Google Cloud → AWS and Azure
AWS and Azure equivalents of Bigtable
The closest AWS equivalents of Google Cloud Bigtable are Amazon DynamoDB global tables, Amazon DynamoDB Streams, Amazon Keyspaces (for Apache Cassandra) and Amazon Timestream for LiveAnalytics; on Azure, Azure Cosmos DB global distribution and multi-region writes, Azure Cosmos DB change feed, Azure Cosmos DB for Apache Cassandra, Azure Managed Instance for Apache Cassandra and Azure Data Explorer.
About Bigtable
| Provider | Google Cloud |
|---|---|
| Status | Generally available |
| What it is | Wide-column NoSQL database for large, low-latency workloads, with an HBase-compatible client, replication across clusters, and change streams. |
| Documentation | docs.cloud.google.com |
| Pricing | Official pricing page |
Mappings (4)
Multi-region, multi-active NoSQL replication Partial match
DynamoDB global tables map partly to Firestore multi-region locations (one database replicated across regions) and Bigtable replication with multi-cluster routing (writes in any cluster, eventually consistent).
- Firestore multi-region locations are fixed configurations, not arbitrary region sets.
- Bigtable reads and writes to different clusters are eventually consistent, as with global tables.
NoSQL change data capture stream Partial match
DynamoDB Streams maps to Bigtable change streams for Bigtable. Firestore changes are consumed through Eventarc triggers or client listeners rather than a stream API.
- Spanner also has change streams for relational data.
Apache Cassandra-compatible wide-column database Partial match
Amazon Keyspaces maps partly to Bigtable (wide-column, HBase-compatible) and to Spanner's Cassandra Adapter, which lets Cassandra clients use Spanner.
- Bigtable isn't CQL-compatible; data models and queries need rework.
- The Spanner Cassandra Adapter keeps CQL clients but runs on Spanner's engine.
Serverless time-series database Partial match
Amazon Timestream has no purpose-built Google Cloud counterpart. Bigtable is commonly used for time-series data, with BigQuery for analysis.
- You design row keys for time-series access yourself.