Complete Feature Mapping: Palantir Foundry to Azure¶
The definitive feature-by-feature reference for mapping every Palantir Foundry capability to its Microsoft Azure equivalent.
Audience: Platform architects, migration leads, and technical evaluators Last updated: 2026-04-30
Summary¶
This reference maps 65 Palantir Foundry features across 10 capability domains to their Azure equivalents. Each mapping includes migration complexity, the CSA-in-a-Box evidence path (where the pattern exists in the repository), and notes on gaps or limitations.
| Metric | Count |
|---|---|
| Total features mapped | 65 |
| Full parity (XS–M effort) | 48 |
| Partial parity (L effort) | 12 |
| Known gaps (XL or roadmap) | 5 |
Migration complexity key¶
| Rating | Description | Typical effort |
|---|---|---|
| XS | Drop-in replacement or native Microsoft capability | < 1 day |
| S | Minor configuration or adaptation required | 1–3 days |
| M | Moderate development; requires design decisions | 1–3 weeks |
| L | Significant development; architectural changes | 1–3 months |
| XL | Major initiative; phased delivery | 3+ months |
1. Data layer¶
| # | Foundry feature | Description | Azure equivalent | Complexity | CSA-in-a-Box evidence | Notes |
|---|---|---|---|---|---|---|
| 1 | Datasets | Versioned, transactional file storage for structured/unstructured data | Delta Lake on ADLS Gen2 / OneLake | S | domains/shared/dbt/models/ | Delta provides ACID transactions, versioning, and time travel |
| 2 | Multimodal Data Plane (MMDP) | Open data architecture with Iceberg, virtual catalogs, multi-engine compute | OneLake + Fabric multi-engine (Spark, SQL, KQL, Python) | M | csa_platform/unity_catalog_pattern/ | OneLake is the unified storage layer; Fabric provides multi-engine compute |
| 3 | Virtual Tables | Zero-copy access to external data without ingestion | OneLake shortcuts + Fabric mirroring + Synapse serverless | S | N/A — use Azure native | OneLake shortcuts provide zero-copy access to S3, ADLS, GCS |
| 4 | Data Connection app | Central management of all data source connections | Azure Data Factory (ADF) management hub | S | domains/shared/pipelines/adf/ | ADF provides central linked service and integration runtime management |
| 5 | Magritte connectors | 200+ pre-built connectors (databases, APIs, cloud, SaaS) | ADF 100+ built-in connectors + Logic Apps 600+ connectors | S | docs/ADF_SETUP.md | Combined ADF + Logic Apps exceeds Foundry connector count |
| 6 | Agent worker/proxy | On-premises data access via installed agent software | Self-hosted Integration Runtime (SHIR) | S | docs/SELF_HOSTED_IR.md | SHIR provides same on-prem connectivity with Windows/Linux support |
| 7 | Batch syncs | Scheduled full or incremental data pulls | ADF Copy Activity with schedule triggers | S | domains/shared/pipelines/adf/ | ADF supports full and incremental copy with watermark patterns |
| 8 | CDC syncs | Change data capture from databases | ADF CDC connector / Fabric mirroring / Debezium + Event Hubs | M | docs/patterns/streaming-cdc.md | Fabric mirroring is the simplest path for supported sources |
| 9 | Streaming syncs | Real-time data ingestion | Event Hubs + Stream Analytics / Fabric Real-Time Intelligence | M | docs/guides/event-hubs.md | Event Hubs handles millions of events/second |
| 10 | Media syncs | File, image, and document ingestion | Azure Blob Storage + Azure AI Document Intelligence | S | N/A — use Azure native | AI Document Intelligence extracts structure from documents |
| 11 | Listeners | HTTPS, WebSocket, Email push-based ingestion | Event Grid + Azure Functions (HTTP trigger) + Logic Apps | M | csa_platform/functions/eventProcessing/ | Event Grid is the native Azure event routing service |
| 12 | REST API plugins | Custom connector framework for REST sources | ADF REST connector + custom ADF activities | S | N/A — use Azure native | ADF REST connector supports auth, pagination, and transformation |
