I'll search for information on US patent 8645517, including any CAFC 2026 docket activity.

I have solid bibliographic data. Now let me retrieve the issued patent's claims and search specifically for Federal Circuit (CAFC) 2026 activity.

US Patent 8,645,517 B2 — Summary

I searched for patent number 8645517 specifically (interpreting the number literally; I did not substitute similar numbers). Note: I could not find any CAFC (Federal Circuit) 2026 docket entries for this patent — my CAFC-specific search returned no results. What I did find is active PTAB and district court litigation, described below.

Bibliographic Data

Inventors (as listed): Paul E. Stolorz (Los Altos, CA), John K. Salmon (Chappaqua, NY), Michael S. Warren (Los Alamos, NM), Jeffrey G. Koller (Oxnard, CA), Maksim Yevmenkin (Thousand Oaks, CA), Mark Brady (San Rafael, CA), David Pfitzner (College Park, AU), Ted Middleton (Moorpark, CA), Aric Hagberg (Santa Fe, NM).

Family context: This is a continuation of US 7,860,964 B2 (App. 11/976,648), itself a continuation-in-part of US 7,822,871 B2 (App. 10/259,497). It is part of the Sandpiper/Digital Island CDN patent portfolio acquired by Level 3 and later sold to Sandpiper CDN, LLC in 2024.

Abstract (as published)

"In a framework wherein resources of a content provider may be delivered to clients from different domains, a method distributes the requests based on content-provider policies. In some cases, the domains include at least two distinct content delivery network (CDN) domains. The domains may include a content provider domain. Responsive to a request, either the content provider domain or one of the two CDN domains is selected, the selection being based at least in part on one or more policies set by the content provider."

Plain-Language Overview of the Independent Claim

Caution: My searches returned the full published application text (US 2011/0145386 A1) and its claim 1, but I could not verify the complete, final claim set of the issued patent from authoritative sources in these searches. The patent text you supplied covers the specification but not the claims. So the following is the claim as published, with the caveat that the issued claims may have been amended during prosecution.

• Claim 1 (method) — In a system where a content provider's resources are delivered from multiple domains — at least two of which are distinct CDN domains — the method: (i) upon receiving a request, selects one of the domains based at least in part on one or more policies set by the content provider, and (ii) provides information (e.g., a DNS answer such as an IP address or CNAME) associated with the selected domain.
  • In plain terms: a content provider can set rules (e.g., geographic, load-share, failover/shedding, or CDN-selection rules) and the system uses those rules to decide, per request, whether traffic goes to the provider's own origin servers or to one of two or more different CDNs.

I was not able to authoritatively confirm the number of independent claims or the final issued claim language from the search results. If you need the exact issued claims, the authoritative sources are the USPTO Patent Center (application 12/978,537) or the PTAB's IPR2025-00806 institution decision, which quotes challenged claims of the '517 patent.

Litigation / Post-Issuance Status (relevant, and the closest thing to "2026 docket" activity found)

• PTAB — IPR2025-00806 (Google LLC v. Sandpiper CDN, LLC): Petition challenges Patent 8,645,517 B2; the Board instituted review on Nov. 10, 2025 (Paper 14). Sandpiper CDN is identified as sole real party in interest; related litigation identified as Sandpiper CDN, LLC v. Google LLC, No. 2:24-cv-03951 (C.D. Cal., filed May 10, 2024). A Request for Director Review was filed Nov. 13, 2025 regarding institution (per Sandpiper's mandatory notices).
• District court: Sandpiper CDN, LLC v. Google LLC, 2:24-cv-03951 (C.D. Cal.) — asserted portfolio includes the '517 patent family.
• District court (Microsoft): Sandpiper CDN, LLC v. Microsoft Corp., 2:25-cv-00664 (E.D. Tex.) — the complaint describes the Sandpiper/Digital Island/Savvis/Level 3 acquisition history and asserts CDN portfolio patents (the '517 patent is part of this portfolio; I did not confirm it is individually asserted in that specific case).
• PTAB — IPR2026-00095 (Microsoft v. Sandpiper CDN): filed Nov. 3, 2025, but it challenges US 8,478,903 (not the '517 patent); Microsoft's brief notes eight portfolio patents on which Google or Microsoft has requested IPR.

Uncertainties / Gaps

1. No CAFC 2026 docket found for 8,645,517 — I have no evidence of a Federal Circuit appeal involving this patent as of the current date. If one exists, it did not surface in my searches.
2. Exact issued claims of the '517 patent were not verifiable from my search results; I provided the published claim 1 and flagged this.
3. Expiration status: Google Patents lists "Expired – Fee Related" with adjusted expiration Aug. 7, 2023, while DocketAlarm shows a maintenance-fee payment event for year 8. The PTAB mandatory notices list the '517 patent as "Expired." I could not fully reconcile these records; the PTAB/patent-owner filings treat the patent as expired.

Based on my searches, here is what is known about litigation involving US Patent No. 8,645,517 ("Policy-based content delivery network selection"). I have prioritized live search results (Google Patents litigation annotations, PACER-derived docket sources, RPX, and PTAB records) over any general training data.

District Court Litigation

Sandpiper CDN, LLC v. Google LLC

• Plaintiff: Sandpiper CDN, LLC (assignee of the '517 patent from Level 3 Communications, LLC; assignment recorded April 26, 2024)
• Defendant: Google LLC (the initial complaint also named Alphabet Inc. per RPX reporting; Google LLC is the defendant in the active docket)
• Jurisdiction: U.S. District Court for the Central District of California
• Case No.: 2:24-cv-03951 (initially assigned to Judge Philip S. Gutierrez, docket No. CV 24-03951 PSG (RAOx); later re-assigned as 2:24-cv-03951-AB (RAO), Judge André Birotte Jr.)
• Filing date: May 10, 2024
• Patents-in-suit: The original complaint asserted six CDN-related patents, including the '517 patent (8,645,517). The six were 8,478,903; 8,595,778; 8,645,517; 8,719,886; 9,021,112; and 10,924,573 (per RPX and the complaint analysis).
• Status/outcome (as of the latest available docket entries):
  • On September 16, 2024, the court dismissed with prejudice Counts II and IV (alleging infringement of the '778 and '886 patents) on § 101 subject-matter-eligibility grounds (Dkt. 28).
  • Sandpiper amended the complaint to add another patent; the '517 patent remained among the five asserted patents (per Google's motion to stay: remaining patents include 8,478,903, 10,924,573, 8,645,517, 9,021,112, and one more).
  • Google filed IPR petitions challenging all asserted claims and moved to stay the case pending the IPRs (motion dated ~May 23, 2025; hearing set June 27, 2025).
  • Case status: Open/pending (no final judgment located in my search results).

