MongoDB Atlas Admin Associate study plan

Deploy an M10 dedicated replica set cluster in a chosen cloud region, then scale it to M20 and back while recording each change and its deployment time

Compare M0/M2/M5 shared tiers against M10-and-above dedicated tiers on RAM, storage, IOPS, and backup availability, then justify a tier choice for a stated workload

Configure a multi-region replica set with a chosen number of electable and read-only nodes, and explain how write concern majority interacts with node distribution

Distinguish when a replica set versus a sharded cluster is the correct Atlas deployment model, including shard key and cluster-tier implications

Configure cluster autoscaling, pause/resume, termination protection, and maintenance windows, and state exactly what triggers each behavior

Enable SCRAM authentication and create database users scoped to a single database with the minimum built-in role required

Map built-in roles (read, readWrite, dbAdmin, clusterMonitor, backup, atlasAdmin) to three distinct job functions and defend each assignment

Author a custom role limited to specific actions and resources, then verify least privilege by running four mongosh commands that should partly fail

Configure federated authentication through a test identity provider and trace how an external group maps to an Atlas organization or project role

Enable encryption at rest with a customer-managed key through a KMIP endpoint, or document prerequisites and every configuration step when no KMS is available

Turn on database auditing, then filter and export audit records for one specific event type such as authentication failures

Create organization and project API keys with least privilege and an IP access list, then rotate a key without breaking an existing automation script

Build IP access list entries for a specific source range and prove both a permitted and a denied connection using mongosh output

Establish private connectivity from a cloud VPC to Atlas using VPC peering or a private endpoint, and document required CIDRs, routes, and DNS settings

Construct correct SRV and standard connection strings, then connect with mongosh and one driver from an application host with TLS enforced

Diagnose three deliberate connectivity failures caused by allowlist gaps, DNS resolution, and TLS or certificate mismatch, citing the log or metric that revealed each

Interpret opcounters, query targeting, scan-and-order ratio, connections, disk IOPS, and cache metrics to explain a degraded workload

Create alert configurations for replication lag, disk utilization, connection count, and query targeting ratio, and route them to a notification channel

Use Performance Advisor and the Query Profiler to isolate one slow query, add the suggested index, and confirm the gain with explain() plans

Run a synthetic load test while reading the real-time performance panel, and identify which metric saturated first

Enable continuous cloud backup with a snapshot schedule and retention policy, and take an on-demand snapshot of a test cluster

Restore a snapshot to a brand-new cluster and verify completeness using document counts and a checksum comparison

Perform a point-in-time restore to a timestamp immediately before a deliberate document deletion and confirm the record is recovered

Compare snapshot restore, point-in-time restore, queryable backup, and snapshot export by the RPO and RTO each one delivers

Automate project, cluster, database-user, and access-list creation with the Atlas CLI and Admin API in one repeatable script

Define a project, cluster, and database user with the Terraform Atlas provider and run plan, apply, then destroy cleanly

Migrate a sample dataset into Atlas with mongodump and mongorestore, then repeat with a live migration approach and record cutover downtime for each

Load data with $out or $merge and configure an Online Archive or Data Federation query, comparing cost and query patterns against keeping the data live

Diagnose three fault scenarios - an election storm, connection pool exhaustion, and storage saturation - and write the ordered diagnostic steps for each

Draft an operational runbook covering major version upgrade, vertical scaling, and maintenance-window behavior with expected impact on clients

Take a timed full-length practice test across all eight domains, then classify every wrong answer by domain and root cause

Re-run the week 1 through week 4 labs that produced the lowest quiz scores and confirm each weak objective now passes