Database

DynamoDB vs Cosmos DB vs Firestore — NoSQL Databases Compared

Compare AWS DynamoDB, Azure Cosmos DB, and Google Cloud Firestore for managed NoSQL. Pricing models, consistency, and scale.

Feature Comparison

FeatureAmazon DynamoDBAzure Cosmos DBGoogle Cloud Firestore
Data modelsKey-value + documentMulti-model (5 APIs)Document
Global distributionGlobal TablesMulti-region, multi-writeMulti-region
Consistency optionsEventual + strong5 levelsStrong
Serverless pricingOn-demand modeServerless tierAlways serverless
Real-time syncAppSync (separate)Change feedBuilt-in

Service Details

Amazon DynamoDB

AWS

Fully managed key-value and document database. Single-digit millisecond performance at any scale with on-demand or provisioned capacity.

On-demand: $1.25/million writes, $0.25/million reads. Provisioned: $0.00065/WCU/hr. Free tier: 25GB + 25 WCU/RCU.
Strengths
  • Predictable single-digit ms latency at any scale
  • True serverless with on-demand mode
  • DAX in-memory caching
  • Global tables for multi-region
Limitations
  • 400KB item size limit
  • Complex query model (no ad-hoc SQL)
  • Can be expensive for scan-heavy workloads

Azure Cosmos DB

Azure

Globally distributed multi-model database. Unique multi-API support (SQL, MongoDB, Cassandra, Gremlin, Table) with tunable consistency.

Serverless: $0.282 per million RUs. Provisioned: $0.008/hr per 100 RUs. Free tier: 1000 RU/s + 25GB.
Strengths
  • 5 consistency levels (unique in the industry)
  • Multi-API (SQL, MongoDB, Cassandra, Gremlin, Table)
  • Global distribution with multi-write regions
  • Serverless tier for intermittent workloads
Limitations
  • RU pricing model is unintuitive
  • Can be expensive at scale
  • Complex cost modeling

Google Cloud Firestore

GCP

Serverless document database with real-time sync and offline support. Strong mobile/web client SDKs via Firebase.

$0.036/100K reads, $0.108/100K writes, $0.018/100K deletes. 1GB free storage. Generous free tier.
Strengths
  • Real-time synchronization
  • Excellent mobile/web SDKs (Firebase)
  • True serverless (no capacity planning)
  • Offline support built-in
Limitations
  • Document-oriented only
  • Query limitations (no cross-collection joins)
  • Less suitable for high-throughput analytics

When to Use Which

Choose DynamoDB for predictable high-throughput key-value workloads on AWS. Choose Cosmos DB for multi-model, multi-region needs with tunable consistency. Choose Firestore for mobile/web apps needing real-time sync and offline support.

NoSQL database costs can spike unpredictably with traffic patterns. CloudExpat monitors your usage patterns and optimizes capacity modes to prevent cost surprises.

Optimize Your Cloud Costs Across All Providers

CloudExpat works with AWS, Azure, and GCP. Connect in 30 seconds and see where you're overspending.