| 13 | S3-compatible access | S3 API access to Foundry data (via Iceberg) | ADLS Gen2 REST API + OneLake REST API + Azure Storage S3 proxy | S | N/A — use Azure native | OneLake provides native REST and S3-compatible access |
| 14 | JDBC access | JDBC connectivity to Foundry data | Fabric SQL endpoint / Azure SQL / Synapse serverless SQL | XS | N/A — use Azure native | Standard JDBC/ODBC drivers available for all Azure SQL services |
2. Transform and compute¶
| # | Foundry feature | Description | Azure equivalent | Complexity | CSA-in-a-Box evidence | Notes |
|---|---|---|---|---|---|---|
| 15 | Pipeline Builder | Visual ETL designer with drag-and-drop transforms | ADF visual pipeline designer / Fabric Data Factory | M | domains/shared/pipelines/adf/ | ADF provides visual design; dbt handles transformation logic |
| 16 | Code Repositories | Web IDE with Git, branching, PRs, CI/CD | GitHub + VS Code + GitHub Actions | S | .github/workflows/deploy.yml | Standard Git workflow with richer ecosystem |
| 17 | Python transforms | PySpark with @transform decorators | dbt Python models / Fabric notebooks / Databricks notebooks | M | domains/shared/dbt/models/, domains/shared/notebooks/ | Remove Foundry decorators; use standard PySpark |
| 18 | SQL transforms | Declarative SQL transforms | dbt SQL models | S | domains/shared/dbt/models/ | dbt is a direct replacement with better testing and documentation |
| 19 | Java transforms | Spark Java API transforms | Databricks Spark jobs (Java) | M | N/A — use Databricks native | Standard Spark Java; less common in modern stacks |
| 20 | Incremental computation | Process only changed rows/files since last run | dbt incremental models / ADF incremental copy | S | domains/shared/dbt/models/ | dbt incremental materialization with merge strategy |
| 21 | Streaming (Flink) | Low-latency Flink-based streaming | Spark Structured Streaming / Stream Analytics / Fabric RTI | M | docs/patterns/streaming-cdc.md | Multiple options depending on latency requirements |
| 22 | Single-node engines | DataFusion, Polars, DuckDB for small datasets | Fabric notebooks (Polars, DuckDB) / Azure Functions | S | N/A — use Azure native | Same open-source engines available in Fabric notebooks |
| 23 | Compute Modules | Containerized BYO compute | AKS / Azure Container Apps / Azure Container Instances | M | N/A — use Azure native | Full container orchestration with more flexibility |
| 24 | External Transforms | Pipeline connectivity to external systems | ADF custom activities + Azure Functions | M | csa_platform/functions/ | Azure Functions provide serverless external compute |
| 25 | Data Expectations | Assertion-based data quality checks | dbt tests + Great Expectations + dbt contracts | S | domains/finance/data-products/invoices/contract.yaml | dbt tests are more expressive; contracts enforce at build time |
| 26 | Health Checks | Monitoring views with alerts | Azure Monitor + dbt source freshness + Data Activator | M | N/A — use Azure native | Azure Monitor provides richer alerting (email, PagerDuty, Slack, Teams) |
| 27 | LLM pipeline transforms | Classification, sentiment, entity extraction via LLMs | Azure OpenAI in ADF custom activities / Fabric notebooks | M | csa_platform/ai_integration/enrichment/ | Azure OpenAI provides the same LLM capabilities via API |
3. Ontology and semantic layer¶
| # | Foundry feature | Description | Azure equivalent | Complexity | CSA-in-a-Box evidence | Notes |
|---|---|---|---|---|---|---|
| 28 | Object Types | Schema definitions for real-world entities | Purview glossary terms + dbt models (gold layer) | L | csa_platform/csa_platform/governance/purview/purview_automation.py | Requires mapping each object type to glossary term + dbt model |
| 29 | Link Types | Relationship definitions with cardinality | Foreign keys in dbt + Power BI model relationships | M | domains/shared/dbt/models/ | Standard relational modeling patterns |
