A Serverless Approach to Adding Notifications Features to Any Analytics Application

Paul Lappas | Intermix

ABOUT THE TALK

More and more data teams are building analytics applications. These applications present pretty charts to internal users, partners, or customers. The applications source data from your data platform. This can be AWS Redshift, Snowflake, or your favorite metric DB.

Your end-users will love logging into your application and looking at charts. Over time, you will start to hear requests for adding notifications and alarms. Your users are busy, and they want to receive an email when graph XYZ exceed level ABC.

You have a choice. You can certainly build a notification system into your application (hey, it sounds easy right??). Alternatively, your users can keep on using their favorite notification tool.

In this talk, we will discuss an approach to adding notification and alarms functionality to your analytics apps, without bearing the cost of build out a notification system. We will show how AWS Kinesis and Lambda can be used to stream your time-series metrics data from AWS Redshift into AWS Cloudwatch. We will also discuss how this architecture can be easily extended to support other notification tools (e.g. Datadog). We'll discuss gotchas and lessons learned from implementing this architecture at intermix.io.

Download Slides

paul lappas

Co-Founder | Intermix.io

Paul is the co-founder of Intermix. He's experienced in building technology, product, and scaling organizations and establishing high-performing engineering cultures. His prominent work includes bringing one of the first IaaS cloud computing service providers to market, developing a data analytics platform and mobile SDKs used by 1B end-users, and currently solving problems in big data.
Paul Lappas

powered by
hakka-labs-logo.png

PO Box 2087
Wilson, WY 83014