{"schema_version":1,"research_id":"eoa_inverse_innovation_exp05_external_evaluation_20260803","source_assessment_id":"layer_decay_and_expiration_management__physics:P5:v0","cell_id":"layer_decay_and_expiration_management__physics","search_queries":["plasma diagnostics data acquisition circular buffer pre-trigger post-trigger waveform event ITER","fusion plasma experiment data acquisition pretrigger buffer synchronized diagnostics disruption data","MDSplus segmented records event data acquisition plasma diagnostics documentation","plasma control system event trigger data acquisition pre-trigger DIII-D","site:iter.org diagnostics data acquisition storage high data rates continuous plasma data CODAC","site:osti.gov DIII-D high speed data acquisition storage limitations diagnostics 2025","fusion experiment continuous data acquisition data volume retention archive diagnostics official","ITER data handling diagnostics data archive pre trigger circular buffer","Wendelstein 7-X continuous data acquisition trigger event pretrigger buffer diagnostic data archive paper","plasma diagnostic event driven data acquisition circular buffer multiple trigger paper fusion","tokamak delayed trigger pre-trigger data acquisition diagnostic synchronized channels","W7-X ArchiveDB chunking finite time boxes data acquisition paper","\"Implementation of high-speed data acquisition at DIII-D\" PDF","\"ITERDB—The Data Archiving System for ITER\" PDF","\"A distributed real-time system for event-driven control\" PDF","\"ArchiveDB—Scientific and technical data archive for Wendelstein 7-X\" PDF","\"High-Performance FPGA Streaming Data Concentrator\" PDF WEST Tokamak","\"A Versatile Board for Event-Driven Data Acquisition\" PDF","site:mdsplus.org \"Long Pulse Extensions\" segmented records","MDSplus segmented records official documentation long pulse","\"A Versatile Board for Event-Driven Data Acquisition\" DOI","site:pubmed.ncbi.nlm.nih.gov \"A Versatile Board for Event-Driven Data Acquisition\"","site:mdpi.com \"A Versatile Board for Event-Driven Data Acquisition\"","\"A Versatile Board for Event-Driven Data Acquisition\" JINST 2024","site:mdpi.com/2079-9292/12/17/3649 published August 2023 Kasprowicz WEST tokamak","site:mdpi.com/1424-8220/24/5/1631 published March 2024 Manduchi"],"sources":[{"source_id":"S1","title":"Implementation of High Speed Data Acquisition at DIII-D","publisher":"International Atomic Energy Agency conference system / General Atomics","url":"https://conferences.iaea.org/event/377/contributions/31637/contribution.pdf","source_class":"PRIMARY_RESEARCH","publication_date":"2024-07-15","accessed_at":"2026-08-03","claims_supported":["DIII-D identifies high-speed collection as critical to many diagnostics and high-resolution analysis.","Nanosecond-capable systems require modification of existing architectures, identifying DIII-D's acquisition organization as a plausible adopter and authorizer.","Reliable acquisition supports plasma operation and future fusion experiments."]},{"source_id":"S2","title":"How to manage 2 petabytes of new data every day","publisher":"ITER Organization","url":"https://www.iter.org/node/20687/how-manage-2-petabytes-new-data-every-day","source_class":"OFFICIAL_ORGANIZATION_DATA","publication_date":"2020-11-30","accessed_at":"2026-08-03","claims_supported":["ITER projects diagnostic systems as the largest raw-data producers and approximately two petabytes of new data per day by about 2035.","ITER expresses needs for storage, secure replication, common data representations and remote analysis.","ITER plans to retain central and backup copies of raw data, which weakens the case for final deletion of selected scientific raw data at ITER even while supporting fast-tier management."]},{"source_id":"S3","title":"NI-DAQmx Data Acquisition Triggering Techniques Using LabVIEW","publisher":"National Instruments","url":"https://www.ni.com/en/support/documentation/supplemental/21/ni-daqmx-data-acquisition-triggering-techniques-using-labview.html","source_class":"OFFICIAL_PRODUCT_DOCUMENTATION","publication_date":"2024-10-08","accessed_at":"2026-08-03","claims_supported":["Commercial DAQ software already implements pretrigger acquisition through a circular buffer that overwrites oldest samples.","A reference trigger returns specified pre- and post-trigger samples.","Capturing only a region of interest is an established way to conserve memory and bandwidth."]},{"source_id":"S4","title":"A distributed real-time system for event-driven control and dynamic data acquisition on a fusion plasma experiment","publisher":"Fusion Engineering and Design / Elsevier","url":"https://www.sciencedirect.com/science/article/abs/pii/S092037960000123X","source_class":"PRIMARY_RESEARCH","publication_date":"2000-08-01","accessed_at":"2026-08-03","claims_supported":["Fusion-specific prior art already demonstrated dynamic pretrigger sampling, event recording and synchronized distributed triggers.","Laboratory configurations and performance were reported, supporting technical feasibility of event-driven synchronized acquisition."]