed claims of all.

---

Source URLs used

• https://patents.google.com/patent/US7031870/en (patent text, "Prior art keywords," incorporations by reference)
• https://ptacts.uspto.gov/ptacts/public-informations/petitions/1557498/download-documents?artifactId=XNyb0Q9RhDQ57KpzvEQXnQWI6mTySbXgYJumBRx_it852nauAinMpNc (P.R. 3‑3(C) disclosures charted to the '870)
• https://ptacts.uspto.gov/ptacts/public-informations/petitions/1557498/download-documents?artifactId=KWYrWI0v2iPGskQGX3nddWBoUXlU7W5v_UETgWl2nId6hxHqVhPP0Qc (Petitioner's statement of prior art — Bash, Patel, Bishop, Feeney, Kochavi)
• https://insight.rpxcorp.com/patent/US6574104B2 and https://webapp1.dlib.indiana.edu/virtual_disk_library/index.cgi/[5705037](/patent/5705037)/FID2772/OG/html/1271-1/us06574104-20030603.html (US 6,574,104 claim text)
• https://uspto.report/patent/grant/7,031,870 (bibliographic corroboration)
• https://scholar.google.co.th/citations?hl=en&user=7Nb-szUAAAAJ (Sharma I / Sharma II citation records; US 7,155,318 "Air re-circulation index")