PTAB / IPR Proceedings Directed to the '517 Patent

Google LLC v. Sandpiper CDN, LLC, IPR2025-00806

• Petitioner: Google LLC
• Patent Owner: Sandpiper CDN, LLC
• Tribunal: Patent Trial and Appeal Board (PTAB)
• Challenged patent: US 8,645,517 (the '517 patent)
• Status: The Google Patents page for the '517 patent lists IPR2025-00806 as "filed (Pending - Instituted)." I could not confirm the exact petition filing date or the institution decision date from my searches; it was filed in the same wave of Google IPR petitions (spring 2025) as related petitions such as IPR2025-00826 (filed April 15, 2025, challenging a different patent).

Notes and Caveats

• The Google Patents page for US 8,645,517 itself flags two litigation items: the C.D. Cal. case 2:24-cv-03951 and PTAB case IPR2025-00806, plus a "first worldwide family litigation" entry (Darts-ip family 46329570). These are consistent with the Google case above.
• I did not find evidence that US 8,645,517 is asserted in Sandpiper's other district court cases (e.g., Sandpiper CDN, LLC v. Comcast Cable Communications, LLC, No. 2:24-cv-00886 (E.D. Tex.); Sandpiper CDN, LLC v. [Microsoft Corp.](/litigations/by-plaintiff/Microsoft%20Corp.), No. 2:25-cv-00664 (E.D. Tex.) — the latter's amended complaint lists patents 8,478,903; 8,924,466; 9,456,053; 6,922,883; 9,762,692; and 10,701,173, not the '517 patent). I could not fully verify the Comcast patent list, so I cannot rule out that the '517 patent appears there.
• The patent's legal status on Google Patents is "Expired - Fee Related" (adjusted expiration August 7, 2023), which may bear on enforceability, but the litigation above is the known record.

If you need the exact IPR2025-00806 filing/institution dates or a complete docket history of 2:24-cv-03951, I'd recommend checking PACER, the PTAB's public API (ptacts.uspto.gov), or Unified Patents' portal directly — those were the sources my searches surfaced but could not fully mine.

I found the key proceeding: IPR2025-00806 — Google LLC v. Sandpiper CDN, LLC. Let me dig into the institution decision, panel, and any other proceedings on this patent.

Proceedings overview

The USPTO ODP block in this prompt reports zero AIA trial proceedings on US 8,645,517 as of the most recent ingest — but that ingest is stale. Live web search surfaces one recently-filed, pending proceeding: IPR2025-00806 (Google LLC v. Sandpiper CDN, LLC), filed 2025-04-14 and instituted 2025-11-10 as to all claims 1–20 and all grounds. Status breakdown: 1 active (instituted, awaiting Final Written Decision), 0 claims invalidated, 0 settled, 0 institution denials.

Bottom line for a defendant: No claims are canceled yet — all 20 claims are under active, instituted IPR challenge, and the parallel district court case (Sandpiper CDN, LLC v. Google LLC, No. 2:24-cv-03951, C.D. Cal.) has been fully stayed until the IPRs are finally resolved, including appeals. If the IPR were won, the entire patent falls; if it is lost, the patent comes out of the IPR "hardened" against Google's art — but untested against other references.

IPR2025-00806 — Google LLC v. Sandpiper CDN, LLC

• Type: Inter Partes Review
• Filed: 2025-04-14
• Status: Trial Instituted (per PTAB docket sources; not yet indexed in the ODP API ingest provided here — flagged as the gap between the ODP block and the live record)
• Judge panel: Administrative Patent Judges Mitchell G. Weatherly (lead), Sheila F. McShane, and Michael T. Cygan — opinion for the Board by APJ McShane, with a concurring opinion by APJ Cygan. The earlier Director-level referral decision (Paper 13) was by Acting Chief Administrative Patent Judge Kalyan K. Deshpande.
• Petition grounds (all claims 1–20 challenged, all under 35 U.S.C. § 103 obviousness):
  • Ground 1: Claims 1, 4–9, 12–20 obvious over Beaumont (US 2002/0169890) in view of Hasan (US 7,082,464). Google argued Beaumont disclosed a policy-based DNS/CDN traffic director with preference values structured as a tree, and Hasan supplied the tree-structured GUI for configuring those policies.
  • Ground 2: Claims 2, 3, 10, 11 obvious over Beaumont and Hasan in view of Biliris (US 2002/0078233), which taught sub-domain-level DNS redirection/outsourcing (e.g., "images.xyz.com" vs. "www.xyz.com").
  • Ground 3: Claims 1–20 obvious over Ben-Shaul (US 2002/0010798) alone or in view of Hasan, Ben-Shaul allegedly teaching a CDN with policy-based DNS "directives" and hierarchical GUIs.
• Institution decision: Granted in full — 2025-11-10 (Paper 14). The panel found a reasonable likelihood of prevailing "with respect to at least one claim" and instituted "as to all of the challenged claims of the '517 patent and all of the asserted grounds of unpatentability in the Petition." (Institution Decision, Paper 14, Nov. 10, 2025)
  • Discretionary-denial skirmish: Patent Owner sought discretionary denial (Paper 6), arguing the expired-patent "settled expectations" doctrine. The Acting Chief APJ denied the request (Paper 13, Sept. 2025), reasoning that the parallel district court case was stayed pending institution, reducing duplication, and that Google persuasively argued the patent expired in 2023 (adjusted expiration per the patent record: 2023-08-07), undermining settled expectations of enforcement. (Director referral decision, Paper 13). Patent Owner's subsequent petition for Director Review (Paper 16) was rejected.
• Final Written Decision: Not yet issued. Statutory FWD deadline is one year from institution — due on or about 2026-11-10 (extendable up to ~2027-05-10 for good cause). No oral-hearing date was public in the sources reviewed.
• Settlement / termination: None reported. The proceeding is active.
• Appeal: Not applicable yet (no FWD to appeal). The only appellate-adjacent event is the denied Director Review of the referral decision.
• Defensive value: All 20 claims are live targets, but nothing is dead yet. The IPR's existence has already neutralized the patent in the co-pending Google litigation — the C.D. Cal. case (filed 2024-05-10) was stayed on 2026-01-22 (Dkt. 93) until Google's IPRs are finally resolved, including appeals. A non-Google defendant cannot ride Google's estoppel, but a Google-facing defendant gets a free stay and a credible threat to the entire patent.