},{"source_id":"S5","title":"A Versatile Board for Event-Driven Data Acquisition","publisher":"Sensors / MDPI","url":"https://www.mdpi.com/1424-8220/24/5/1631","source_class":"PRIMARY_RESEARCH","publication_date":"2024-03-01","accessed_at":"2026-08-03","claims_supported":["An ITER Neutral Beam Test Facility system uses a circular FPGA buffer to retain recent history and transfer pre- and post-trigger samples.","The implementation supports multiple triggers, timestamps, streamed operation and has been used in several facility applications.","The paper identifies overflow and cascades of high-rate events as practical constraints and reports successful implementation of the core acquisition primitives."]},{"source_id":"S6","title":"High-Performance FPGA Streaming Data Concentrator for GEM Electronic Measurement System for WEST Tokamak","publisher":"Electronics / MDPI","url":"https://www.mdpi.com/2079-9292/12/17/3649","source_class":"PRIMARY_RESEARCH","publication_date":"2023-08-31","accessed_at":"2026-08-03","claims_supported":["WEST prior art aggregates, buffers, time-sorts and synchronizes data from multiple plasma-diagnostic channels.","Its FPGA supports trigger control and local selection to reduce transferred volume under long, intense pulses.","The component was laboratory-tested and used during the 2023 WEST campaign, supporting deployability of synchronized event selection."]},{"source_id":"S7","title":"LongPulseExtensions: Segmented Records","publisher":"MDSplus Project","url":"https://www.mdsplus.org/index.php?open=101583195325138468913&page=Documentation%2FUsers+Guide%2FLong+Pulse+Extensions&title=Documentation%3AUsers%3ALongPulseExtensions","source_class":"OFFICIAL_PRODUCT_DOCUMENTATION","publication_date":"undated","accessed_at":"2026-08-03","claims_supported":["MDSplus already stores continuous acquisition data as timestamped segments indexed by start and end time.","Its documented double-buffering and per-segment retrieval provide implementation primitives for immutable time chunks and interval reconstruction.","The documentation does not describe bounded provisional event pins, owner release, reason-coded deletion markers or dependency-gated purge."]},{"source_id":"S8","title":"ArchiveDB—Scientific and technical data archive for Wendelstein 7-X","publisher":"Max Planck Institute for Plasma Physics","url":"https://pure.mpg.de/pubman/item/item_2376563_5/component/file_3212408/Hennig_ArchiveDB.pdf","source_class":"PRIMARY_RESEARCH","publication_date":"2016-11-15","accessed_at":"2026-08-03","claims_supported":["W7-X continuously archives time-stamped scientific and technical streams in finite time chunks indexed by absolute time.","The system reports demanding throughput, reliability and lifetime-preservation requirements and sustained operational use.","W7-X archives triggers and associated timing, demonstrating that synchronized chunk identity and long-term archival are established practice rather than distinctive features of the candidate."]}],"problem_evidence":{"support":"MODERATE","rationale":"High diagnostic rates, finite processing/network/storage capacity and the scientific importance of reliable synchronized data are directly visible at DIII-D, ITER and W7-X. Circular buffers demonstrably overwrite oldest samples, and event bursts can overflow implementation buffers. However, none of the eight sources documents an actual plasma-science incident in which delayed recognition caused an essential precursor interval to be overwritten, nor the prevalence of orphaned pins or ambiguous deletion gaps. ITER's stated plan to preserve raw data also limits generalization from fast-tier pressure to final deletion.","source_ids":["S1","S2","S3","S5","S8"]},"stakeholder_evidence":{"support":"MODERATE","rationale":"DIII-D's diagnostic acquisition organization, ITER CODAC/data-management stakeholders, WEST diagnostic teams and W7-X CoDaC are identifiable adopters or authorizers that publicly express needs for reliable high-rate acquisition, synchronization and scalable storage. No source requests this exact provisional-pin, dependency-gate and gap-marker lifecycle or commits resources to adopting it.","source_ids":["S1","S2","S6","S8"]},"prior_art":{"proximity":"SUBSTANTIAL_COLLISION","closest_analogues":[{"name":"NI-DAQmx pretrigger circular buffering","similarity":"Already preserves a configurable interval before and after a later reference trigger while overwriting the oldest unclaimed samples in fixed memory.","remaining_difference":"No