Keyspaces

In this lesson we cover Amazon Keyspaces — AWS’s managed, Cassandra-compatible database service for high-throughput, low-latency workloads. Amazon Keyspaces (for Apache Cassandra) delivers a serverless Cassandra experience: AWS manages the underlying infrastructure, automatically scales capacity, and provides built-in security, availability, and monitoring. Applications use Cassandra Query Language (CQL) and standard Cassandra drivers to interact with Keyspaces, minimizing migration or tooling changes. Key highlights:

Fully managed, Cassandra-compatible service using familiar CQL and Cassandra drivers.
Serverless scaling with on-demand billing or provisioned throughput (with auto scaling).
High availability across Availability Zones, encryption at rest, IAM integration, and VPC endpoints.
Built-in TTL (time-to-live) for automatic expiration of data and managed backups.
Support for multi-region (active-active) deployments with storage-based asynchronous replication.

What is Amazon Keyspaces Amazon Keyspaces is designed to remove the operational burden of running Apache Cassandra clusters: provisioning, patching, node management, and scaling. Because Keyspaces is compatible with CQL and Cassandra drivers, you can often reuse existing application code with minimal changes. How applications use Keyspaces

Use CQL and standard Cassandra drivers (Java, Python, Node.js, etc.) to read and write data.
Keyspaces encrypts data at rest and in transit, integrates with AWS IAM for auth, and supports VPC endpoints for private connectivity.
TTL (time-to-live) is fully managed by Keyspaces — you don’t provision capacity separately for TTL operations.

Quick CQL examples Create a keyspace (CQL):

CREATE KEYSPACE IF NOT EXISTS demo_keyspace
WITH REPLICATION = {'class': 'SingleRegionStrategy'};

Create a table and use it:

CREATE TABLE IF NOT EXISTS demo_keyspace.devices (
  device_id uuid,
  ts timestamp,
  temperature double,
  PRIMARY KEY (device_id, ts)
) WITH CLUSTERING ORDER BY (ts DESC);

INSERT INTO demo_keyspace.devices (device_id, ts, temperature)
VALUES (uuid(), toTimestamp(now()), 72.5);

SELECT * FROM demo_keyspace.devices WHERE device_id = 123e4567-e89b-12d3-a456-426655440000;

Capacity modes Keyspaces supports two throughput capacity modes. Choose the mode that matches your workload characteristics:

Mode	How it works	Best for
On-demand capacity	You pay per read/write request. Keyspaces instantly scales to adapt to unpredictable or spiky traffic.	Applications with irregular, bursty, or unpredictable traffic.
Provisioned capacity	You specify read/write capacity units; enable auto scaling to adjust capacity as load changes.	Stable, predictable workloads where cost can be optimized by provisioning.

Choose on‑demand for unpredictable or bursty workloads and provisioned capacity (with auto scaling) for predictable workloads where cost optimization matters.

You can switch a table’s capacity mode (on‑demand ↔ provisioned) at most once per day. Plan capacity changes accordingly.

Multi-region Keyspaces Keyspaces supports multi-region (global) tables implemented as replica tables — one replica per region — that share a single logical schema. Writes in any region are propagated to other regions using storage-based asynchronous replication. Typical replication lag is low (often under one second), and replication occurs in storage so it has minimal effect on query performance. Benefits of multi-region Keyspaces:

Active‑active global reads and writes with low latency for local users.
Improved business continuity: if a region fails, traffic can be routed to another active region.
Built-in conflict resolution for write conflicts across regions.

When to use Keyspaces Keyspaces is a strong fit when you need Cassandra compatibility without operational overhead. Typical use cases include:

High-volume telemetry and time-series ingestion (industrial IoT).
Real-time trading and market monitoring systems.
Fleet management, routing, or any system that needs low-latency reads and high write throughput.

Key advantages:

No Cassandra cluster management (provisioning, patching, repair).
Seamless use of existing CQL-based tools and drivers.
Built-in security and monitoring integrations.

Key features (quick reference)

Feature	Description
CQL & driver compatibility	Use existing Cassandra drivers and CQL tools (cqlsh, DataStax drivers).
Serverless scaling	On-demand mode scales automatically; provisioned mode supports auto scaling.
Multi-AZ HA	Data is replicated across Availability Zones for availability.
Multi-region replication	Active-active global tables with storage-based replication.
Security & compliance	Encryption at rest/in transit, IAM integration, and VPC endpoints.
Managed TTL & backups	Automatic TTL handling and AWS-managed backups/retention.

Summary Amazon Keyspaces is a fully managed, Cassandra-compatible, serverless database service that reduces operational complexity while preserving the familiar CQL interface. Choose on-demand capacity for unpredictable traffic and provisioned capacity (with auto scaling) for steady, predictable workloads. Multi-region active-active tables provide global reads and writes with low latency and built-in conflict resolution, while the service handles security, replication, TTL, and monitoring for you.

AWS Databases - Part 1

AWS Databases - Part 2

Migrations & Security

Introduction

Watch Video