Strategic summary

Claim map — CANCELED vs. SUSTAINED vs. UNTESTED. As of today, no claim of the '517 patent has been canceled or sustained by a final PTAB decision. All 20 claims (1–20) — the complete claim set — are under active challenge in IPR2025-00806, and the institution decision covered every claim and every ground. So the honest snapshot is: 0 canceled, 0 sustained, 20 untested-but-challenged. Until the FWD issues, any demand letter citing any claim of the '517 patent is at least colorable on validity — but the risk of total cancellation is real, and any defendant facing Sandpiper should be telling the patent owner that the entire patent is one Board decision away from extinction.

Estoppel landscape. No FWD has issued, so § 315(e)(2) estoppel has not yet attached to anyone. When the FWD issues, Google and its privies will be barred from re-raising in the district court any § 102/§ 103 ground on instituted claims that it raised or reasonably could have raised — and Google has already given a Sotera-style stipulation to that effect. For a new defendant (not Google or its privy), the § 315(b) one-year bar runs from service of the complaint, and unused art remains fully available: the instituted grounds are limited to Beaumont, Hasan, Biliris, and Ben-Shaul, so alternative CDN/DNS prior art (e.g., Akamai-era, Cisco DistributedDirector, other 1990s–2001 DNS load-balancing and CDN routing references) is untouched and can support a fresh petition or a district court § 282 defense. Note also that the patent is expired (2023-08-07) — damages are limited to pre-expiration conduct, which sharply constrains any plaintiff's leverage.

Pattern signals. This is a coordinated, multi-patent campaign: Sandpiper CDN, LLC — a Delaware entity formed in 2024 that acquired the patent family from Level 3 — sued Google in C.D. Cal. on 2024-05-10, and Google filed IPRs against all five asserted family patents: IPR2025-00806 ('517), IPR2025-00826 (US 9,021,112), IPR2025-00860, IPR2025-00969, and IPR2025-01010 — all referred to the Board in Sept.–Oct. 2025 and all instituted in Nov.–Dec. 2025. Sandpiper has fought institution hard (discretionary-denial briefing, Director Review petitions) and lost. The "Unified Patents" reference on the Google Patents page is only a data-attribution source, not a petitioner or aggregator in the chain — the actual petitioner is Google. A separate Microsoft Corporation v. Sandpiper CDN, LLC (IPR2026-00095) exists and references the '517 IPR's institution decision, but the sources reviewed do not confirm which patent Microsoft challenges — treat it as related-family activity and verify on PTAB E2E before relying on it.

Recommended next steps

1. If you are a defendant facing the '517 patent today: Do not wait. The single most valuable document on the record is the Institution Decision (Paper 14, 2025-11-10), which put every claim in play — cite it in any stay motion and in settlement leverage: "the Board instituted IPR as to all of the challenged claims of the '517 patent and all of the asserted grounds" (Paper 14). The patent is expired; damages are capped at pre-August-2023 conduct.
2. Milestones to watch: FWD due ~2026-11-10 (one-year statutory deadline from institution; PTAB may extend to ~2027-05-10); oral hearing typically scheduled 1–3 months before the FWD. The C.D. Cal. stay (Dkt. 93, 2026-01-22) lifts 30 days after the IPRs are "finally resolved, including any appeals," so a CAFC appeal of any FWD could push the stay well into 2027–2028.
3. If you are not Google and are newly served: run the § 315(b) clock from service, and immediately assess non-Beaumont/Hasan/Biliris/Ben-Shaul prior art — those grounds are reserved for Google. A fresh IPR using different references is the cleanest path, because a successful Google FWD will cancel the claims for everyone (collaterally), while a Google loss does not bar your independent challenge.
4. Verify the live docket before any filing: the ODP ingest in this prompt predates the proceeding — confirm current status, papers, and deadlines on USPTO PTAB E2E/PTACTS (search "IPR2025-00806") or CourtListener. No Federal Circuit appeal exists yet, so nothing to monitor there until the FWD issues.

I have strong leads. Let me now dig into the specific USPTO assignment records — reel/frame numbers and correspondents — for both the 2012 chain-of-title cleanup and the 2024 Sandpiper transfer.

I have enough verified material to reconstruct the chain, though I must be transparent that the USPTO Assignment Center's reel/frame and correspondent-of-record data for the 2012 records was not retrievable through the sources available to me in this session. I will not fabricate those fields.

Inventors

All nine named inventors — Paul E. Stolorz, John K. Salmon, Michael S. Warren, Jeffrey G. Koller, Maksim Yevmenkin, Mark Brady, David Pfitzner, Ted Middleton, Aric Hagberg — are associated with the "adaptive traffic control" (ATC) / Footprint CDN technology lineage. The specification itself references the ATC name-server framework, "customer.footprint.net," "customer.sandpiper.net," and "Level 3 Communications' Content Delivery Network," placing the underlying work in the Sandpiper Networks → Digital Island → Cable & Wireless chain.

• Employer at time of invention (priority 2001-09-28): Digital Island, Inc. (which became Cable & Wireless Internet Services, Inc. after Cable & Wireless acquired Digital Island in 2001). This is consistent with the recorded 2012-01-17 "Change of Name" from Digital Island, Inc. to Cable & Wireless Internet Services, Inc.
• Employer at time of the continuation filing (2010-12-24): Savvis Communications Corporation (the then-owner per the recorded chain; Level 3's acquisition of Savvis closed in 2011 and the chain-of-title records were recorded January 2012).
• Unusual pattern: None confirmed. I found no evidence of mass inventor departure or inventor-side assignments; the inventors appear to have been Digital Island employees whose work was assigned to the company in the ordinary course. Note that none of the nine are Sandpiper Networks founders Farber/Swart — those two are named only on the sibling '903 patent, not this one.

Original assignee

Level 3 Communications LLC is the entity recorded on the issued patent (per Google Patents "Original Assignee").

