DP-500 study plan

Map the DP-500 domain weights to your own experience gaps and rank the gaps by exam impact

Sketch a layered lakehouse architecture (ingestion, bronze/silver/gold, serving layer) and justify each layer's technology choice in writing

Provision a Synapse Analytics workspace with a dedicated SQL pool and a Spark pool, then run one query against each to confirm connectivity

Compare Parquet, Delta, and CSV storage for a 1 GB sample dataset and record file size, query latency, and schema-evolution behavior

Build an ingestion pipeline that lands raw files into a bronze layer and promotes cleaned data to silver using a notebook or Data Factory activity

Implement incremental loading with a watermark column and demonstrate that a re-run does not duplicate rows

Configure a dedicated SQL pool table with hash distribution on a high-cardinality key and compare query time against round-robin distribution

Write and execute three queries that use result-set caching or materialized views and note the performance delta

Design a star schema for a sales scenario, defining grain for the fact table and conformed dimensions

Implement the schema physically and populate it from the silver layer with a transformation notebook

Model a slowly changing dimension Type 2 and verify historical rows are preserved after two updates

Explain in writing when to choose star versus snowflake versus a wide denormalized table for analytical workloads

Apply column-level sensitivity classifications and dynamic data masking to a table containing personal data

Implement row-level security with a predicate function and prove a restricted user sees only permitted rows

Configure object-level and schema-level permissions for two personas (analyst, engineer) and test each with the actual account

Document a data lineage path from source file to curated table using the catalog or workspace lineage view

Publish a semantic model over the curated layer and define relationships, hierarchies, and at least five explicit measures

Write DAX measures for year-to-date, prior-year comparison, and a ratio using DIVIDE with a safe-divide pattern

Implement row-level security in the model and validate it with the role-testing feature

Design an incremental refresh policy and explain which parameters it depends on

Build a two-page report using bookmarks, drillthrough, and a custom tooltip to support a guided analysis flow

Optimize a slow visual by reducing cardinality or moving logic into the model, and measure the improvement with Performance Analyzer

Configure a deployment pipeline across development, test, and production stages and promote content through it

Set up a scheduled refresh with gateway or cloud credentials and confirm a successful run

Enable logging and query the monitoring data to identify the three longest-running queries of the past week

Interpret an execution plan for one slow query and apply one fix such as statistics update or index/partition change

Configure alerts on a key pipeline failure and on a refresh failure, then trigger each deliberately

Right-size a workload by comparing compute utilization against cost and proposing a scaling or pausing policy

Re-take a full-length timed practice set and log every missed item by domain and root cause

Re-study only the two weakest domains using your own notes and re-test those areas

Recite from memory the decision criteria for distribution types, security layers, and refresh modes

Review your deliverables from weeks 1 through 7 and close any incomplete item