evidence of cross-diagnostic provisional bundles, owner/TTL release, dependency-aware archival disposition or reason-coded markers for removed intervals.","source_ids":["S3"]},{"name":"Fusion-specific distributed dynamic pretrigger acquisition","similarity":"Demonstrated dynamic pretrigger sampling, event recording and accurate synchronization across distributed fusion-experiment nodes in 2000.","remaining_difference":"The abstract does not establish bounded overlapping pins, reviewer confirmation, downstream dependency checks or deletion-state semantics.","source_ids":["S4"]},{"name":"ITER Neutral Beam Test Facility event-driven FPGA recorder","similarity":"Uses a circular buffer, pre/post-event pointers, multiple triggers, timestamps and streamed transfer in an operational fusion-related facility.","remaining_difference":"It captures events but does not show the proposed owned event-pin lifecycle across diagnostics or archival purge governance.","source_ids":["S5"]},{"name":"WEST multichannel FPGA data concentrator","similarity":"Synchronizes, buffers, time-sorts and selectively transmits triggered measurements from many plasma-diagnostic channels under volume pressure.","remaining_difference":"Selection occurs inside one diagnostic system; no published bounded look-back claims, provisional-review workflow, gap markers or post-analysis dependency gate were found.","source_ids":["S6"]},{"name":"MDSplus and W7-X timestamped chunk archives","similarity":"Provide timestamped segment identity, continuous chunk storage, trigger timing, interval retrieval and durable scientific archives.","remaining_difference":"They generally retain/archive data and do not document selective event pin expiry, immediate unpinned eviction or reason-coded tombstones.","source_ids":["S7","S8"]}],"distinctive_claim_remaining":"At identical fast-buffer capacity and on identical time-synchronized plasma traces, an owned, TTL-bounded cross-diagnostic provisional pin followed by reviewer confirmation, immediate eviction of only unpinned chunks, dependency-gated off-tier disposition and reason-coded gap markers will preserve a larger fraction of independently adjudicated delayed-event windows than FIFO, fixed-TTL or trigger-only capture without increasing acquisition-admission failures. The claim is falsified by no completeness improvement, any increase in admission loss, routinely unbounded pinned occupancy, materially worse synchronization, or marker errors that obscure the true disposition of missing intervals.","confidence":"HIGH"},"implementation_evidence":{"support":"STRONG","rationale":"Every central technical primitive—circular pretrigger buffers, pre/post-event extraction, multiple triggers, distributed synchronization, multichannel time sorting, timestamped segments and continuous archival—has been implemented in commercial or fusion-specific systems. The proposed end-to-end policy composition and its behavior during overlapping false triggers remain untested. Workflow integration requires authoritative event adjudication, reliable channel/time metadata and a complete dependency registry; no legal prohibition was found, but local retention obligations and ITER's retain-raw-data policy must override purge.","source_ids":["S3","S4","S5","S6","S7","S8"]},"scores":{"meaningful_impact":{"score":4,"rationale":"Loss of an essential synchronized precursor can irreversibly weaken event interpretation, while uncontrolled pins could interrupt acquisition; the consequence is scientifically meaningful although incidence is unmeasured.","source_ids":["S1","S3","S5"]},"stakeholder_pull":{"score":3,"rationale":"Named fusion organizations express strong adjacent needs for reliable high-rate acquisition and scalable data handling, but no source asks for the exact lifecycle intervention.","source_ids":["S1","S2","S6","S8"]},"incremental_advantage":{"score":2,"rationale":"Core precursor capture and synchronized event selection substantially collide with established systems; superiority of the added governance layer has no comparative evidence.","source_ids":["S3","S4","S5","S6"]},"distinctiveness_plausibility":{"score":2,"rationale":"The remaining distinction is a specific workflow combination—owned expiring cross-diagnostic pins, dependency-gated disposition and explicit gap semantics—not a new acquisition primitive. No exact match appeared in the bounded search, but world novelty was not measured.","source_ids":["S3","S4","S5","S6","S7","S8"]},"technical_implementability":{"score":4,"rationale":"Deployed fusion and commercial systems establish feasibility of buffering, synchronization, triggering, chunking and archival. Correct behavior during trigger cascades and incomplete dependency graphs remains a material integration