Amazon Kinesis Data Streams
aws.amazon.com
What it is
Amazon Kinesis Data Streams is a cloud, serverless service for capturing, processing, and analyzing live data streams. It can continuously capture gigabytes per second from hundreds of thousands of sources, then make records available to multiple real-time analytics applications, Amazon Managed Service for Apache Flink, or AWS Lambda within 70 milliseconds of collection. Data is synchronously replicated across three Availability Zones in an AWS Region. Each stream supports up to 20 consumers with dedicated read throughput, and retention can be configured for as long as 365 days. Server-side encryption uses AWS Key Management Service keys, while Amazon VPC provides private access. Integrations include DynamoDB, Aurora, Database Migration Service, CloudWatch, Lambda, Managed Service for Apache Flink, and Kinesis Data Firehose. Records can be delivered to Amazon S3 Tables as Apache Iceberg tables or to S3 buckets in source format. It supports log and event streaming, real-time analytics, event-driven applications, and real-time AI or machine-learning workloads. Access is through an API or web interface. Pricing is paid from $0.03, with no free plan or AWS Free Tier availability.
Spec sheet
- What it does
- Amazon Kinesis Data Streams is a serverless, real-time data streaming service for capturing, processing, and analyzing data streams at any scale.Read 1 Oct 2026 · aws.amazon.com
- Integrations
- AWS lists integrations with Amazon DynamoDB, Amazon Aurora, AWS Database Migration Service, Amazon CloudWatch, AWS Lambda, Amazon Managed Service for Apache Flink, and Amazon Kinesis Data Firehose.Read 1 Oct 2026 · aws.amazon.com
- Security
- The service supports server-side encryption with AWS Key Management Service keys and private access through Amazon VPC.Read 1 Oct 2026 · aws.amazon.com
- Availability
- Streaming data is synchronously replicated across three Availability Zones in an AWS Region.Read 1 Oct 2026 · aws.amazon.com
- Consumer fan-out
- A Kinesis data stream can have up to 20 consumers, each with dedicated read throughput.Read 1 Oct 2026 · aws.amazon.com
- Data delivery
- Kinesis Data Streams can deliver records directly to Amazon S3 Tables as Apache Iceberg tables and to general-purpose Amazon S3 buckets in source format.Read 1 Oct 2026 · aws.amazon.com
- Free tier
- Amazon Kinesis Data Streams is not currently available in the AWS Free Tier.Read 1 Oct 2026 · aws.amazon.com
- Getting started
- Users can create a stream through the AWS Management Console or CreateStream API, configure producers, and build applications to read and process records.Read 1 Oct 2026 · aws.amazon.com
- Latency
- Streaming data can be made available to multiple real-time analytics applications, Amazon Managed Service for Apache Flink, or AWS Lambda within 70 milliseconds of collection.Read 1 Oct 2026 · aws.amazon.com
- Retention
- Kinesis Data Streams supports retention for up to 365 days.Read 1 Oct 2026 · aws.amazon.com
- Scale
- It can continuously capture gigabytes of data per second from hundreds of thousands of sources.Read 1 Oct 2026 · aws.amazon.com
- Security model
- AWS states that security is a shared responsibility and provides guidance for data protection, IAM access control, compliance validation, resilience, and infrastructure security.Read 1 Oct 2026 · docs.aws.amazon.com
- Use cases
- AWS describes use cases including log and event streaming, real-time analytics, event-driven applications, and real-time AI and machine-learning applications.Read 1 Oct 2026 · aws.amazon.com
Plans and prices
Strong points and limits
Strong points
- Captures gigabytes per second from hundreds of thousands of sources
- Data available to consumers within 70 milliseconds
- Retention configurable for up to 365 days
- Synchronous replication across three Availability Zones
- Supports Go, Java, Python, and many other languages
Limits
- No free plan or AWS Free Tier availability
- Paid pricing starts at $0.03
- At-least-once delivery guarantee
- Cloud deployment only
Who it suits
It suits teams building cloud applications that capture high-volume streams for analytics, event handling, or machine-learning workloads. API and web access, stream processing, schema registry, and support for many client languages fit application developers.
Verdict
Amazon Kinesis Data Streams combines high-volume ingestion, sub-second availability, configurable retention, and AWS integrations. It is a paid cloud service with no free tier, so usage and retention choices affect cost.
Where Specifiction places it
- Event Streaming Platforms#16 of 277.4