Industrial IoT / SCADA · API / SDK

Cognite Data Fusion APIs

Cognite

Build applications over contextualized industrial data and assets.

Official documentation Download development spec

COGNITE DATA FUSION / DEVELOPER FIELD GUIDE

Choose an interface. Define a verifiable result.

Company or project context, ten technical entry points and a practical analysis of every reference.

2026-10-08 · Integration not tested

Company / project overview

Original summary of official company or project statements; use the source for the full original page.

Cognite Data Fusion · About
Company	Cognite develops industrial data and AI software for teams that need to discover, connect and use operational information.
Platform	CDF combines project-scoped APIs, contextualized data models and industrial analytics integrations.

Complete official About page ↗

About reference — TA analysis
Audience	Buyers, partners and product owners shortlisting an AI provider.
Our assessment	Use positioning and company history for initial fit. Obtain project-specific deployment, support and commercial terms separately.

Official sources reviewed. No provider API or inference was run; results are illustrative internal contracts.

Top 10 technical entry points

Ten documented technical entry points selected for implementation decisions, not ten independent SDKs or an official ranking.

Choose by the work you need to complete
Interface	Purpose and audience
01 · REST resource accessAPI / SDK / developer guide	Read and update project-scoped industrial resources.Integration developers
02 · Python SDKAPI / SDK / developer guide	Use typed clients for CDF resources.Python data engineers
03 · JavaScript/TypeScript SDKAPI / SDK / developer guide	Integrate CDF from typed JavaScript clients.Web/backend developers
04 · Identity and accessAPI / SDK / developer guide	Map IdP identities to CDF group capabilities.Identity administrators
05 · Cursor paginationAPI / SDK / developer guide	Retrieve complete paginated resource sets.Bulk retrieval developers
06 · Time-series aggregationAPI / SDK / developer guide	Compute aggregates/interpolation with explicit granularity.Process analytics engineers
07 · Unit-aware dataAPI / SDK / developer guide	Standardize units and request supported conversions.Measurement integrators
08 · Entity matchingAPI / SDK / developer guide	Suggest links between source and target entities.Contextualization teams
09 · Data Modeling ServiceAPI / SDK / developer guide	Model equipment relationships with spaces, containers, views and instances.Industrial data architects
10 · API calendar versionsAPI / SDK / developer guide	Pin stable API behavior through Cdf-Version.Platform maintainers

Reference analysis

Capabilities summarize official documentation. Constraints and proposed tests are our engineering assessment.

API / SDK / developer guide

REST resource access

REST resource access · TA
Target audience	Integration developers
Documented capability	Read and update project-scoped industrial resources.
Input → output	Project/resource request → data/status/cursor.
Execution environment	Correct cluster, project and authenticated identity.
Our decision constraint	Object IDs are not globally unique across projects.
Our suggested validation	Keep project identity with every resource reference.

Official reference · REST resource access ↗

API / SDK / developer guide

Python SDK

Python SDK · TA
Target audience	Python data engineers
Documented capability	Use typed clients for CDF resources.
Input → output	SDK calls → resource objects and errors.
Execution environment	Pinned cognite-sdk release and compatible Python runtime.
Our decision constraint	SDK convenience does not replace project capability grants.
Our suggested validation	Verify pagination, exception mapping and client configuration.

Official reference · Python SDK ↗

API / SDK / developer guide

JavaScript/TypeScript SDK

JavaScript/TypeScript SDK · TA
Target audience	Web/backend developers
Documented capability	Integrate CDF from typed JavaScript clients.
Input → output	OAuth-backed client call → resource lists.
Execution environment	Pinned @cognite/sdk and appropriate browser/service auth flow.
Our decision constraint	Backend client secrets must not enter browser bundles.
Our suggested validation	Check installed SDK migration guide before using older tutorial methods.

Official reference · JavaScript/TypeScript SDK ↗

API / SDK / developer guide

Identity and access

Identity and access · TA
Target audience	Identity administrators
Documented capability	Map IdP identities to CDF group capabilities.
Input → output	OAuth/OIDC identity → scoped resource access.
Execution environment	Configured IdP, approved app and project group membership.
Our decision constraint	Successful login does not imply access to every resource.
Our suggested validation	Validate intended read/write scope and denied neighboring resources.

Official reference · Identity and access ↗

API / SDK / developer guide

Cursor pagination

Cursor pagination · TA
Target audience	Bulk retrieval developers
Documented capability	Retrieve complete paginated resource sets.
Input → output	Stable filters and cursor → next page/nextCursor.
Execution environment	Resource-specific limit and authorized project.
Our decision constraint	A short page can still carry nextCursor; do not stop on page length.
Our suggested validation	Continue until cursor exhaustion with unchanged request parameters.

Official reference · Cursor pagination ↗

API / SDK / developer guide

Time-series aggregation

Time-series aggregation · TA
Target audience	Process analytics engineers
Documented capability	Compute aggregates/interpolation with explicit granularity.
Input → output	Series, interval and aggregate → bucket results.
Execution environment	Time-series read access and chosen step/continuous interpretation.
Our decision constraint	Average is time-weighted, not arithmetic mean; interpolation is not a raw reading.
Our suggested validation	Test sparse intervals and label interpolated/aggregated values.

Official reference · Time-series aggregation ↗

API / SDK / developer guide

Unit-aware data

Unit-aware data · TA
Target audience	Measurement integrators
Documented capability	Standardize units and request supported conversions.
Input → output	Source unit and target unit → converted values/unit metadata.
Execution environment	Recognized unit identifiers and compatible measured quantity.
Our decision constraint	A free-text label alone is not a validated physical conversion.
Our suggested validation	Check dimensions, scale and offset with known reference values.