| 30 | Properties | Typed attributes with classifications | Column definitions + Purview classifications | M | csa_platform/csa_platform/governance/purview/classifications/ | Four classification taxonomies shipped (PII, PHI, Gov, Financial) |
| 31 | Interfaces | Abstract type definitions enabling polymorphism | dbt abstract/base models + Purview classification hierarchies | L | N/A — use dbt patterns | Polymorphism modeled through dbt macros and base models |
| 32 | Materialization | Datasets merging input data with user edits | dbt incremental models + SCD Type 2 snapshots | M | domains/shared/dbt/models/ | Standard dimensional modeling patterns |
| 33 | Scenarios | What-if branching of the ontology | Git branches + dbt environments + Fabric workspaces | M | N/A — use Git native | Git branching provides isolation; dbt targets switch environments |
| 34 | Ontology SQL | Read-only parameterized queries over objects | Fabric SQL endpoint / Azure SQL views / Power BI DAX | S | N/A — use Azure native | Standard SQL views provide equivalent read-only access |
| 35 | Object Views | UI components for embedding ontology objects | Power BI visuals + Power Apps components + PCF | M | portal/react-webapp/src/pages/ | Multiple embedding options depending on use case |
4. Functions and actions¶
| # | Foundry feature | Description | Azure equivalent | Complexity | CSA-in-a-Box evidence | Notes |
|---|---|---|---|---|---|---|
| 36 | TypeScript functions | Server-side TypeScript with Ontology access | Azure Functions (Node.js) + Fabric user data functions | M | csa_platform/functions/ | Remove Foundry decorators; use standard Azure SDK |
| 37 | Python functions | Server-side Python with Ontology edits | Azure Functions (Python) + Fabric notebooks | M | csa_platform/functions/ | Remove Foundry decorators; use standard Azure SDK |
| 38 | Function-backed columns | Computed properties evaluated at runtime | Power BI DAX measures / Fabric computed columns | M | csa_platform/semantic_model/ | DAX measures provide same runtime computation |
| 39 | Function-backed actions | Ontology-editing functions for action logic | Event-driven Azure Functions + Power Automate | M | csa_platform/data_activator/functions/ | Event Grid triggers Azure Functions for complex logic |
| 40 | Actions | State-changing operations with rules and side effects | Data Activator rules + Event Grid + Power Automate | L | csa_platform/data_activator/rules/ | Data Activator handles rules; Event Grid fans out to compute |
| 41 | External Functions | Call external REST APIs from Foundry | Azure API Management + Azure Functions (HTTP trigger) | S | N/A — use Azure native | APIM provides governance; Functions provide compute |
| 42 | Webhooks | Receive inbound HTTP calls from external systems | Event Grid + Azure Functions (HTTP trigger) + Logic Apps | S | csa_platform/functions/eventProcessing/ | Event Grid is the native inbound event service |
| 43 | Automate | Trigger-based automation (time, data, combined) | Power Automate + Data Activator + Azure Functions (timer) | M | csa_platform/data_activator/ | Power Automate for workflow; Data Activator for data-driven triggers |
5. Applications¶
| # | Foundry feature | Description | Azure equivalent | Complexity | CSA-in-a-Box evidence | Notes |
|---|---|---|---|---|---|---|
| 44 | Workshop | Low-code operational app builder (60+ widgets) | Power Apps canvas apps + Power BI Embedded | L | portal/powerapps/ | Widget-by-widget mapping required; see App Migration |
| 45 | Slate | Custom HTML/CSS/JS application framework | Power Pages + custom React (Azure Static Web Apps) | M | portal/react-webapp/src/ | Standard web development; more flexibility than Slate |
| 46 | Object Explorer | Ontology search, browse, drill-down | Purview Data Catalog + CSA-in-a-Box portal marketplace | M | csa_platform/data_marketplace/ | Purview provides catalog; portal provides data product browse |
