These next few posts will be about AWS Kinesis, to ensure each point is covered I will divide over a number of posts. First off we have Kinesis Data Streams, these allow you to capture and process data produced by, you guessed it, producers. This data is then processed by Kinesis and sent to a consumer who is able to analyze said data. The data stream is divided into shards with a capacity of 1 MB/s in and 2 MB/s out. If you have more data to analyze, then you need more shards. The retention period for this data is 1-365 days and is immutable, meaning it can’t be deleted. The two capacity modes offered for the data streams are provisioned and on-demand. Provisioned, you pay per share provisioned per hour, this option is excellent if you know exactly how much you are processing. On-demand, you pay per stream and amount of data in and out, this option is more expensive but is a useful option if you want to get a determination of how much capacity you are actually using. It is important to note the on-demand scales automatically based on usage. In terms of security HTTPS is utilized and AWS KMS is available as well.