Ravindra BagaleCourses & study guides मराठी Track your progress

Ravi Sir's Cards › AWS

Streams vs Firehose

This revision card explains Streams vs Firehose. You may only need S3/Redshift delivery without custom consumers. It describes data Streams — real-time consumers you build. Firehose: managed delivery to S3/OpenSearch/Redshift with buffering. The how-to section covers how to use Streams for fraud scoring Lambda; Firehose to land raw JSON in S3 for later Athena. An example on the card is clickstream Firehose buffer 60s , then parquet in s3://lake/clicks/.

Streams vs Firehose You may only need S3/Redshift delivery without custom consumers. Data Streams: real-time consumers you build. Firehose: managed delivery to S3/OpenSearch/Redshift with buffering. Use Streams for fraud scoring Lambda; Firehose to land raw JSON in S3 for later Athena. Clickstream Firehose buffer 60s → parquet in s3://lake/clicks/.
Streams vs Firehose · Download PNG
Read card text

What this card says

Why do we need it?

You may only need S3/Redshift delivery without custom consumers.

What is it?

Data Streams: real-time consumers you build. Firehose: managed delivery to S3/OpenSearch/Redshift with buffering.

Firehose

How to do it

Use Streams for fraud scoring Lambda; Firehose to land raw JSON in S3 for later Athena.

Example

Clickstream Firehose buffer 60s → parquet in s3://lake/clicks/.

See all AWS revision cards