Launched in November 2013, Kinesis is a managed Amazon Web Services (AWS) solution for real-time processing and analytics of streaming data. Kinesis offers a couple different streaming services:
- Data Streams to process data records without managing servers.
- Video Streams to capture and store media streams, e.g. for playback.
There is already an abundance of documentation online which dives into the workings of Kinesis. Amazon’s own documentation would be a good reference to begin with. In this blog post, we will focus on Kinesis Data Streams monitoring, following a similar format to my previous post about EventBridge Rules.
We will start with an overview regarding how Kinesis metrics work before covering current Data Stream metric offerings of notable interest. Where applicable, we will take note of their limitations. We will then finish with suggested steps that can add customer value to Data Stream observability.
Like other AWS services, Kinesis offers CloudWatch metrics to monitor Data Stream performance. Two levels of metrics are available: stream level and shard level. For…