Data Architecture Expert
Master the design of modern data architectures, from data lake to data mesh, and drive corporate data governance strategy.
Key skills
- ✓Design scalable data architectures (batch, streaming, hybrid)
- ✓Master modern patterns: data lake, lakehouse, data mesh, data fabric
- ✓Implement data governance, quality and security
- ✓Drive technology choices and architecture migrations
The question bank for this exam is prepared after your purchase. The exam is available no later than 48 hours after payment; you will receive an email as soon as it opens.
Is this right for you?
✓ This certification is right for you if:
- →Data Architects wishing to validate their expertise on modern patterns
- →Senior Data Engineers moving into architecture roles
- →IT Solutions Architects specialising their career path towards data
✗ This certification is not suitable if:
- →Data beginners: first consider a Data Engineering Foundations or SQL/Database Fundamentals certification to acquire the basics.
- →Senior Data Architects with 8+ years of experience: consider our Master 801 Data Strategy & Leadership certification to drive enterprise-scale data transformation.
Prerequisites & eligibility
No formal prerequisites, but 3 years of experience in data engineering or architecture strongly recommended.
Purchase options
Exam voucher, LMS pathway, bundle packs — build your order
Options d'achat
Composez votre parcours — voucher d'examen, préparation LMS, packs entreprise.
Skills assessed
Modern architectures
Design data lake, lakehouse, data mesh and data fabric architectures suited to business needs.
Data governance
Implement governance, quality, security and compliance (GDPR, lineage) strategies.
Scalability & performance
Optimise batch and streaming pipelines for massive volumes and high availability.
Technology choices
Evaluate and select suitable stacks: cloud, on-premise, open source or proprietary.
Exam format
Curriculum
Fundamentals of data architecture
Enterprise architecture principles applied to data. Conceptual modelling (entity-relationship), logical (relational schemas, star schema, snowflake) and physical (partitioning, indexing). Normalisation (3NF, BCNF) vs denormalisation for performance. Architectural differences between data warehouse (Kimball, Inmon), data mart, ODS and data lake. Architectural patterns: Lambda (batch + streaming), Kappa (streaming-first), Medallion (Bronze/Silver/Gold). Compute/storage separation principles. Event-driven architecture (EDA).
Modern and distributed architectures
Data lake: HDFS, S3, ADLS architecture, zones (raw, curated, trusted). Data lakehouse: Delta Lake, Apache Iceberg, Apache Hudi (ACID, time travel, schema evolution). Data mesh: domain-oriented ownership, data as a product, self-serve platform, federated governance. Data fabric: unified data management, active metadata, knowledge graph. Multi-cloud and hybrid architectures: replication strategies, data residency, vendor lock-in. Streaming architecture: Apache Kafka (topics, partitions, consumer groups), Apache Pulsar, AWS Kinesis, Azure Event Hubs. Change Data Capture (CDC) with Debezium, Airbyte. Real-time analytics with Apache Flink, ksqlDB.
Governance, quality and security
Governance frameworks: DAMA-DMBOK, DCAM. Data catalog and metadata management: Collibra, Alation, Azure Purview, AWS Glue Catalog, OpenMetadata. Data lineage: end-to-end traceability, impact analysis. Data quality: dimensions (accuracy, completeness, consistency, freshness), tools (Great Expectations, Deequ, Soda). Master Data Management (MDM): golden records, matching, merging. GDPR compliance: privacy by design, right to erasure, portability, pseudonymisation. Security: encryption at rest (AES-256, KMS) and in transit (TLS), data masking, tokenization, RBAC/ABAC, data loss prevention (DLP). Audit and logging.
Data technologies and stacks
Relational databases: PostgreSQL, MySQL, Oracle Database, SQL Server, HA architectures (replication, failover). NoSQL: MongoDB (document), Cassandra (wide-column), DynamoDB, Redis (key-value), Neo4j (graph). Cloud data warehouses: Snowflake (multi-cluster architecture), Google BigQuery (columnar storage, BI Engine), Amazon Redshift (Spectrum, RA3 nodes), Azure Synapse Analytics. Distributed processing engines: Apache Spark (RDD, DataFrame, Catalyst optimizer), Apache Flink (stateful streaming), Databricks (Delta Engine, Photon). Orchestration: Apache Airflow (DAGs), Prefect, Dagster, AWS Step Functions. File formats: Parquet, ORC, Avro, JSON, CSV. Compression: Snappy, Gzip, LZ4.
How to fund your certification
3 ways to cover the cost of your certification
OPCO — Employer funding
Ask your employer to cover the cost through your OPCO (employee training plan).
Personal — €450 incl. VAT
Secure online payment by credit card (Stripe).
Business rate: €585 excl. VAT
Business — custom quote
Teams, cohorts or private catalogue: a quote tailored to your volumes and goals.
Frequently asked questions
What is the difference with cloud certifications (AWS, Azure, GCP)?
PROVA Data Architecture Expert is cloud-agnostic and covers all modern patterns (data mesh, lakehouse, data fabric) with a strategic architectural approach, not just operational on a specific cloud.
Do I need to know all the tools mentioned in the syllabus?
No. The exam assesses your ability to choose and compare technologies based on architectural criteria (scalability, cost, governance), not exhaustive mastery of each tool. Practical knowledge of 2-3 stacks is sufficient.
Does this certification cover machine learning and AI?
Partially: it covers the architecture of data pipelines for ML (feature stores, MLOps), but not modelling or training. For that, consult our specialised AI certifications.
Can I take the exam in English?
Yes, the exam is available in French and English. You choose the language at the time of booking.
Your certification pathway
Before, during, after — the recommended progression
Droits du candidat
Transparence
Les critères de décision et le barème sont documentés et accessibles avant l'examen.
Droit d'appel
Toute décision peut être contestée dans les 30 jours. Examen par un tiers indépendant.
Plainte
Toute personne peut signaler un dysfonctionnement. Formulaire public accessible sans compte.