• Product embodiment: Yes. Level 3 operated a real CDN — the former Sandpiper/Digital Island "Footprint" network — and the '517 claims (policy-based content-delivery-network selection via DNS) describe the ATC system Level 3 operated commercially. The complaint in Sandpiper CDN v. Google itself pleads that Level 3 "became one of the foremost CDN operators in the United States."
• Line of business: Tier-1 telecom / content delivery network operator.
• Current status: Subsidiary of Lumen Technologies (f/k/a CenturyLink, which acquired Level 3 in 2017). Per the Sandpiper complaint and RPX, Level 3 exited the CDN market in 2023 and sold its CDN patent assets to Sandpiper CDN in 2024. The patent itself lapsed (Expired – Fee Related) on 2023-08-07 — i.e., it was expired before it was sold and asserted.

Assignment timeline

The USPTO Assignment Center does have records for this patent. I verified the 2024 Level 3 → Sandpiper transfer (Reel 068256, Frames 0091–0115, cited as PTAB exhibit EX1036 in IPR2026-00190) and the 2012 chain-of-title cluster (from Google Patents legal events). I was not able to retrieve the reel/frame numbers or the recorded correspondent-of-record names for the 2012 records in this session — those fields are marked as not retrieved rather than guessed. Execution dates for the 2012 cluster are not shown in the available source; the dates below are recording dates.

• 2012-01-03 (recorded) — reel/frame not retrieved

  • Conveyance: Merger
  • Assignor: Savvis, Inc.
  • Assignee: Savvis Communications Corporation
  • Correspondent: not retrieved (USPTO Assignment Center lookup required)
  • Context: Internal corporate step in the Savvis lineage — Savvis, Inc. merged into/renamed as Savvis Communications Corporation, part of the title cleanup around Level 3's 2011 acquisition of Savvis.

• 2012-01-03 (recorded) — reel/frame not retrieved

  • Conveyance: Assignment of Assignors' Interest
  • Assignor: Cable & Wireless Internet Services, Inc.
  • Assignee: Savvis Asset Holdings, Inc.
  • Correspondent: not retrieved
  • Context: Retroactive perfection of the 2004 sale of Cable & Wireless America (including the Digital Island CDN business) to Savvis.

• 2012-01-03 (recorded) — reel/frame not retrieved

  • Conveyance: Change of Name
  • Assignor: Savvis Asset Holding, Inc.
  • Assignee: Savvis, Inc.
  • Correspondent: not retrieved
  • Context: Name-change step in the same Savvis corporate chain.

• 2012-01-03 (recorded) — reel/frame not retrieved

  • Conveyance: Security Agreement
  • Assignor: Savvis Communications Corporation
  • Assignee: Wells Fargo Foothill, Inc., as Administrative Agent
  • Correspondent: not retrieved
  • Context: Financing lien against the patent under Savvis's credit facility (a debt-security interest, not an ownership transfer; no release appears in the events I retrieved).

• 2012-01-05 (recorded) — reel/frame not retrieved

  • Conveyance: Assignment of Assignors' Interest
  • Assignor: Savvis Communications Corporation
  • Assignee: Mount Shasta Acquisition LLC
  • Correspondent: not retrieved
  • Context: Transfer into Level 3's named acquisition vehicle for the Savvis deal — the step that moved the family out of the Savvis chain.

• 2012-01-05 (recorded) — reel/frame not retrieved

  • Conveyance: Merger
  • Assignor: Mount Shasta Acquisition LLC
  • Assignee: Level 3 Communications LLC
  • Correspondent: not retrieved
  • Context: Merger of the acquisition vehicle into Level 3 Communications LLC, vesting title in the operating company. Note: this "original assignee" of record on the issued patent in fact acquired the application via this 2012 chain.

• 2012-01-17 (recorded) — reel/frame not retrieved

  • Conveyance: Change of Name
  • Assignor: Digital Island, Inc.
  • Assignee: Cable & Wireless Internet Services, Inc.
  • Correspondent: not retrieved
  • Context: Retroactive recordation of the 2001–2002 Digital Island → Cable & Wireless name change, closing the loop on the oldest link in the chain.

• 2024-04-24 (executed per RPX; complaint pleads agreement of 2024-03-29) / recorded 2024-04-26 — Reel 068256, Frames 0091–0115

  • Conveyance: Assignment of Assignors' Interest
  • Assignor: Level 3 Communications, LLC
  • Assignee: Sandpiper CDN, LLC
  • Correspondent: not retrieved from my sources; the PTAB exhibit (EX1036, IPR2026-00190) confirms reel/frame only. Sandpiper's litigation counsel in the Google suit is Robert H. Reckers / Mayela C. Montenegro-Urch — noted for completeness, but that is docket counsel, not assignment correspondent, and I will not conflate the two.
  • Context: Fire-sale-style monetization transfer of 80+ US CDN patents from an operating company (Lumen's Level 3 subsidiary) to a Delaware LLC formed 2024-03-21 (RPX), immediately before suit.

Timeline diagram

timeline
    title Ownership of US 8645517
    2001 : Family priority date
    2002 : Parent application filed
    2010 : Continuation application filed
    2012 : Digital Island to Level 3 chain recorded
    2014 : Patent issued to Level 3
    2017 : Level 3 acquired by CenturyLink
    2023 : Patent expired
    2024 : Assigned to Sandpiper CDN LLC
         : Suit filed against Google

NPE / troll-pattern signals

1. Shell-entity transfer — present. Reel 068256/Frames 0091–0115, recorded 2024-04-26: Level 3 Communications, LLC → Sandpiper CDN, LLC. Sandpiper CDN is a Delaware LLC formed 2024-03-21 (RPX), with no products and no business other than asserting the acquired Level 3 CDN patents; its own complaint pleads that Level 3 "decided to exit the CDN market in 2023 and began selling off its CDN assets." The name is an homage to the defunct Sandpiper Networks — naming alone is not a finding, but formation date + zero commercial operations + immediate assertion are concrete evidence.

2. Known asserter in the chain — present. Sandpiper CDN, LLC is not on the classic Acacia/Marathon/IPNav lists, but RPX explicitly identifies it as an NPE formed to assert Level 3 CDN patents, and its profile matches a high-frequency plaintiff: it sued Google (C.D. Cal. 2:24-cv-03951) asserting six patents (8,595,778; 8,645,517; 8,719,886; 8,478,903; 9,021,112; 10,924,573) and is defending multiple IPRs (IPR2025-00806 — instituted Nov. 10, 2025; plus IPR2025-00826, -00969, and Microsoft's IPR2026-00190/00095). The '517 patent was expired when sold (2023-08-07), so the assertion is pure damages monetization.

