Amazon Keyspaces

Amazon Keyspaces is a managed, serverless database service compatible with Apache Cassandra. It supports the Cassandra Query Language API, existing application code, Cassandra drivers, and developer tools, so Cassandra workloads can use familiar interfaces without managing database infrastructure. Tables can use on-demand or provisioned read and write capacity; provisioned capacity can scale automatically. AWS describes single-digit-millisecond read and write performance, with no limit on table size or row count. Multi-Region deployments offer up to 99.999% availability, and data is replicated three times across multiple Availability Zones. Keyspaces also integrates with CloudWatch metrics and can stream changes to Amazon OpenSearch Service. Keyspaces Streams provides change data capture, while point-in-time recovery can restore table data to a selected second within the previous 35 days. Data is encrypted at rest by default, with AWS-owned or customer-managed keys. The free tier lasts three months after the first Keyspaces resource is created, with one free tier per payer account and allowances of 30 million write request units, 30 million read request units, and 1 GB of storage.

Amazon Keyspaces combines Cassandra compatibility with serverless operations, capacity choices, and recovery features. The free tier has a defined term and usage allowance, so teams should account for ongoing paid usage.

It suits teams running Cassandra-compatible applications that need a managed database with adjustable capacity and multi-region options. AWS lists low-latency applications, IoT data, game player profiles, and time-series data among its use cases.

Overview

Amazon Keyspaces is a managed database service for applications built with Apache Cassandra’s query language and tooling. It is best suited to teams migrating Cassandra workloads or building low-latency applications that want AWS to manage the database infrastructure. The trade-off is usage-based capacity and a free tier that ends after three months.

Key features

Keyspaces supports the Cassandra Query Language API, existing application code, Cassandra drivers, and developer tools. That makes it a credible migration target when retaining a Cassandra-compatible interface matters; it is less compelling if a team has no reason to use that ecosystem.

Tables offer on-demand or provisioned read and write capacity. On-demand adjusts to traffic and charges for reads and writes performed, which suits variable workloads but makes usage costs important to monitor. Provisioned capacity lets teams specify expected throughput and use auto scaling, providing more control over capacity settings while still incurring charges for provisioned capacity, storage, and other applicable usage.

AWS describes single-digit-millisecond read and write performance, with no limit on table size or row count; the maximum row size is 1 MB. Data is replicated three times across multiple Availability Zones. Multi-Region deployments offer up to 99.999% availability, a useful option for workloads with regional availability requirements, though that figure applies to Multi-Region deployments.

Operational features include CloudWatch performance metrics, streaming changes to Amazon OpenSearch Service, and change data capture through Keyspaces Streams’ API or the Kinesis Client Library. Point-in-time recovery continuously backs up data and can restore a table to any second in the preceding 35 days. Managed Time to Live settings expire rows and attributes without requiring users to manage tombstones or compaction.

Data is encrypted at rest by default, using AWS-owned or customer-managed keys through AWS Key Management Service. IAM integration, TLS, and AWS PrivateLink connectivity to Amazon VPC provide access and network controls. The service also supports tunable consistency and transactions.

Pricing

Keyspaces is paid software with no free trial. Its ongoing capacity plans have custom pricing, so teams should estimate reads, writes, storage, and other applicable usage before committing.

• On-demand capacity: custom pricing. Pay for reads and writes actually performed as capacity adjusts to application traffic. This suits variable demand, but usage charges continue beyond the free allowance.
• Provisioned capacity: custom pricing. Specify expected reads and writes per second and pay for provisioned capacity, storage, and other applicable usage. Auto scaling can adjust capacity; this is a fit for teams that want to set capacity expectations rather than rely entirely on on-demand throughput.
• AWS Free Tier: free. For the first three months after creating the first Keyspaces resource, it includes 30 million write request units, 30 million read request units, and 1 GB of storage. The allowance is limited to one free tier per payer account, so it is a short evaluation window, not a permanent free plan.
• Database Savings Plans: custom pricing, with a one-year usage commitment and up to 18% savings stated by AWS. It is aimed at teams prepared to commit to usage over that term.

Platforms

Keyspaces is available through API and web platforms. It is a managed, serverless service, with CQL as its query interface and horizontal scaling.

Who it's for

Keyspaces fits teams that want Cassandra compatibility without managing infrastructure, particularly for Cassandra workload migration, low-latency applications, IoT device data, game player profiles, and time-series workloads. Its capacity choices and recovery features also suit teams that need managed scaling and point-in-time restoration. Teams with unpredictable use should weigh on-demand costs, while teams able to forecast throughput can consider provisioned capacity.

Pros and cons

• Pro: Cassandra API, code, drivers, and tools support migration without requiring a different query interface.
• Pro: On-demand and provisioned capacity modes accommodate different traffic and planning needs, with auto scaling for provisioned capacity.
• Pro: Multi-Region availability, multi-AZ replication, and point-in-time recovery support resilient deployments and restoration.
• Con: The free tier lasts only three months and is capped at 30 million read units, 30 million write units, and 1 GB of storage per payer account.
• Con: Ongoing plan prices are custom, and both capacity modes add usage-related charges, requiring cost planning.

Alternatives

Google Cloud Bigtable is worth considering for teams that prefer its Enterprise or Enterprise Plus editions and have a free trial option.

Apache Cassandra is the free, self-managed alternative for teams willing to operate an open-source Cassandra database themselves.

Apache HBase is another free, open-source distributed database, with operators configuring and managing deployments.

SWC-DB is a free, self-hosted option; commercial-purpose licensing may require inquiry where GPLv3 limits use.

Apache Accumulo is a free, open-source sorted distributed key/value store under the Apache License 2.0.

Baidu Tera, Cloudera Data Lake Service, and Elassandra are other alternatives to consider.

Browse more options in Wide-Column Databases and Database Software.

Verdict

Choose Amazon Keyspaces if your application depends on Cassandra compatibility and you want a managed, automatically scaling service with recovery and multi-region options. Its clearest advantage is avoiding infrastructure management without abandoning Cassandra interfaces; look elsewhere if you need permanent free usage or pricing that can be assessed without estimating workload consumption.