risk.","source_ids":["S3","S4","S5","S6","S7","S8"]},"adoption_authority_feasibility":{"score":3,"rationale":"DAQ leads, diagnostic owners and data stewards are plausible authorities and an offline simulator avoids production authority initially. Actual decision rights, retention rules and partner commitment were not externally verified.","source_ids":["S1","S2","S8"]},"evidence_readiness":{"score":3,"rationale":"A deterministic replay is technically bounded and comparators are clear, but representative traces, trigger delays and blinded scientific event-window labels require proprietary facility data and expert field participation.","source_ids":["S1","S5","S6"]},"safety_net_benefit":{"score":4,"rationale":"Expiring pins, reason-coded gaps, dependency gates and isolated restore tests could make loss and rollback states safer and more inspectable, though these benefits are still design claims rather than measured outcomes.","source_ids":["S7","S8"]},"scalability":{"score":3,"rationale":"Chunking, FPGA selection and distributed archives scale in deployed settings, but overlapping trigger bursts can expand pinned occupancy faster than the fixed tier and are the intervention's central unresolved scaling failure mode.","source_ids":["S2","S5","S6","S8"]}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"10K_TO_50K","scope":"Four- to eight-week isolated replay using copied traces, implementation of four policies, blinded event-window adjudication, metric reporting and restoration of a small sample.","confidence":"MODERATE","assumptions":["A partner can export 100–300 representative discharges or equivalent traces without a new acquisition campaign.","Existing timestamps and channel mappings are usable.","One DAQ engineer, one analyst and limited plasma/diagnostic panel time are available.","No production hardware or source data are modified."],"source_ids":["S1","S3","S5","S7"]},"initial_deployment_startup":{"band_2026_usd":"50K_TO_250K","scope":"Single-diagnostic non-authoritative shadow prototype, metadata schema, pin-state service, dashboard, synthetic-pressure testing and security/retention review.","confidence":"LOW","assumptions":["Existing DAQ hardware exposes copied or mirrored streams.","Commercial or existing FPGA buffering is reused rather than redesigned.","The prototype covers one diagnostic cluster and one archive interface.","Estimate is resource-equivalent, not based on vendor quotations."],"source_ids":["S3","S5","S6","S7"]},"operational_launch":{"band_2026_usd":"250K_TO_1M","scope":"Production hardening across several synchronized diagnostics, redundant services, integration with timing/archive systems, failover and overload validation, operator training and governed rollout.","confidence":"LOW","assumptions":["No wholesale replacement of acquisition chassis is required.","Facility timing and archive APIs are available.","Launch includes formal failure-injection and rollback testing before authoritative use.","Radiation-hardened custom hardware or facility-wide storage expansion would move cost higher."],"source_ids":["S1","S5","S6","S8"]},"annual_recurring":{"band_2026_usd":"50K_TO_250K","scope":"Part-time engineering and data-steward ownership, policy review, dependency cleanup, archive-restore drills, monitoring and modest metadata/storage overhead.","confidence":"LOW","assumptions":["Waveform archive capacity already exists.","Recurring cost excludes major storage-centre expansion and full-time 24/7 staffing.","One or two restore campaigns and periodic policy reviews occur annually.","Estimate is resource-equivalent without labor-rate or storage-price quotations."],"source_ids":["S2","S7","S8"]}},"verified_pipeline_gates":{"externally_supported_problem":{"status":"UNCERTAIN","reason":"High-rate diagnostic acquisition and capacity strain are verified, but published evidence of delayed-recognition precursor loss, ambiguous eviction gaps or its prevalence was not found.","source_ids":["S1","S2","S3","S5","S8"]},"externally_credible_adopter_or_authorizer":{"status":"YES","reason":"DIII-D acquisition teams, ITER CODAC/data-management stakeholders, WEST diagnostic teams and W7-X CoDaC are identifiable organizations with relevant operational authority and expressed adjacent needs.","source_ids":["S1","S2","S6","S8"]},"distinct_testable_incremental_claim":{"status":"YES","reason":"The remaining claim compares adjudicated event-window completeness and admission continuity at equal capacity against FIFO, fixed-TTL and trigger-only policies, with explicit failure conditions.","source_ids":["S3","S4","S5","S6"]},"bounded_next_evidence_step":{"status":"YES","reason":"A copied-trace, isolated deterministic replay can bound inputs, capacity, comparators, labels, metrics and rollback without touching production