| 47 | Contour | Point-and-click dataset analysis tool | Power BI reports with Direct Lake semantic models | M | csa_platform/semantic_model/ | Power BI is the industry standard; Copilot adds NL capability |
| 48 | Quiver | Ontology-aware object analysis with charts | Power BI + Fabric Copilot + Fabric Data Agent | M | csa_platform/semantic_model/ | Power BI + Copilot provides NL querying over semantic models |
| 49 | Insight | Point-and-click analysis for modeled data | Power BI with relationships + Fabric SQL endpoint | M | N/A — use Azure native | Power BI relationships enable same drill-through patterns |
| 50 | Fusion | Spreadsheet with dataset writeback | Excel + Analyze in Excel (Power BI live connection) | XS | N/A — native Microsoft | Zero new work — native Microsoft capability |
| 51 | Notepad | Collaborative rich text with embedded charts | Microsoft Loop + OneNote with embedded Power BI visuals | S | N/A — native Microsoft | Loop is the modern equivalent with real-time collaboration |
| 52 | Code Workbook/Workspaces | Notebooks (JupyterLab, RStudio) | Fabric notebooks + Databricks notebooks | S | domains/shared/notebooks/ | Standard Jupyter/VS Code notebooks with richer ecosystem |
| 53 | Vertex | System graphs, process visualization, simulation | Azure Digital Twins + Power BI custom visuals | L | N/A — use Azure native | ADT for complex system-of-systems; Power BI for simpler cases |
| 54 | Pilot | AI app generator from natural language | Copilot Studio + Power Apps AI Builder + GitHub Copilot | M | N/A — use Azure native | Multiple AI-assisted app building tools available |
| 55 | Marketplace | Product discovery, installation, management | Power Platform Environment + AppSource + managed solutions | M | N/A — use Azure native | AppSource provides similar discovery and distribution |
6. AI and AIP¶
| # | Foundry feature | Description | Azure equivalent | Complexity | CSA-in-a-Box evidence | Notes |
|---|---|---|---|---|---|---|
| 56 | AIP (Language Model Service) | Multi-provider LLM access with governance | Azure OpenAI Service + Azure AI Foundry model catalog | M | csa_platform/ai_integration/ | Azure OpenAI provides GPT-4o, GPT-4.1, o3, o4-mini + open models |
| 57 | AIP Chatbot Studio | Agent builder with retrieval context and tools | Copilot Studio + Microsoft 365 Agents SDK | L | Gap — see CSA-0008 | Copilot Studio provides agent building; CSA Copilot on roadmap |
| 58 | AIP Logic | No-code visual LLM function builder | Power Automate + Azure Functions + Semantic Kernel | M | csa_platform/ai_integration/ | Semantic Kernel provides flexible agent orchestration |
| 59 | AIP Assist | In-platform AI assistant | Microsoft 365 Copilot + GitHub Copilot | XS | N/A — native Microsoft | Available across all Microsoft apps; no migration needed |
| 60 | AIP Evals | LLM evaluation, benchmarking, test generation | Azure AI Studio evaluations + Prompt Flow eval | M | N/A — use Azure native | Prompt Flow provides evaluation pipelines with custom metrics |
| 61 | BYOM | Bring your own model (registered models) | Azure ML model registry + AI Foundry model catalog | S | csa_platform/ai_integration/model_serving/ | Standard MLflow-compatible model registry |
| 62 | Model lifecycle | Training, deployment, serving, monitoring | Azure ML pipelines + MLflow on Databricks | M | csa_platform/ai_integration/model_serving/ | Full MLOps lifecycle with experiment tracking |
7. Developer tools¶
| # | Foundry feature | Description | Azure equivalent | Complexity | CSA-in-a-Box evidence | Notes |
|---|---|---|---|---|---|---|
| 63 | OSDK | Auto-generated TypeScript/Python/Java SDKs | Microsoft Graph API + Fabric REST APIs + Data API Builder | M | docs/tutorials/11-data-api-builder/ | Data API Builder auto-generates REST/GraphQL from databases |
