1. 4SO
  2. Products
  3. 4SO Network

Network source of truth

4SO Network

Network data with the provenance and confidence of every record.

4SO Network turns scattered network information into a machine-readable model: inventory, interfaces, L2/L3, routing, topology and history are kept with clear provenance, and inference never silently becomes fact.

  • Inventory
  • Topology
  • L2 / L3
  • Provenance
Conceptual 4SO Network scene; the globe and routes are synthetic.
Who it is for
Network, operations and architecture teams whose network knowledge is scattered across inventories, configuration files, diagrams and people.
What it does
Turns inventory, interfaces, VLAN and VRF, addressing, routing (static, BGP, OSPF) and topology into one queryable model, keeping the source, time and confidence of every record.
What sets it apart
A relationship that is only inferred stays a candidate until another source corroborates it; it never silently becomes fact.
CORRELATION · CANDIDATE → FACTFrom scattered evidence to a relation you can rely on
01

Snapshots from config, inventory and diagrams are recorded with source identity and time.

01Problem

A network usually has more than one “truth”.

Inventory, diagrams, configuration, routing state and operator knowledge can disagree with one another. A good single source of truth does not hide that disagreement; it keeps the provenance and confidence of each piece of data attached to the data itself.

  1. 01

    Data is scattered across many sources

    Devices, interfaces, addressing, VLAN/VRF, routing and topology usually live in separate formats.

  2. 02

    A diagram is not always live reality

    A relationship derived only from a diagram or naming must stay separate from an observed fact.

  3. 03

    Inference must be labeled

    An inferred link candidate or site/role is not promoted to a definitive authority until it is corroborated.

  4. 04

    History matters for incidents

    Knowing where data came from, and when, is essential for change and incident analysis.

02Capabilities

Network data is broken down into analyzable domains.

The goal is to turn config dumps into an operational model that supports queries, correlation and forensics.

  1. 01

    Device Inventory

    A structured model for devices and operational metadata, independent of the source format.

  2. 02

    Interface Model

    Interfaces, descriptions, bundles and their relationships to other domains.

  3. 03

    VLAN and port-channel

    An L2 model for segments, membership and aggregation.

  4. 04

    IP and VRF

    Addressing and routing context in a separate, queryable structure.

  5. 05

    Routing Facts

    Static, BGP and OSPF data for path and policy analysis.

  6. 06

    Topology relationships

    A node/edge model carrying the source and confidence of every relationship.

  7. 07

    Service Dependencies

    Network service dependencies such as DNS, time, AAA and logging in a common model.

  8. 08

    Provenance and history

    Timestamps, sources and change history for analysis and forensics.

4SO

In this product, authority is not a simple boolean. Every record must keep its own source, timestamp and confidence so that correlation and automation can tell fact from inference.

03Deployment model

The source matters; so does the confidence.

Instead of a flat table that treats everything as definitive truth, data is classified by its origin and extraction method.

DomainExample recordsAuthority / ConfidenceUse
InventoryDevice identity, platform/role metadata, source mappingFact within the scope of the source snapshotInventory, ownership and normalization
Interfaces / L2Interface, description, VLAN, port-channelConfiguration-derived + source timestampLink-layer analysis and aggregation
L3 / RoutingIP/subnet, VRF, static routes, BGP, OSPFObserved/config-derived recordPath, routing context and policy analysis
TopologyNode, edge, device matching, link candidateFact / Candidate / Derived with confidenceCorrelation without turning inference into truth
DependenciesDNS, time, AAA, logging relationshipsSource-owned relation or bounded inferenceService dependency analysis
HistorySnapshot identity, source history, change contextTimestamp + provenance preservedForensics and change comparison
04Lifecycle

Network data has a quality lifecycle too.

Until a snapshot has been parsed, classified and correlated, it is only raw input; a derived result does not take the place of a fact until it has been reviewed.

  1. 01AcquisitionAcquire
  2. 02ParsingParse
  3. 03NormalizationNormalize
  4. 04CorrelationCorrelate
  5. 05ClassificationClassify
  6. 06ReviewReview
  7. 07Derived viewProject
  8. 08History comparisonCompare
05Profiles

Each network domain answers a different question.

Separating domains keeps analysis from depending on one large, ambiguous CSV.

  • Inventory

    Device identity and metadata.

