Evolve - SaaS Lens

This whitepaper contains additional guidance not found in the SaaS Lens of the AWS Well-Architected Tool.


SaaS OPS 2: How are you capturing and surfacing metric data that can be used to analyze the usage and consumption trends of individual tenants?

To continually evolve your SaaS operations experience, you’ll need to have access to a rich collection of operational data that can be used to analyze your multi-tenant SaaS environment. This often means instrumenting and publishing a much richer collection of tenant metrics from your application that can accurately capture the activity and consumption patterns of the tenants that are exercising your environment.

SaaS metrics go beyond the fundamental of infrastructure consumption (CPU, memory, etc.), identifying specific operational usage patterns that are fundamental to understanding the operation of multi-tenant loads in your environment. Analysis of this data will allow SaaS organizations to assess trends that are happening across their system and identify opportunities to introduce policies or changes to the underlying architecture that will continually improve the reliability, scalability, cost efficiency, and overall agility of a SaaS offering.

There are two distinct elements of capturing and surfacing metric data. First, your application must publish the metric data that can provide useful operational insights into your SaaS environment. You’ll need to identify key points in your application where you’ll want to capture and publish these metrics.

The other piece of the puzzle here is the ingestion, aggregation, and surfacing of this data. There is a wide spectrum of tools that you can choose here from AWS or one of the AWS Partner Network (APN) Partners. Ultimately, the ingestion component of this becomes more of a data warehouse and business intelligence question. The SaaS architecture scenarios listed previously in this document includes a Tenant Insights scenario. This scenario outlines an architecture for ingesting metric data. That architecture represents one of the patterns you can apply to address this need.

While our focus here is on the operational view of these tenant metrics, you can imagine that these metrics have usage in multiple contexts across a SaaS business. The operational requirement is for your architecture to ensure that the organization is actively collecting and analyzing tenant activity and consumption trends to identify opportunities to evolve your SaaS system. This fits with the broader theme of building the foundational tooling that can be used to assess the continually shifting mix of tenants and tenant workloads.