DP-203 study plan
DP-203 · Give each week a clear focus.
8 weeks · About 6 hours per week
Design and implement data storage
Map each Azure Storage service (Blob, Data Lake Storage Gen2, Files, Tables, Queues) to the DP-203 storage design scenarios it satisfies
Configure a hierarchical namespace-enabled storage account and explain how it changes blob semantics
Provision Blob storage tiers and lifecycle management rules for a sample dataset
Write and run CLI or PowerShell commands that create containers, upload data, and set access tiers
Your checkpoint
A one-page storage service selection matrix plus a working storage account with three containers, lifecycle rules, and uploaded sample files
Design and implement data storage
Design partitioned layouts for files in Data Lake Storage Gen2 and justify the partition key choices
Create external tables and views over Parquet and CSV files using a serverless SQL pool
Implement a distribution and index strategy for a dedicated SQL pool table
Compare rowstore, columnstore, hash, round-robin, and replicated designs against query patterns
Your checkpoint
A written design note covering partition layout, file format, and distribution choices, validated by three working queries against the same dataset
Design and implement data storage & Develop data processing
Build a dimensional model (fact and dimensions) for a sample retail scenario and load it into a dedicated SQL pool
Configure PolyBase or COPY INTO to ingest files from Data Lake Storage into the warehouse
Implement slowly changing dimension logic for at least one dimension table
Document the ingestion pattern and its failure handling
Your checkpoint
An end-to-end load pipeline that populates a star schema from raw files, with a run log showing a successful and a retried execution
Develop data processing
Author a pipeline with a copy activity that parameterizes source and sink connections
Add control-flow activities (ForEach, If Condition, Lookup) to drive dynamic ingestion
Configure triggers and schedule the pipeline to run on a defined cadence
Capture pipeline run history and interpret activity-level error output
Your checkpoint
A parameterized pipeline that ingests multiple source files via ForEach, triggered on a schedule, with a screenshot-free run summary documenting success and failure paths
Develop data processing
Implement a mapping data flow that joins, aggregates, and derives columns
Use a data flow sink with partitioning and compare it to a copy activity for the same task
Write a Databricks or Spark notebook that reads and transforms the same dataset
Choose between pipeline data flows, Spark notebooks, and stored procedures for three given scenarios and justify each choice
Your checkpoint
Two working implementations of the same transformation (data flow and Spark notebook) plus a decision memo comparing their cost, latency, and maintainability
Develop data processing & Secure, monitor, and optimize data storage and data processing
Implement stream ingestion with Event Hubs or Stream Analytics and land results in a storage sink
Configure windowing and watermark behavior for a streaming query
Apply column-level and row-level security to a warehouse table
Set up dynamic data masking and verify masked output for a non-privileged user
Your checkpoint
A streaming job producing aggregated output into storage, plus a security test log showing masked and restricted results for a low-privilege account
Secure, monitor, and optimize data storage and data processing
Configure managed private endpoints and firewall rules for a storage account and a Synapse workspace
Implement credential management with managed identities or stored secrets in a key vault
Create monitoring alerts for pipeline failures, long-running queries, and storage capacity
Tune a slow query using distribution, index, and statistics changes and measure the improvement
Your checkpoint
A hardening checklist with evidence of private connectivity, secret-free authentication, three active alerts, and before-and-after query timings
Design and implement data storage & Develop data processing & Secure, monitor, and optimize data storage and data processing
Rehearse storage design decisions under time pressure using timed scenario prompts
Rehearse pipeline and transformation choices under time pressure using timed scenario prompts
Review every missed item and write a corrected rationale for each
Consolidate weak areas into a final one-page reference sheet
Your checkpoint
A scored practice run above 80 percent with a written remediation list for every missed item
Milestones
Week 2: Can explain and defend partition, file format, and distribution choices for a new dataset without notes
Week 4: Can build a parameterized, scheduled pipeline from scratch in under 45 minutes
Week 6: Can implement both a streaming aggregate and a row-level security policy unaided
Week 8: Scores above 80 percent on a timed mixed-domain practice run and has closed every listed gap