| 64 | Developer Console | SDK generation, OAuth, app sharing | Azure Portal + Entra ID app registrations | S | N/A — use Azure native | Standard Azure identity and app management |
| 65 | REST APIs | Programmatic access to platform resources | Azure REST APIs + Fabric REST APIs + Power BI REST APIs | S | N/A — use Azure native | Comprehensive public REST APIs with OpenAPI specs |
| 66 | VS Code Integration | Palantir VS Code extension for local dev | VS Code + Azure extensions + GitHub Copilot | S | N/A — use VS Code native | Richer extension ecosystem with AI assistance |
| 67 | Continue | AI coding assistant in Code Repositories | GitHub Copilot + GitHub Copilot Workspace | XS | N/A — native GitHub | Industry-leading AI coding assistant |
| 68 | Palantir MCP | Model Context Protocol for builder access | GitHub Copilot + Azure DevOps extensions | M | N/A — emerging capability | MCP adoption growing across Azure and Fabric |
| 69 | Ontology MCP | MCP for consumer access to ontology data | Fabric MCP server + custom MCP implementations | M | N/A — emerging capability | Fabric MCP announced; see Agentic Fabric blog |
| 70 | Global Branching | Coordinated branching across resources | Git feature branches + GitHub Environments + PR workflows | M | .github/workflows/ | Standard Git workflow with environment-based deployment |
8. DevOps and deployment¶
| # | Foundry feature | Description | Azure equivalent | Complexity | CSA-in-a-Box evidence | Notes |
|---|---|---|---|---|---|---|
| 71 | Apollo | Continuous delivery, zero-downtime upgrades | Azure DevOps + GitHub Actions + Bicep IaC | M | deploy/bicep/, .github/workflows/ | Standard CI/CD with deployment slots for zero-downtime |
| 72 | Product packaging | Bundle resources into installable products | Bicep modules + Helm charts + Azure Artifacts | M | deploy/bicep/ | Bicep modules provide composable infrastructure packages |
| 73 | Upgrade Assistant | Platform upgrade management | Azure Advisor + Azure Resource Graph + Dependabot | S | N/A — use Azure native | Azure Advisor provides proactive recommendations |
| 74 | Environment management | Dev, staging, production environments | Azure subscriptions + resource groups + deployment slots | S | deploy/bicep/ | Standard Azure environment isolation patterns |
9. Security and governance¶
| # | Foundry feature | Description | Azure equivalent | Complexity | CSA-in-a-Box evidence | Notes |
|---|---|---|---|---|---|---|
| 75 | Projects | Primary security boundary | Azure resource groups + Purview collections + Fabric workspaces | S | deploy/bicep/ | Standard Azure RBAC boundaries |
| 76 | Organizations | User silos for multi-tenant separation | Entra ID tenants / management groups | S | N/A — use Azure native | Standard Azure multi-tenant patterns |
| 77 | Markings | Mandatory access controls (PII, PHI) | Purview classifications + Microsoft Information Protection | M | csa_platform/csa_platform/governance/purview/classifications/ | Four classification taxonomies shipped |
| 78 | Encryption | In-transit (TLS) and at-rest (AES-256) | Azure encryption (platform-managed or CMK) + TLS 1.3 | XS | N/A — Azure default | Encryption is default on all Azure services |
| 79 | Audit logging | Comprehensive audit trail | Azure Monitor + Log Analytics + tamper-evident audit (CSA-0016) | M | Audit logger module (CSA-0016) | CSA-in-a-Box adds tamper-evident hash-chained audit |
| 80 | Data lineage | End-to-end lineage tracking | Purview lineage + dbt DAG + ADF lineage | M | csa_platform/csa_platform/governance/purview/purview_automation.py | Auto-capture from ADF, Fabric, and dbt |
| 81 | Sensitive data scanning | Automated PII/PHI detection | Purview automated scanning + sensitivity label auto-labeling | S | csa_platform/csa_platform/governance/purview/classifications/ | Purview provides same automated scanning capability |
| 82 | Retention policies | Configurable data retention | Azure Storage lifecycle management + Purview retention | S | N/A — use Azure native | Standard storage lifecycle policies |