Official reference · Unit-aware data ↗

API / SDK / developer guide

Entity matching

Entity matching · TA
Target audience	Contextualization teams
Documented capability	Suggest links between source and target entities.
Input → output	Entity attributes and model → ranked candidates/scores.
Execution environment	Configured matching model and relevant data permissions.
Our decision constraint	Scores are not universally calibrated probabilities or verified identity.
Our suggested validation	Review ambiguous candidates and validate links against labeled samples.

Official reference · Entity matching ↗

API / SDK / developer guide

Data Modeling Service

Data Modeling Service · TA
Target audience	Industrial data architects
Documented capability	Model equipment relationships with spaces, containers, views and instances.
Input → output	Schema and instance definitions → queryable industrial graph.
Execution environment	Project data-model permissions and explicit model/view versions.
Our decision constraint	Legacy asset-centric assumptions may not match the chosen data model.
Our suggested validation	Verify property types, relationships and selected view versions.

Official reference · Data Modeling Service ↗

API / SDK / developer guide

API calendar versions

API calendar versions · TA
Target audience	Platform maintainers
Documented capability	Pin stable API behavior through Cdf-Version.
Input → output	v1 path and calendar header → versioned endpoint behavior.
Execution environment	Supported stable calendar version and compatible SDK.
Our decision constraint	Playground is deprecated; beta is not recommended for production.
Our suggested validation	Review deprecations in addition to the version header.

Official reference · API calendar versions ↗

Official documentation

Official documentation · Supporting reference analysis
TA	Developers and technical product owners
Our validation advice	Pin the release and follow the target-specific support requirements before implementation.

Official reference · Official documentation ↗

Deprecated and retired features

Deprecated and retired features · Supporting reference analysis
TA	Maintainers of existing CDF integrations
Our validation advice	Do not start new Vision API or totalVariation dependencies. Review retired CDF CLI and Cognite Functions runtime requirements before reuse.

Official reference · Deprecated and retired features ↗

04 / INPUT · EXPECTED OUTPUT · ACCEPTANCE

Expected results before implementation

Illustrative industrial asset/time-series retrieval contract only. No CDF project, model or API was executed.

Illustrative contract · no API call or model execution · not measured provider output

Complete contextual retrieval

Expected behavior — illustrative, not executed
Input	Authorized asset query with related time series.
Expected result	Project-scoped records with source identifiers and units.
Acceptance	Verify relationship correctness, cursor exhaustion and source provenance.

Sparse interval

Expected behavior — illustrative, not executed
Input	Valid aggregate query with no qualifying points and no gap fill.
Expected result	No aggregate bucket; no invented zero reading.
Acceptance	Display missing data explicitly and separate interpolated values.

Short page with cursor

Expected behavior — illustrative, not executed
Input	Page has fewer records than requested but contains nextCursor.
Expected result	retrieval_complete=false until subsequent pages finish.
Acceptance	Do not truncate the dataset based on page length.

Insufficient capability

Expected behavior — illustrative, not executed
Input	Valid identity lacks permission for requested project resource.
Expected result	Access error with resource scope; no empty-success substitution.
Acceptance	Request appropriate capability through the owner; do not widen scopes automatically.

Map the expected result to your application

Map your actual runtime response into this proposed application contract. All values are invented examples; no credentials or live execution are included.

Expected output JSON — illustrative, not measured · CogniteDataFusion-expected-results.json

{
  "schema_version": "1.0.0",
  "example": true,
  "provider": "Cognite Data Fusion",
  "status": "ok",
  "input_id": "sample-001",
  "data": {
    "project": "demo-project",
    "external_id": "demo-pump-1",
    "retrieval_complete": true,
    "records": [
      {
        "temperature_c": 23.5,
        "value_kind": "raw-example"
      }
    ],
    "next_cursor": null
  },
  "error": null,
  "verification": "illustrative-only;not-provider-response"
}

Download specification, example and acceptance plan

Engineering handoff — our proposed contract and gates
State semantics	Valid empty results differ from errors. Pending work must not appear successful before completion.
Evidence	Record service and SDK versions, request identity, configuration and response mapping. No execution or performance was measured here.
Acceptance proposal	Agree numerical tolerances and workload targets before testing on representative authorized samples.

Development assessment

Industrial asset evidence browser

Development concept · integration not tested

Provider capabilities above are based on official documentation or repositories. The proposed product, inputs, deliverable and acceptance criteria below are our development assessment.

Integration pilot

Validate runtime, schemas or hardware together before estimating a deployable product.

Proposed inputs: A test device or simulator, a tag map and documented units.
Proposed deliverable: Timestamped device values with freshness, unit and connection state.
Acceptance criterion: Verify scaling and timestamps against the source; a disconnected device must not appear healthy.
Dependencies: A gateway or service account, an equipment schema and access to a test environment.

Development sequence

Confirm access to Cognite Data Fusion APIs, license and the exact supported version.
Prepare the sample above and implement one documented operation for “Industrial asset evidence browser”.
Normalize the result with source, time and explicit error state; keep the provider response for review.
Run the acceptance criterion before estimating rollout effort or committing a customer deliverable.

How to validate demand

Record product views, documentation clicks, specification downloads and contextual hub clicks. These are event counts, not unique people or completed integrations.

This feasibility assessment uses implementation conditions. No traffic-based rank or delivery-time promise is assigned.

Cognite Data Fusion APIs

Choose an interface. Define a verifiable result.

Company / project overview

Top 10 technical entry points

Reference analysis

Official documentation

Deprecated and retired features

Expected results before implementation

Complete contextual retrieval

Sparse interval

Short page with cursor

Insufficient capability

Map the expected result to your application

Industrial asset evidence browser

Development sequence

How to validate demand

Related products