Kinesis Streams is a fully managed service for real-time processing of high-volume streaming data. It processes data in real time, and enables highly scalable parallel processing. There are source libraries for sending data to and reading data from a stream. It synchronously replicates your data across three facilities, and is integrated with many AWS and third-party services. It also supports SSL and automatic encryption of data once it is uploaded.

AWS Kinesis makes large-scale data ingestion a lot simpler. It is easy to administer. You have to essentially create a new stream and set the desired capacity in terms of shards. You can scale your capacity to match your data throughput rates and expected volumes. It is typically used to build real-time cloud-based applications, for example, the continuous processing of streaming log data. It provides a cost-efficient mechanism for streaming workloads of any scale.

With Kinesis you can process very high volumes of streaming data (currently up to 15 million records at peak load and growing). As a managed service you also don’t have to predict storage and volume requirements as you would have to do with Apache Kafka and ZooKeeper clusters. It also comes with cross AZ replication. However, remember to configure the max data retention settings (from 24 hours to 7 days) or you risk losing your records.