Kinesis Data Streams vs Amazon Data Firehose
People think Firehose is just another name for Kinesis Data Streams. Kinesis Data Streams keeps events for a while, and your own apps read them, as many times as they like. Firehose keeps nothing to replay. It gathers events into batches and drops them into a store, like Amazon S3. One is a stream you read. The other is a chute that delivers.
Amazon Kinesis Data Streams
AWS's managed service for event streams. Streams are split into shards.
Amazon Kinesis Data Streams is the managed event streaming service on AWS. A stream is split into shards, and each shard carries a set amount of traffic. You add shards to carry more, or use on-demand mode and let AWS scale it. Records are kept for 24 hours by default, and that can be raised to 365 days. AWS runs the servers for you.
Amazon Data Firehose
An AWS service that takes a stream and drops it in a store. No servers to run.
Amazon Data Firehose is an AWS service that delivers streaming data to a store. Apps send records to a Firehose stream. Firehose holds them for a short time, then writes them in batches. It can deliver to Amazon S3, Amazon Redshift, Splunk, Apache Iceberg tables and more. There are no servers to run. It can also change the data on the way.
In one line
- Amazon Kinesis Data Streams
- AWS's managed service for event streams. Streams are split into shards.
- Amazon Data Firehose
- An AWS service that takes a stream and drops it in a store. No servers to run.