Achieving Global Observability in Multi-Account and Multi-Region AWS Environments Using Sumo Logic

December 14, 2020 Katie Gray

By Katie Gray, Partner Marketing Manager at Sumo Logic
By Rishi Divate, Principal Product Manager at Sumo Logic
By Bashyam Anant, Sr. Director of Product Management at Sumo Logic

By Srinivas Kesanapally, Principal Partner Solution Architect at AWS

Sumo Logic-Logo-2
Connect with Sumo Logic-2

There is a demand placed on operations teams to implement digital processes into code, yet the number of engineers available to do that work is finite.

This is the case for many customers that work with Sumo Logic, a cloud-based machine data analytics service and AWS Advanced Technology Partner. Sumo Logic has AWS Competencies in Data & Analytics, DevOps, Security, and Containers.

Sumo Logic ingests data across your entire application stack to provide observability and, therefore, reliability of mission-critical apps. It eases the burden placed on engineers, simplifying data collection, unifying visibility, and leveraging machine learning (ML) to speed troubleshooting.

Sumo Logic’s Observability solution is designed specifically for Amazon Web Services (AWS) and seamlessly detects anomalies and events of interest correlated with application incidents.

By leveraging all of the telemetry generated by AWS services, Sumo Logic accelerates issue resolution, automatically determines root cause of failures, and helps customers optimize AWS usage to improve uptime and performance.

In this post, we will walk you through how Snoop uses Sumo Logic and AWS to improve reliability of their application. In the end, this enables them to spend less time fixing problems and more time building revenue generating features.

Snoop’s Observability Implementation Using AWS and Sumo Logic

Snoop is a UK-based consumer fintech company that recently launched a mobile app that uses open banking to offer customized and tailored money saving tips to customers based on their credit card and banking transactions.

“Every day, we think about scale, performance, security, and whether we have the right tools in place to speed the safe delivery of features, mature our operating model, and deliver a great customer experience,” says Jem Walters, Co-Founder and CTO at Snoop.

Snoop’s architecture is completely on AWS, on-demand, and built on serverless wherever possible. The frontend uses Amazon API Gateway, while the backend consists of AWS Lambda and Amazon DynamoDB. It uses Amazon Elastic Container Service (Amazon ECS), AWS Fargate, and Amazon RDS for PostgreSQL for data analytics, data ingestion, and data science, respectively.


Figure 1 – Snoop architecture on AWS.

Given the array of contributing services, enabling observability across the entire stack was a key driver in Snoop implementing Sumo Logic from day one. Snoop wanted to spend their time building products for their customers and less time managing their infrastructure.

“When you’re running a modern app-based startup, real-time is the lifeblood of the business,” says Walters.

Overview of Sumo Logic’s AWS Observability

Sumo Logic’s AWS Observability solution was built just for organizations like Snoop. The heart of the solution is the single pane of glass view of your key AWS services.

As data flows into Sumo Logic, the architectural hierarchy of AWS accounts, regions, services, and entities is automatically populated. This provides high-level account visibility and the ability to drill down into specific entities.

Easy Install via AWS CloudFormation

Sumo Logic automates the collection of AWS telemetry including AWS services added after setup. Using AWS CloudFormation, the collection is configured to pull Amazon CloudWatch logs and metrics, and AWS CloudTrail logs into Sumo Logic.

When data reaches Sumo Logic, it’s enriched with value metadata that enables intuitive hierarchical navigation, and seamlessly pivots from metrics to related log data and vice versa.


Figure 2 – Sumo Logic’s AWS Observability solution.

Check out Sumo Logic’s product documentation to learn more about how to set this up.

Saving Time with Automated Root Cause Exploration

The crux of the Sumo Logic’s AWS Observability solution provides visibility, as well as enabling more efficient root cause analysis.

The Sumo Logic Root Cause Explorer goes beyond the “what” to help customers quickly get to the “why” behind problems. This capability automatically correlates data across affected AWS services, and provides a cause and effect relationship that helps determine the underlying root cause of issues affecting application availability and performance.

For customers like Snoop, who use a wide range of AWS services, the challenge lies in correlating data and determining the cause and effect relationship of issues across different services. Sumo Logic’s Root Cause Explorer was designed for this purpose.

The Root Cause Explorer is powered by machine learning and a decade of industry knowledge to cut down on noise and surface high anomaly events of interest. By first establishing a resource-specific baseline for what a normal day or week of CloudWatch metrics might look like, Sumo Logic can highlight out of ordinary events.

AWS Observability in Action at Snoop

When Snoop implemented Sumo Logic’s AWS Observability solution, they immediately got the broad visibility they were looking for, but were also rewarded with a few other benefits.

“Errors are relatively easy to find. Latency is harder. This solution makes it easier to see these things,” says Andy Makings, Head of DevSecOps at Snoop.

In the Snoop environment, they likened the solution to “seeing the wood from the trees.” There is often so much data that it can be difficult to get the big picture. “Sumo helps bubble up the issues without you looking too hard. This also saves you time,” adds Makings.

In addition, the Snoop team was able to gain insight into the operation of their fully on-demand serverless architecture.

“First thing in the morning, when most people have their breakfast and look at the app, we have to scale up automatically and quickly. There is a bit of latency. It’s not currently causing an issue, but it’s good to know it is happening,” says Makings.

Sumo Logic has helped build a complete view of the dependences of the various services that make up Snoop’s application. For example, if load suddenly increases, they can track the spike on the API gateway, or if there’s a new data ingestion Lambda might slow down on the backend.

The visibility into the operation of each AWS service has given Snoop the ability to be proactive about issues rather than reactive.


With the increasing demand placed on operations and engineering teams, the only solution is better automation of their process, which includes automating observability.

Misconfigurations within critical underlying AWS services can cause downtime or performance degradation, so improving detection and resolution is critical to the reliability of mission-critical applications.

To get started with Sumo Logic’s AWS Observability solution, check out AWS Marketplace.

If you don’t have a Sumo Logic account, sign up for a free trial. For more information about Snoop, check out


Sumo Logic – AWS Partner Spotlight

Sumo Logic is an AWS Competency Partner and secure, cloud-native, machine data analytics service that delivers real-time, continuous intelligence to customers.

Contact Sumo Logic | Partner Overview | AWS Marketplace

*Already worked with Sumo Logic? Rate the Partner

*To review an AWS Partner, you must be a customer that has worked with them directly on a project.

Previous Article
Building a Single Source of Truth with a Data Hub from Semarchy
Building a Single Source of Truth with a Data Hub from Semarchy

Organizations need a comprehensive data management solution that includes data quality, cleansing, de-dupli...

Next Article
How to Mitigate Security and Privacy Resistance in Secure Data Migrations
How to Mitigate Security and Privacy Resistance in Secure Data Migrations

One of the first steps for organizations adopting AWS is migrating data from an on-premises environment or ...