3. Repeat correspondent across the chain — unclear. I could not retrieve the recorded correspondent-of-record names for any entry in this session; the only corroborated reel/frame is 068256/0091–0115. I will not speculate. (Sandpiper's litigation counsel, Robert H. Reckers, appears repeatedly on the docket, but that is not assignment-correspondent evidence.)

4. Cascading transfers — not present (in the NPE sense). The five 2012 records (recorded 2012-01-03 through 2012-01-17) are a compressed corporate chain-of-title cleanup — name changes, a merger, an acquisition-vehicle transfer, and a lender security agreement — executed to perfect Level 3's title after its acquisition of Savvis. That is a reorg cascade, not an LLC-shell cascade. The NPE step is a single transfer (Level 3 → Sandpiper).

5. Pre-litigation transfer — present. Assignment recorded 2024-04-26 (Reel 068256/0091–0115); first infringement suit Sandpiper CDN, LLC v. Google LLC, 2:24-cv-03951 (C.D. Cal.) filed 2024-05-10 — roughly two weeks between recordation and filing, on a patent that had already expired. The transfer was plainly arranged to enable assertion with clean standing.

6. Bankruptcy fire-sale — not present. Level 3 was not in bankruptcy at the time of the 2024 sale; it was a Lumen subsidiary exiting the CDN market. (Level 3's own 2003 Chapter 11 reorganization predates and is unrelated to this chain.)

7. Privateering — present (moderate). An operating company (Lumen/Level 3) transferred 80+ patents to a newly formed NPE that immediately sued Google over competing CDN products, after Level 3 exited the market. RPX expressly frames this as "not Level 3's first transfer to an NPE," citing the earlier April 2017 transfer of ~110 US assets to Optic153 LLC / Equitable IP Corporation (John T. Meli Jr.). The complaint's own narrative (CDN competition depressing Level 3's revenue, then selling the patents) reads as monetization against a market rival.

8. Defensive aggregator (anti-NPE) — not present. The chain terminates at Sandpiper CDN, an asserting entity, not at RPX/AST/LOT/Unified/OIN.

Verdict

NPE — high confidence.

Two strong signals drive this: (1) the shell-entity transfer — Level 3 → Sandpiper CDN, a Delaware LLC formed 2024-03-21 with no products, recorded 2024-04-26 at Reel 068256/Frames 0091–0115; and (5) the pre-litigation transfer — recordation two weeks before the 2024-05-10 suit in C.D. Cal. (2:24-cv-03951), on a patent already expired 2023-08-07. RPX independently characterizes Sandpiper CDN as an NPE and notes this is Level 3's second such transfer (after Optic153/Equitable IP in 2017), supporting a privateering reading. The one gap — correspondent-of-record data for the 2012 records — does not weaken the verdict, which rests on the 2024 transfer and suit timing rather than attorney-recurrence.

Verification: search patent number 8645517 at the USPTO Assignment Center (https://assignmentcenter.uspto.gov/) — Reel 068256/Frames 0091–0115 is the Level 3 → Sandpiper CDN record; the 2012 chain entries (Savvis/Cable & Wireless/Digital Island/Mount Shasta) are the earlier records whose reel/frame and correspondent details I could not retrieve here and should be confirmed directly against that database.

1. Patent Identification (verified)

US 8,645,517 B2 — "Policy-based content delivery network selection"

• Inventors: Paul E. Stolorz; John K. Salmon; Michael S. Warren; Jeffrey G. Koller; Maksim Yevmenkin; Mark Brady; David Pfitzner; Ted Middleton; Aric Hagberg
• Original Assignee: Level 3 Communications, LLC (now listed: Mount Shasta Acquisition LLC / Sandpiper CDN LLC)
• Application No. 12/978,537, filed 2010-12-24; granted 2014-02-04; effective priority claimed to 2001-09-28 (Provisional 60/325,177 via U.S. 10/259,497 → US 7,822,871 B2; continuation chain also includes US 7,860,964 B2)
• Status: Expired – Fee Related (adjusted expiration ~2023-08-07); active litigation (C.D. Cal. 2:24-cv-03951) and instituted PTAB IPR2025-00806 per the Google Patents family record.

Important caveat on claims: The excerpt of the granted patent you supplied ends in the Detailed Description and does not include the granted claims of 8,645,517. For claim mapping below I use the representative independent claims of the continuation family (same title, same specification), as published in US 2008/0215718 A1 and granted in US 7,860,964 B2:

• Claim 1 (family): In a framework where resources are delivered from a plurality of domains comprising at least two distinct CDN domains, selecting one of the domains responsive to a request based on one or more policies set by the content provider, and providing information associated with the selected domain.
• Claim 11 (family): Obtaining content-provider policies including a relative load share for each of at least two CDNs; receiving a DNS request from a client's resolver; selecting one CDN based on the policies; providing a domain name (CNAME) of the selected CDN.
• Claim 19/23/24 (family): Selecting between a content-provider domain (returning an IP address) and one of a plurality of distinct CDN domains (returning a CNAME), based on load-share policies; repeated policy refresh at regular intervals.

2. Most Relevant Cited Prior Art (references listed on the face of 8,645,517)

The full "References Cited" list is reproduced at freepatentsonline.com/8645517.html and uspto.report/patent/grant/8645517. I could not retrieve every entry within the search budget, so the following are the most relevant references I could verify from the citation record. Each is analyzed for potential anticipation under pre-AIA 35 U.S.C. § 102 (the statute applicable to this filing chain).

A. US 7,185,052 B2 — "Meta content delivery network system" (Mark Day; Digital Island/Level 3 family)

• Publication: 2007-02-27 (filed 2002-05-16)
• Description: A "meta-CDN" layer that mediates delivery of content across multiple, distinct content delivery networks, selecting among CDNs and routing client requests to a chosen CDN/domain.
• Anticipation analysis: Most relevant to the independent claims. Discloses the claimed framework of multiple distinct CDN domains, request-driven selection among them, and supplying the client with the selected CDN's name/address. Potentially anticipates family claims 1, 11, 19, 23, 24 (multi-CDN selection), provided the load-share policy limitation is found (the '052 disclosure is policy/rules-based but load-share weighting across CDNs may be weaker).
• Timing caveat: '052 was filed 2002-05-16, after the 2001-09-28 provisional priority date of 8,645,517. It is § 102 prior art only if the asserted claims are not entitled to the 2001 priority date (e.g., if the multi-CDN selection limitations are treated as new matter added in the 2007/2010 continuations). This is the pivotal validity question for this reference.

