MongoDB Data Modeler Associate study plan
MongoDB Data Modeler Associate · Give each week a clear focus.
8 weeks · About 7 hours per week
Data Modeling
Define the four core modeling questions (what does the app do, which entities exist, what access patterns dominate, what is the read/write ratio) and answer them for two sample applications
Distinguish embedding from referencing for at least six concrete field pairs and justify each choice in one sentence
Identify the three MongoDB document size and nesting constraints that force a schema decision and state each from memory
Sketch an entity-relationship diagram for a small e-commerce catalog using only entities, attributes, and cardinalities
Your checkpoint
A one-page written model for a sample app that states access patterns, cardinalities, and embed-vs-reference decisions with justification for every relationship
Data Modeling & Schema Design Patterns
Apply the extended reference pattern to denormalize the three most-read fields of a frequently joined collection
Apply the subset pattern to cap an unbounded array at a fixed number of most-recent items and write the update logic that maintains it
Compare the computed, attribute, and schema-versioning patterns by writing a short use case for each
Convert a normalized relational schema of five tables into an embedded document model and list every trade-off introduced
Your checkpoint
A before/after schema pair for one workload, with a table listing each pattern applied, the problem it solves, and the cost it adds
Schema Design Patterns
Select the appropriate pattern for five stated problems (bucket, outlier, archive, polymorphic, tree) and defend each pick
Design a bucketing scheme for time-series readings that bounds document count and write the bucket key formula
Model a category tree using both the parent-reference and materialized-path approaches and compare query cost for subtree reads
Document a polymorphic schema with a discriminator field and enumerate the query and index implications
Your checkpoint
A pattern selection matrix covering at least six patterns with the trigger condition, the anti-pattern it removes, and a one-line example document for each
Schema Design Patterns & Document Structure and Relationships
Choose between one-to-few, one-to-many, and one-to-squillions for three cardinality scenarios and state the threshold numbers used
Write the two-sided reference pattern for a many-to-many relationship and show the query needed to read one side
Identify which arrays must be bounded and describe the failure mode when they are not
Explain how document growth affects storage and write performance and name the mitigation for each case
Your checkpoint
A relationship decision guide covering three cardinality bands with a sample document, the read query, and the write query for each
Document Structure and Relationships & Indexing for the Data Model
Derive the compound index key order from three named query patterns using the equality-sort-range rule
Map five sample queries to indexes and mark each as covered, partially covered, or unindexed
Distinguish multikey, partial, sparse, and wildcard indexes by the document shapes each one serves
Diagnose why a given index is not used and list the four most common causes
Your checkpoint
An index plan for one collection: every query pattern listed against its supporting index, with the key order rationale written out
Indexing for the Data Model
Design indexes that serve the embedded-array and subset patterns without creating runaway multikey entries
Apply the ESR rule to three compound indexes and reorder keys where the rule is violated
Evaluate the index cost of a schema change by counting how many indexes a new field would touch
Write the checklist used to review a schema and its indexes together before release
Your checkpoint
A schema-plus-index review checklist of at least ten items that a reviewer could apply to any collection without further context
Data Modeling & Schema Design Patterns & Document Structure and Relationships & Indexing for the Data Model
Redesign a deliberately flawed schema that embeds an unbounded array, duplicates mutable data, and lacks a supporting index
Produce the corrected model with pattern names, relationship types, and index definitions
Time yourself answering mixed-domain scenario questions and log every miss by domain
Re-study only the domains where your logged miss rate exceeds one in five
Your checkpoint
A full redesign document for the flawed schema: original problems, corrected model, index list, and a short rationale per change
Data Modeling & Schema Design Patterns & Document Structure and Relationships & Indexing for the Data Model
Take two full timed practice sets and record scores per domain
Rewrite your personal decision rules for embed-vs-reference and for index key order from memory and check them against your notes
Review every missed item until you can state the correct reasoning without notes
Confirm you can justify each pattern choice by naming the access pattern it optimizes
Your checkpoint
A final one-page decision cheat sheet covering cardinality thresholds, pattern triggers, and index key ordering, written from memory
Milestones
Week 2: You can look at any two collections and state within a minute whether to embed or reference, and name the pattern that applies
Week 4: You can classify any relationship into a cardinality band and produce the sample document plus read and write queries without notes
Week 6: Given a query and an index, you can say whether the index serves it and why, and reorder keys correctly under ESR
Week 8: You score at least 80 percent on a full timed practice set with no domain below 70 percent