acquisition.","source_ids":["S1","S3","S5","S7"]},"no_unresolved_safety_or_authority_stop":{"status":"YES","reason":"The first step is read-only with respect to authoritative data and production controls. Production deployment remains prohibited until acquisition owners and data stewards approve retention, overload and rollback behavior.","source_ids":["S1","S2","S8"]},"credible_cost_scope_and_range":{"status":"UNCERTAIN","reason":"Scopes and resource-equivalent bands are bounded, but no facility architecture inventory, labor rates, storage prices or vendor quotations were available; startup and recurring estimates are low-confidence.","source_ids":["S1","S2","S5","S6","S8"]}},"next_evidence_step":"With one fusion-facility partner, export an immutable, non-authoritative set of 100–300 discharges or equivalent continuous traces containing ordinary operation, calibration intervals, independently confirmed events, deliberately delayed triggers, overlapping/false triggers, timing faults and buffer-pressure episodes. Before retention outcomes are revealed, a plasma physicist plus diagnostic owners must specify each event's required channels and pre/post interval. Replay identical samples at identical simulated capacity through FIFO, fixed-TTL, trigger-only and event-pinned policies; include a deeper-FIFO sensitivity analysis but do not treat extra capacity as an equal-resource comparator. Measure complete adjudicated windows, missing required chunks, admission failures, peak and duration of pinned occupancy, false-pin release latency, cross-channel synchronization error, marker/reason accuracy and restore success including calibration linkage. Falsify the intervention if it does not improve complete-window retention over the best equal-capacity comparator, causes any material increase in admission loss, permits pinned occupancy to remain unbounded, evicts a known required window, produces incomplete/mis-synchronized confirmed bundles, or yields inaccurate gap markers. Do not connect to live acquisition or change source files.","blocking_evidence":["No direct external evidence quantifies delayed-trigger precursor overwrite or ambiguous-gap incidence in a plasma experiment.","No named facility has endorsed or funded the exact bounded-pin lifecycle.","No equal-capacity replay result demonstrates incremental benefit over existing pretrigger capture.","No field-derived distribution of trigger delays, overlapping-trigger bursts or required cross-diagnostic window sizes is available.","Dependency-registry completeness and post-analysis purge authority have not been verified at a facility.","No architecture-specific labor, hardware, storage or integration quotations support the deployment cost bands.","ITER's stated intent to retain raw data may eliminate final-purge benefits there, leaving only fast-tier lifecycle value."],"research_disposition":"PARTNERED_RESEARCH_PROGRAM","world_novelty_boundary":"The eight-source search establishes substantial prior art for circular pretrigger buffers, fusion event recording, synchronized multichannel selection, timestamped chunks and continuous archival. It did not locate the exact combined governance workflow, but absence from this bounded search is not evidence of world novelty. Patentability, freedom to operate, market size, prevalence and realized impact remain unmeasured.","arm":"COMPLETE_PROPOSAL_PORTFOLIO","candidate_version":0,"controller_recommendation":{"action":"STOP_EMPIRICAL_RESEARCH_NEEDED","repairable":false,"material_progress_observed":true,"progress_targets":["Obtain a representative copied trace and blind expert labels for required event windows and channels.","Quantify baseline prevalence of precursor loss, ambiguous gaps, admission failures and unmanaged retained bundles.","Demonstrate higher complete-window retention than the best equal-capacity comparator without materially increasing admission loss.","Bound peak pinned occupancy and false-pin release time under realistic overlapping-trigger bursts.","Validate synchronization, reason-coded markers and end-to-end archive restoration against calibration references.","Verify facility-specific retention authority, dependency ownership and whether raw-data purge is permissible.","Replace resource-equivalent deployment bands with an architecture inventory, staff estimates and vendor/storage quotations."],"reason":"Web evidence verifies the technical primitives, relevant institutions and capacity pressure, while also showing substantial collision with established pretrigger and synchronized-event acquisition. The remaining incremental claim can only be resolved with proprietary traces, blinded scientific adjudication and live workflow/authority information; bounded web search cannot establish its effect or prevalence."},"proposal_index":5}