B. US 7,054,935 B2 — "Internet content delivery network" (Farber et al.; Digital Island)

• Publication: 2006-05-30 (filed 1999-07-16; family priority to 1998-02-09/1998-07-14, e.g., US 6,185,598 "Optimized network resource location")
• Description: DNS-based global content delivery: a content provider's hostname is resolved by an authoritative DNS system that selects among distributed servers/CDN nodes based on client (resolver) location, network latency, server load, and availability.
• Anticipation analysis: Clearly § 102 art (well before 2001). Discloses DNS-request-driven selection of a serving domain based on provider-configurable criteria and returning the selected server's IP/CNAME. Potentially anticipates family claims 1 and 23 (selecting a serving domain responsive to a request based on provider policies; returning address information). Does not clearly disclose splitting load among two distinct CDN domains per a content-provider load-share policy (claims 11, 19, 24).

C. US 7,010,578 B1 — "Internet content delivery service with third party cache interface support" (Lewin et al.; Akamai)

• Publication: 2006-03-07 (filed 2000-09-20)
• Description: CDN service with third-party cache-interface support; client DNS requests are redirected (e.g., via CNAME) into the CDN, which selects an edge server.
• Anticipation analysis: § 102 art. Discloses DNS redirection to a CDN domain and returning server addresses. Potentially anticipates the single-CDN selection aspects of claims 1, 11, 19, 23 but not the two distinct CDNs with a content-provider-specified load share limitation.

D. US 6,553,420 B1 — "Method and apparatus for distributing requests among a plurality of resources" (Karger et al.; MIT)

• Publication: 2003-04-22 (filed 1998-09-29)
• Description: Distributes client requests among replicated resources (servers/caches) by mapping requesters to resources (e.g., consistent hashing, load balancing) in a global hosting system.
• Anticipation analysis: § 102 art. Relevant to the load-share/request-distribution mechanics of claims 4, 5, 14–17, 22, 25 (distributing requests among resources according to relative weights). Not directed to CDN-domain selection or CNAME answers; weaker on the "distinct CDN domains" element of the independent claims.

E. US 6,996,616 B2 — "HTML delivery from edge-of-network servers in a content delivery network (CDN)" (Leighton et al.; Akamai)

• Publication: 2006-02-14
• Description: DNS-based CDN that serves HTML from edge servers selected based on client location/network conditions; provider content is delivered from the CDN domain.
• Anticipation analysis: § 102 art. Supports the DNS-selection/address-return elements of claims 1, 23; no multi-CDN load-share selection.

F. US 7,103,645 B2 — "Method and system for providing content delivery to a set of participating content providers" (Leighton et al.; Akamai)

• Publication: 2006-09-12
• Description: CDN service where participating content providers' content is delivered via the CDN under service policies; DNS-based request routing.
• Anticipation analysis: § 102 art. Relevant to provider-policy-driven routing to a CDN domain (claims 1, 11, 23); no disclosure of dividing traffic between two distinct CDN domains by load share.

G. US 7,096,266 B2 — "Extending an Internet content delivery network into an enterprise" (Lewin et al.; Akamai)

• Publication: 2006-08-22
• Description: Extends CDN services into enterprise networks, routing requests to the best available serving location.
• Anticipation analysis: § 102 art. Marginal; relevant only to generic request-routing/address-return elements.

H. US 2007/0055764 A1 — "Method and system for tiered distribution in a content delivery network" (Dilley et al.; Akamai)

• Publication: 2007-03-08 (application claiming priority to 2002-04-08)
• Description: Tiered content distribution within a CDN, with hierarchical/overflow server selection and failover.
• Anticipation analysis: Relevant to tiered failover/overflow limitations (family claims of the parent '871 patent; also claim 24's "content source vs. CDN" fallback). Timing depends on priority-date entitlement, as with '052.

I. US 7,058,706 B2 — "Method and apparatus for determining latency between multiple servers and a client" (Iyer et al.; Cisco)

• Publication: 2006-06-06
• Description: Client-aware selection of a server based on measured latency between servers and the client.
• Anticipation analysis: § 102 art. Relevant only to location/latency-based selection aspects, not the multi-CDN policy selection.

J. US 7,007,089 B2 / US 2006/0143293 A1 — "Content delivery network map generation using passive measurement data" (Freedman)

• Publication: 2006-02-28 / 2006-06-29
• Description: Building network maps from passive measurements to route CDN requests based on network proximity.
• Anticipation analysis: § 102 art. Supports the "network-aware/adaptive" selection background, not the CDN-selection policy claims.

K. US 6,976,090 B2 — "Differentiated content and application delivery via internet" (Ben-Shaul et al.)

• Publication: 2005-12-13
• Description: Differentiated delivery of content/applications over the Internet via distributed platforms.
• Anticipation analysis: § 102 art. Background art for distributed delivery; not a strong single-reference anticipation candidate for the multi-CDN claims.

L. Older load-balancing references visible in the citation list (uspto.report)

• e.g., US 5,774,660 (Brendel, 1998) — "World-wide-web server with delayed resource-binding for resource-based load balancing" and US 5,774,668 (Choquier, 1998) — "System for allocating resources among requesters…"
• These are early (1998) request-routing/load-allocation patents; § 102 art that undermines the generic request-distribution elements but does not address multi-CDN policy selection.

3. Bottom-Line Anticipation Assessment

Most relevant single reference: US 7,185,052 B2 ("Meta content delivery network system") — it is the only face-of-patent citation squarely directed to a meta layer selecting among multiple distinct CDNs, which is the heart of the independent claims. Its utility as § 102 prior art turns on whether the 8,645,517 claims are entitled to the 2001-09-28 priority date; if not, '052 (filed 2002-05-16) is the best anticipation candidate. If the 2001 date holds, the strongest unequivocally prior art is the Digital Island/Akamai CDN line (7,054,935; 7,010,578), which anticipates the single-CDN aspects but leaves the multi-CDN load-share limitation as the likely point of patentability.

Completeness caveat: The complete citation list (several dozen U.S. patents, published applications, and likely non-patent literature) is available at freepatentsonline.com/8645517.html and uspto.report/patent/grant/8645517. I verified the references above from the search-returned citation records; I could not exhaustively retrieve and verify every entry within the available search steps, and I did not fabricate filing details for entries I could not confirm. For a formal IPR/validity analysis, pull the full front-page citation list from the USPTO Patent Center record (Application 12/978,537) and confirm the granted claim set of 8,645,517 B2 directly, since the claim text was not included in the supplied excerpt.

