Warm Throughput and Planned Spikes
Understand the concept of warm throughput in Amazon Keyspaces and how pre-warming prepares tables to handle sudden traffic surges. Learn when and how to apply warm throughput settings for migration cutovers, launch events, and scheduled spikes, while emphasizing the importance of proper partition-key design to prevent hot partition throttling. Gain insights into monitoring strategies, sizing best practices, and operational controls to maintain reliable database performance during planned high-demand periods.
Replication readiness across Regions, covered in the previous lesson, is only one dimension of capacity planning. A table that can replicate globally still needs enough local throughput to absorb traffic the instant it arrives. When an Amazon Keyspaces table receives a sudden, large burst of reads or writes during a migration cutover, a product launch, or a scheduled promotional event, the table's internal partitions may not yet be scaled to handle that load immediately.
Warm throughput is the read and write capacity a table can sustain right now, without waiting for adaptive scaling to detect rising demand and react. By default, a newly created or lightly used table starts with a modest warm-throughput level. Adaptive scaling gradually increases capacity as traffic grows. For steady, organic growth, this works well, but for a known spike that arrives all at once, the ramp-up lag causes throttling. Requests are rejected because the underlying partitions have not yet expanded.
Pre-warming is the mechanism that lets operators declare a higher warm-throughput target ahead of time so the table is ready the moment traffic hits. Think of it like preheating an oven before baking: you set the temperature in advance so cooking begins immediately rather than waiting for the oven to reach the right heat.
This lesson covers how pre-warming works, when to use it, how it relates to partition distribution, and why it does not replace good partition-key design.
The following diagram illustrates the contrast between a pre-warmed table and one that relies solely on adaptive scaling when a planned spike arrives.
How pre-warming works
Operators can specify a warm-throughput target when creating a new table or update it on an existing table through the Amazon Keyspaces console, AWS CLI, or API. When the target is set, Amazon Keyspaces pre-allocates internal partition resources ...