    • Device model
    • Platform / role metadata
    • Source mapping
  • Interface

    Interface and aggregation detail.

    • Interfaces
    • Descriptions
    • Port-channel relationships
  • L2 / L3

    Segment and routing context.

    • VLAN
    • IP / subnet
    • VRF
  • Routing

    Routing data for path analysis.

    • Static routes
    • BGP data
    • OSPF data
  • Topology

    Nodes and edges with explicit authority.

    • Topology sources
    • Device matching
    • Link candidates
  • Forensics

    Change tracking and confidence.

    • Source history
    • Last-change context
    • Provenance-aware comparison
Technical reference

Architecture, workflows and operations in detail

For engineers who want to know how the product works underneath: components, boundaries, failure behavior and the operations it runs.

06Architecture

The pipeline must preserve provenance, not erase it.

Sources are ingested separately, parsers normalize them, and the single source of truth keeps records with their provenance and confidence. Analytical views are only projections.

  1. INPUTS
    Inventory · Topology · Configuration · HistoryEach source has its own identity and timestamp.
  2. NORMALIZATION
    Parser & NormalizerConverts different formats into uniform domain records.
  3. SOT
    Structured network source of truthFact + provenance + confidence; inference is never automatically turned into truth.
  4. INTELLIGENCE
    Derived Views / Analytics / ForensicsCorrelation and analysis over source-owned records without creating a second authority.
TECH

Inference Guardrails. 4SO Network keeps an explicit distinction between what a source actually stated, what has been correlated across several sources, and what has only been inferred.

Evidence typeClassificationUsage rule
Configuration snapshotFact within the scope of that snapshotInterface, addressing, VLAN/VRF and routing facts are valid only with their own source and timestamp.
Topology/diagram sourceSource-scoped factA relationship shown in a diagram does not prove the live state of the network.
Naming / Description correlationCandidateLink/role/site inference is not promoted to definitive authority before corroboration.
Multi-source corroborationConfidence-bearing relationSupporting evidence and sources stay attached to the relation; confidence is never separated from provenance.
Derived viewProjectionSummaries, path views or analytics are built from source-owned data and do not create a second SoT.
Historical comparisonChange contextThe difference between two snapshots shows change; it does not claim the cause of an incident without additional evidence.
07Workflows and failure

The core job: turn network evidence into an analysable model without losing provenance.

Rather than treating everything as truth, the pipeline keeps source-owned facts, correlation and inference in separate stages.

  1. 01AcquireSOURCE SNAPSHOT
  2. 02ParseSTRUCTURED RECORDS
  3. 03NormalizeDOMAIN MODEL
  4. 04CorrelateMULTI-SOURCE
  5. 05ClassifyFACT / CANDIDATE
  6. 06ReviewCORROBORATE
  7. 07ProjectDERIVED VIEW
EVIDENCE CLASSIFICATION

Fact / Candidate / Derived

Every relation is stored with its own evidence type.

SOURCE SNAPSHOT
↓
EVIDENCE
CLASS
↓
FACT
CANDIDATE
DERIVED
A naming hint without corroboration is not a fact.
MULTI-SOURCE CORRELATION

Several sources, one explainable relation

Correlation does not remove source ownership.

CONFIG
INVENTORY
DIAGRAM
↓
CORRELATE + ATTACH PROVENANCE
↓
EXPLAINABLE RELATION
source identity · timestamp · confidence · supporting evidence
Conflicts between sources are preserved and do not collapse into a fabricated truth.
CORROBORATION LOOP

Candidate to fact only with evidence

Confidence and provenance remain throughout the review loop.

HINT
→
CANDIDATE
↓
REVIEW
↑
EVIDENCE
←
FACT /
UNRESOLVED
confidence changes only with traceable evidence
Absence of evidence = unresolved, not automatic confirmation or rejection.
  1. Flow 01

    Ingest Snapshot

    Source identity is preserved before parsing.

    1. Source
    2. Hash / Time
    3. Acquire
    4. Parse
    5. Persist Raw Context
    6. Record Provenance

    Failure and recoveryA parse failure does not overwrite the previous source.

  2. Flow 02

    Normalize Domains

    Different formats are converted into a canonical model.