Obviousness Analysis — U.S. Patent 8,645,517 B2 ("Policy-Based Content Delivery Network Selection")

A. Scope of analysis and sources used

The full patent text provided (description, drawings list, and assignee/priority history) does not reproduce the claims or the "References Cited" section verbatim. I have therefore reconstructed the claim scope from (1) the published abstract, (2) the specification text provided, (3) the sibling/family applications (US 7,860,964 B2 and US 7,822,871 B2, which share the same priority chain and inventors), and (4) the examiner-cited reference lists for US 8,645,517 B2 as published by Justia (patents.justia.com/patent/8645517), USPTO.report (uspto.report/patent/grant/8645517), and PubChem/Google Patents. Where a claim element cannot be quoted verbatim, I flag it as a "representative limitation" rather than fabricating claim text.

Patent posture relevant to obviousness:

• Effective priority date: provisional 60/325,177 (2001-09-28) → US 10/259,497 (2002-09-30, now US 7,822,871 B2) → US 12/978,537 (2010-12-24, the '517 patent). Because the application was filed 2010 and claims priority through a 2001 provisional, the § 103 analysis must first confront whether the 2001 priority date is valid for each claim; if a claim limitation lacks written-description support in the 2001/2002 disclosures, references dated after 2001-09-28 become available as § 102/§ 103 prior art.
• The patent is actively contested: PTAB IPR2025-00806 (Unified Patents, pending/instituted) and California Central District Court case 2:24-cv-03951 are listed on the Google Patents record. The institution of IPR2025-00806 indicates that at least one obviousness combination has already been found to present a "reasonable likelihood of prevailing" on at least one challenged claim.
• The '517 patent carries a terminal disclaimer relative to its family members (per the face of sibling US 7,860,964 B2), confirming the family's claims are substantively overlapping.

Representative claim scope. The invention, as disclosed, is a DNS-based "adaptive traffic control" (ATC) framework in which a content provider's traffic is distributed among the provider's own servers and one or more distinct CDN domains based on content-provider-defined policies. Representative limitations across the independent claims (as reflected in the abstract and specification) include:

1. Receiving a DNS request (e.g., a resolver query for a hostname).
2. Determining a location/attribute of the requester (e.g., resolver IP, geo/CIDR block).
3. Accessing a policy database containing subscriber policies (and possibly operator/regulatory policies).
4. Selecting one or more servers/domains — including at least one CDN domain — based on the policies (load share, shed fraction, tiered failover, geographic, regulatory).
5. Returning IP addresses and/or CNAME answers (e.g., pointing to "customer.footprint.net" / "demo.c.footprint.net" CDN domains), including distributing traffic between two distinct CDNs in a specified ratio (e.g., the 1.5:1 "Level 3 CDN" vs. "Other CDN" example of FIGS. 10(l)-10(n)).
6. Dynamic monitoring/polling of server state (monitored/managed servers, XML load-feedback files) and updating policies accordingly.

B. Key prior-art references (from the "References Cited" record of the '517 patent)

The examiner cited an unusually large field of U.S. patents. The most probative references for a § 103 attack, all of which are pre-September 2001 (i.e., available even under the claimed priority date), are:

Non-patent literature on the face of the patent: IBM TDB "Local Area Network Server Replacement Procedure" (Jan. 1995) (server availability/failover), and the Gilbane Report (Jun. 1999) on content-management/CDN industry practice.

Not prior art: The family members themselves (US 7,860,964 B2, US 7,822,871 B2, US 9,203,636 B2) share the same priority chain and cannot be used as § 102/§ 103 references against each other. Note, however, that several of the best references (Farber 6,185,598; 5,978,791; 6,415,280) trace to Digital Island/Sandpiper — the very CDN technology that later became Level 3's footprint.net infrastructure described throughout the '517 specification. This lineage is analytically important: the specification's own examples (CNAMEs to "customer.sandpiper.net" and "customer.footprint.net") show the claimed subject matter operating on infrastructure that the Farber patents already taught how to select and hand off to.

C. Combinations rendering the claims obvious

Combination 1 — Leighton (6,108,703) + Farber (6,185,598) + Brendel (5,774,660)

Claim elements addressed:

• DNS-request-driven, map/location-based selection among replicated servers/CDNs (Leighton: global host system receiving DNS queries, consulting a network map and conditions, returning a server IP; Farber: CNAME-based handoff to a CDN).
• Selection among multiple distinct domains including at least two CDN domains (Farber teaches directing traffic to a CDN via CNAME; the specification's own multi-CDN example is a load-shared set of two CNAMEs).
• Weighted distribution according to policy (Brendel: DNS answers weighted by server capacity/load share — the same "load share" concept the '517 claims use to split traffic 1.5:1 between two CDNs).

Motivation to combine (KSR v. Teleflex, 550 U.S. 398 (2007)): A person of ordinary skill in 2001 designing a content-delivery traffic director would have faced the known problem of distributing load across multiple content sources while maintaining availability. Each reference supplies a complementary, well-known solution in the same art: Leighton solves where to send a request (network-aware DNS selection), Farber solves how to hand off to a third-party CDN (CNAME), and Brendel solves in what proportion (weighted load sharing). Combining weighted load-share answers (Brendel) with DNS/CDN selection (Leighton + Farber) to let a content provider split traffic among its own servers and two CDN domains is the predictable application of known techniques to achieve a known result — exactly the KSR "combination of familiar elements according to known methods" paradigm. The '517 specification itself describes the multi-CDN load-share example as a matter of configuring "load share" values, which is the Brendel mechanism applied to Farber-style CNAME leaves.

Combination 2 — Rabinovich (6,125,394 and/or 6,052,718) + Choquier (5,774,668) + Kenner (6,003,030)

Claim elements addressed:

• Policy-based selection responsive to a request (Rabinovich 6,125,394: "dynamic on-demand routing" selects among replicas according to policy rules tied to client attributes; Gifford 6,052,718: replica routing policy).
• Failover/overflow to alternate servers (Choquier: policy-controlled failover among replicated transaction servers; Kenner 6,003,030: probing server availability and redirecting when a server is down — mirroring the '517 "monitored server" polling at 30-second intervals).
• Policy hierarchy / decision tree (Rabinovich's rule-based selection and Choquier's policy table map onto the '517 policy decision tree with branch nodes and resource leaves).