| 83 | Row/column security | Fine-grained data access | Fabric RLS + Power BI RLS + dynamic data masking | M | N/A — use Azure native | RLS in Power BI; column masking in Fabric/SQL |
| 84 | SSO/SAML | Identity federation | Entra ID (SAML, OIDC, WS-Fed) | XS | N/A — use Azure native | Standard federation; likely already in place |
| 85 | Checkpoint justifications | Require justification for sensitive data access | Entra Privileged Identity Management (PIM) + access reviews | M | N/A — use Azure native | PIM provides just-in-time access with justification |
10. Monitoring and observability¶
| # | Foundry feature | Description | Azure equivalent | Complexity | CSA-in-a-Box evidence | Notes |
|---|---|---|---|---|---|---|
| 86 | Data Health | Monitoring app for resource health/freshness | Azure Monitor + dbt source freshness + Data Activator | M | N/A — use Azure native | Richer alerting and dashboard capabilities |
| 87 | Metrics | Near-real-time metrics for functions/actions | Azure Monitor metrics + Application Insights | S | N/A — use Azure native | Standard APM with distributed tracing |
| 88 | Workflow Lineage | Execution history and filtering | ADF Monitor + Azure Monitor activity logs | S | N/A — use Azure native | ADF provides visual pipeline monitoring |
| 89 | Trace views | End-to-end tracing of functions/LLM calls | Application Insights distributed tracing + OpenTelemetry | M | docs/patterns/observability-otel.md | OpenTelemetry standard with Azure Monitor backend |
| 90 | Log export | Export logs for custom dashboards | Azure Monitor diagnostic settings + Log Analytics export | S | N/A — use Azure native | Export to storage, Event Hubs, or partner tools |
| 91 | Alerts | Email, PagerDuty, Slack, Foundry notifications | Azure Monitor alerts (email, SMS, Teams, PagerDuty, Slack, webhook) | S | N/A — use Azure native | Broader notification channel support |
11. Interoperability¶
| # | Foundry feature | Description | Azure equivalent | Complexity | CSA-in-a-Box evidence | Notes |
|---|---|---|---|---|---|---|
| 92 | Power BI connector | Connect Power BI to Foundry data | Direct Lake / DirectQuery / Import (native) | XS | csa_platform/semantic_model/ | Power BI connects natively to Azure data sources |
| 93 | Tableau connector | Connect Tableau to Foundry data | Tableau connector to Azure SQL / Fabric SQL endpoint | XS | N/A — use Azure native | Standard JDBC/ODBC connection |
| 94 | Jupyter/RStudio | Notebook integration | Fabric notebooks / Databricks notebooks (native) | XS | domains/shared/notebooks/ | Native notebook support in both Fabric and Databricks |
| 95 | Consumer mode | External-facing apps with limited permissions | Power Pages + Azure AD B2C | M | N/A — use Azure native | Power Pages is purpose-built for external-facing apps |
Gap summary¶
| Gap | Foundry feature | Status | Resolution |
|---|---|---|---|
| CSA Copilot | AIP Agents / Quiver deeper NL | CSA-0008 (roadmap) | Copilot Studio provides partial coverage today |
| Object Explorer UX | Pixel-perfect ontology drill UX | CSA-0129 (partial) | Purview Data Catalog + portal marketplace |
| IL6 coverage | Classified SCI workloads | Out of scope | Recommend Foundry or bespoke Azure Top-Secret tenant |
| Vertex full fidelity | Complex system-of-systems simulation | N/A | Azure Digital Twins for complex cases; Power BI for simple |
| MCP ecosystem | Palantir MCP / Ontology MCP maturity | Emerging | Fabric MCP announced; ecosystem maturing rapidly |
Related resources¶
- Migration Playbook — Phased project plan and worked example
- Ontology Migration — Deep dive on migrating the Ontology
- AI Migration — Deep dive on migrating AIP
- App Migration — Widget-by-widget Workshop mapping
- Benchmarks — Performance comparisons
- TCO Analysis — Cost comparisons
Last updated: 2026-04-30 Maintainers: CSA-in-a-Box core team