    1. Device
    2. Interface
    3. L2
    4. L3
    5. Routing
    6. Dependency

    Failure and recoveryUnknown fields are not fabricated; missing stays missing.

  3. Flow 03

    Correlate Sources

    Relations are built without removing provenance.

    1. Select Keys
    2. Match
    3. Compare Evidence
    4. Create Relation
    5. Attach Sources
    6. Confidence

    Failure and recoveryConflicts between sources are not turned into a single fabricated truth.

  4. Flow 04

    Infer → Corroborate

    Inference is not fact until independent evidence exists.

    1. Hint
    2. Candidate
    3. Confidence
    4. Review
    5. Corroborate
    6. Fact / Unresolved

    Failure and recoveryNo evidence = unresolved; neither a fabricated confirmation nor a fabricated rejection.

  5. Flow 05

    Historical Compare

    A change view is a projection, not a new authority.

    1. Snapshot A
    2. Snapshot B
    3. Diff
    4. Context
    5. Derived View
    6. Safe Export

    Failure and recoveryA snapshot difference does not claim the cause of an incident without additional evidence.

Additional analytical flows

Path analysis and conflict resolution are carried out without turning a projection into an authority.

  1. Flow 06

    Path / Dependency Analysis

    A probable path is computed from structured evidence, and the source of every hop is preserved.

    1. Select Endpoints
    2. Load L2 / L3
    3. Load Routing
    4. Correlate Dependencies
    5. Compute Candidate Path
    6. Attach Evidence
    7. Derived View

    Failure and recoveryIf link or routing evidence is incomplete, confidence is lowered; the system does not fabricate the missing path.

  2. Flow 07

    Source Conflict Resolution

    Source conflicts are kept explicit before any promotion to fact.

    1. Detect Conflict
    2. Preserve Values
    3. Compare Source / Time
    4. Authority Hint
    5. Review
    6. Corroborate
    7. Reclassify

    Failure and recoveryIf there is not enough independent evidence, the conflict stays unresolved and no source is artificially declared the winner.

Action / FlowAdmission / PreconditionsExecution / LockSuccess CriterionFailure / Recovery
Ingest SnapshotSource identity/hash/timestamp must be knownThe parser materializes only that source's dataRecords are traceable back to their snapshot/sourceA parse failure does not overwrite the original source and does not record partial success
NormalizeSchema/domain parser for Device/Interface/L2/L3/RoutingDifferent formats are converted into the canonical domain modelIdentity and source mapping are preserved in the recordUnknown fields are not fabricated; missing data stays missing
Correlate SourcesMatch keys and provenance available on both sidesCorrelation builds the relation but does not remove source ownershipEvery relation can show its supporting evidence/sourcesConflicts between sources are not hidden and are not automatically turned into a single truth
Infer CandidateNaming/description/topology hints used only for bounded inferenceThe candidate is recorded with its confidence and inference methodCandidates are distinguishable from facts and reviewableInference does not enter definitive authority without corroboration
Review / CorroborateIndependent evidence or operator confirmation requiredThe candidate is evaluated against additional evidenceClassification/confidence are preserved with a traceNo evidence = unresolved; no fabricated rejection/confirmation
Compare / Export Safe ViewSpecific snapshots and a sensitivity policy selectedDerived view/history/report is built from existing recordsReproducible projection without creating a second SoTPublic export removes real IP/ASN/device/site/topology data; root cause is not claimed without evidence
FLOW

Core rule: Projection ≠ Authority. Summaries, path views, analytics or forensics may be built from several sources, but the original fact stays bound to its own source and timestamp.

08Operations catalog

The operations catalog is data engineering for the network.

  • Ingest SnapshotIngest a new source with a defined identity and timestamp.
  • NormalizeConvert devices, interfaces and network constructs into uniform data structures.
  • CorrelateLink records from multiple sources while preserving provenance.
  • Classify ConfidenceSeparate fact, candidate and derived information.
  • Derive ViewsBuild summaries and analytical views without creating a new authority.
  • Compare HistoryCompare snapshots for change and incident analysis.
  • Review InferenceConfirm or reject proposed items with additional evidence.
  • Export Safe ViewDeliver a dataset or report matched to the domain and sensitivity of the information.

Turn your network into data you know how far to trust.

4SO Network is built for decisions on network data, with Provenance and Confidence as part of the model itself.