Motivation: The '517 claims' "shed fraction" and "tiered failover" behavior is, functionally, the combination of (a) Rabinovich/Gifford policy rules choosing among resources, and (b) Kenner/Choquier availability monitoring triggering re-routing. The prior art uniformly recognized that policy-based routing must react to server state; combining a policy engine with an availability monitor is the standard design response to the known problem of server overload/failure. A PHOSITA would be motivated to add Kenner's polling to Rabinovich's policy router to prevent the policy engine from selecting dead servers — the same reason the '517 specification gives for classifying servers as "monitored."

Combination 3 — Jindal (6,092,178) + Wallis (6,282,569) + Farber (6,185,598)

Claim elements addressed:

• DNS server receiving a hostname query and returning a selected server address (Jindal).
• Global load balancing across geographically distributed sites using DNS and network metrics (Wallis).
• CNAME redirection to a CDN domain (Farber), including the "overflow server is a CNAME to a CDN" limitation (FIG. 9(c): overflow defined by CName pointing to "customer.footprint.net").

Motivation: The '517 overflow-server limitation — "when no answers are selected by the Static or Managed (or Dynamic) Server selection process, return a CNAME to a CDN" — is the obvious marriage of a DNS-based selector (Jindal/Wallis) with the well-known CNAME-delegation technique for routing to a CDN (Farber). The art recognized CNAME delegation as the standard mechanism for a subscriber to "hand off" name service to an outsourced infrastructure, and the '517 specification itself describes the CNAME method as a conventional activation technique. Combining a DNS load balancer with a CNAME-to-CDN fallback would be an obvious solution to the known problem of ensuring availability when a subscriber's origin servers are saturated.

Combination 4 (if any claim limitation is not entitled to the 2001 priority date)

If, in the IPR or district court, a claim is construed to require features (e.g., XML load-feedback status files, GUI policy editing, or the specific ATC administrative/backup architecture) that lack 2001 written-description support, additional post-2001 references become available: Karger et al. (6,430,618, Aug. 2002, consistent-hash distributed caching), Schneider (6,442,549, Aug. 2002, DNS-based dynamic binding), Lumelsky et al. (6,460,082, Oct. 2002, management of service-oriented resources across heterogeneous networks), and Medin (6,370,571, Apr. 2002, CDN assignment). These would be combined with the pre-2001 references above in the same manner: DNS-based selection plus policy-configurable load distribution plus CDN handoff.

D. Why a PHOSITA would have been motivated to combine (Graham factor analysis)

1. Same field of endeavor / analogous art. All primary references are in the field of network traffic management, DNS-based server selection, load balancing, and content delivery — the identical problem domain as the '517 patent. The '517 specification itself cites most of them on its face, which is strong evidence that the applicant (or examiner) recognized them as the closest art addressing the same problems.

2. Complementary, not conflicting, teachings. No reference teaches away from combining policy-configurable load share with DNS-based CDN selection. To the contrary: Brendel explicitly weights DNS answers by capacity; Farber explicitly delegates to CDNs via CNAME; Leighton explicitly selects among replicated servers using network conditions; Rabinovich explicitly conditions selection on policy rules. A system combining all four is the natural aggregation of these compatible mechanisms.

3. Known design need and "obvious to try." By 2001, content providers faced the well-documented need to (i) scale via multiple data centers and CDNs, (ii) control costs by shifting traffic between CDNs, and (iii) maintain availability through failover. The prior art collectively supplied each piece; a PHOSITA with a finite set of identified, predictable solutions (weighted DNS answers, CNAME handoffs, policy rules, availability polling) would have combined them with a reasonable expectation of success. Under KSR, this is the "obvious to try" and "predictable variation" paradigm.

4. Level of ordinary skill. A PHOSITA would be a network engineer or distributed-systems designer familiar with DNS, HTTP, load balancing, and CDN architectures — precisely the audience of the cited references (Akamai, Digital Island, IBM, Lucent, Cisco).

5. Secondary considerations. The applicant's own prosecution record (the terminal disclaimers and continuation filings) and the fact that the specification relies on standard building blocks (DNS, CNAME, XML, HTTP polling) suggest the claimed combination is a conventional aggregation. Notably, the specification's own examples use infrastructure ("footprint.net," "sandpiper.net") whose selection and CNAME-handoff mechanics were taught by the Farber CDN patents. No evidence of long-felt need, industry skepticism, or unexpected results appears on the face of the file.

E. Claim-mapping caveats and litigation context

• Exact claim text: The provided excerpt omits the claims. The mapping above is based on representative limitations drawn from the abstract, specification, and family claims (US 7,860,964 B2, 25 claims). Before finalizing an invalidity position, the verbatim independent claims of the '517 patent must be checked (available from the USPTO PatentCenter record, the Justia page, and the IPR2025-00806 petition).
• Priority-date issue: The availability of post-2001 references (Karger 6,430,618; Schneider 6,442,549; etc.) turns on whether each challenged claim is fully supported by the 2001-09-28 provisional. The PTAB petition and district court pleadings will contest this; the analysis in Combination 4 is contingent on that issue.
• Pending litigation: IPR2025-00806 (instituted) and C.D. Cal. case 2:24-cv-03951 involve this patent; the petitioner's asserted grounds are the most current articulation of the obviousness case and should be reviewed for the specific reference combinations and claim charts. My analysis above independently identifies the strongest combinations: Leighton 6,108,703 + Farber 6,185,598 + Brendel 5,774,660 and Rabinovich 6,125,394 + Choquier 5,774,668 + Kenner 6,003,030.

F. Bottom line

On the available record, the pre-2001 references cited on the face of US 8,645,517 B2 provide a substantial, well-supported basis for a § 103 obviousness challenge to the representative claims. The invention's core — DNS-based selection among a content provider's own servers and one or more CDN domains governed by content-provider policies, weighted load sharing, monitored failover, and CNAME-based overflow to a CDN — is a predictable combination of known DNS load-balancing (Leighton, Brendel, Jindal, Wallis), policy-based replica routing (Rabinovich, Gifford, Choquier), CDN CNAME handoff (Farber), and availability monitoring (Kenner). A person of ordinary skill would have been motivated to combine these references to achieve exactly the configurable, adaptive global traffic-control system the '517 patent claims, and the pending IPR2025-00806 institution is consistent with that conclusion.

Note: Because the claims were not reproduced in the material provided, the element-by-element mappings above should be validated against the exact claim language before use in any submission.