Data Engineering · Editable reference architecture
Data lakehouse architecture
An editable data lakehouse architecture reference showing Sources → Object storage → Table transaction layer and downstream dependencies. Open files still need transactional table metadata.
Open & Edit in DrawArc (Free, No Login Required)Your changes stay in a separate draft on this browser. Download an editable project to keep a portable copy.

6 components · 5 connections · Open the editor to zoom, move nodes and edit source.
Architectural overview
Data lakehouse architecture addresses the responsibilities and boundaries shown in this reference rather than prescribing an entire production deployment. The primary flow starts at Sources and reaches Analytics through Object storage, Table transaction layer, SQL engine. Use this view to explain which component initiates work, which component owns a decision, and where a dependency can affect the caller. Open files still need transactional table metadata.
The diagram contains 6 distinct components and 5 directed relationships. A separate path connects Table transaction layer → ML workloads. These arrows describe logical interactions; they do not establish authentication, network access, atomic transactions or delivery guarantees. Label the actual protocol, payload contract and owner when adapting the drawing to your system. Keep synchronous and asynchronous boundaries explicit so that a reviewer can identify where a request ends and a later operation begins.
Before implementing this architecture, define expected load, data classification, latency budgets and recovery objectives. Give SQL engine a bounded dependency timeout and a documented failure response. For Object storage, Table transaction layer, specify ownership, retention, access control and a tested recovery procedure. Validate the decision above against your workload instead of treating the presence of a component as proof of reliability. Open the editable canvas, replace generic labels with real deployment units, and document the decisions that differ from this reference.
The real-world problem
Design or explain data lakehouse architecture with explicit ownership and dependency paths. Open files still need transactional table metadata.
Data flow
Connections from this diagram. Branches and control signals are not a single chronological sequence.
- Sources → Object storage. Define the interaction contract and trace the resulting response or state change.
- Object storage → Table transaction layer. Define the interaction contract and trace the resulting response or state change.
- Table transaction layer → SQL engine. Define the interaction contract and trace the resulting response or state change.
- SQL engine → Analytics. Define the interaction contract and trace the resulting response or state change.
- Table transaction layer → ML workloads. Define the interaction contract and trace the resulting response or state change.
Security boundaries
- Authenticate callers at the boundary before Object storage accepts protected work; authorize the specific operation rather than trusting network placement.
- Use separate service identities and least-privilege permissions for Object storage, Table transaction layer, SQL engine. Encrypt sensitive traffic, protect credentials and avoid placing confidential payloads in logs.
- Restrict and audit access to Object storage, Table transaction layer; define retention and deletion requirements for the actual data.
Failure handling
SQL engine is unavailable or slow
Response: Apply bounded timeouts and controlled backoff only where retrying is safe. Return an explicit failure or durable acceptance; do not report success before required work is committed.
Verify: Interrupt SQL engine in an isolated test and measure errors, recovery and any repeated side effects.
Object storage loses or returns stale state
Response: Use workload-specific durability, reconciliation and tested recovery. Specify how stale reads or a failed write affect callers.
Verify: Exercise the failure path with representative records and verify both the visible response and the resulting state.
An invalid or incompatible input reaches Object storage
Response: Validate the input contract before effects are committed. Preserve rejected input identifiers without exposing secrets, and document a safe replay or correction procedure.
Verify: Send malformed and older-version inputs through Sources and verify rejection, observability and unchanged valid state.
Trade-offs and alternatives
Open files still need transactional table metadata.
- Benefit
- Makes the ownership and interaction boundaries visible for a focused design review.
- Cost
- The 6 illustrated components still require contracts, monitoring and operational ownership; extra boundaries can add latency and failure modes.
- Alternative
- Use a simpler deployment when it meets the same requirements, or split a boundary only after measuring a specific need.
Logical boundaries versus deployment units in Data lakehouse architecture
- Benefit
- Keeps the responsibilities of Sources, Object storage, Table transaction layer visible without requiring a separate server for each box.
- Cost
- A logical view does not establish isolation, independent scaling or high availability; those require a separate deployment design and operational tests.
- Alternative
- Co-locate components in a simpler deployment first when ownership and workload requirements permit it.
When to use this architecture
- Explain data lakehouse architecture to engineers or stakeholders using a concrete visual flow.
- Review this decision before deployment: Open files still need transactional table metadata.
- Adapt a starting diagram for your own requirements, then record and validate every changed boundary.
Key components
- Sources
- Sources participates in data lakehouse architecture. Sources → Object storage. Define its contract, owner and access permissions in your implementation.
- Object storage
- Object storage participates in data lakehouse architecture. Sources → Object storage; Object storage → Table transaction layer. Define its contract, owner and access permissions in your implementation.
- Table transaction layer
- Table transaction layer participates in data lakehouse architecture. Object storage → Table transaction layer; Table transaction layer → SQL engine; Table transaction layer → ML workloads. Define its contract, owner and access permissions in your implementation.
- SQL engine
- SQL engine participates in data lakehouse architecture. Table transaction layer → SQL engine; SQL engine → Analytics. Define its contract, owner and access permissions in your implementation.
- Analytics
- Analytics participates in data lakehouse architecture. SQL engine → Analytics. Define its contract, owner and access permissions in your implementation.
- ML workloads
- ML workloads participates in data lakehouse architecture. Table transaction layer → ML workloads. Define its contract, owner and access permissions in your implementation.
Questions about this template
What does this data lakehouse architecture diagram show?
It shows 6 components with the core flow Sources → Object storage → Table transaction layer → SQL engine → Analytics. Open files still need transactional table metadata.
Can I edit this diagram without signing in?
Yes. Open the template in the browser-local canvas to change nodes, connections and source. AI generation and cloud account storage require sign-in.
Is this a complete production implementation?
No. It is an editable reference. Validate the security controls, dependency behavior, capacity and recovery plan against your own requirements before implementation.
Validate your implementation
Use the relevant platform documentation to verify service limits, supported features and configuration. This diagram is an original reference rather than a reproduction of a provider deployment.