Accolade Doubles Developer Productivity with Elastic Observability on AWS

As its customer base grew and its service offerings expanded, Accolade faced new challenges in managing and processing observability data from disparate platforms and formats. While its previous observability vendor provided helpful pre-aggregated metrics, Accolade leaders found that the overall approach was too inflexible to serve its needs. It was especially challenging that Accolade didn’t have full ownership of the data and that it was frequently difficult to access the information it needed.

In addition, the proliferation of logs, metrics, and trace data scattered across the multi-platform, multi-account organization made data analysis and tracking increasingly time consuming and difficult. It slowed down the workflow of Accolade developers, made it increasingly challenging to track service performance, and reduced the time available for innovation and differentiating activities.

Accolade leaders onboarded Elastic because of its unified observability platform. With Elastic, Accolade developers can easily access, navigate, and correlate application and infrastructure data across multiple environments and different AWS services from a single interface. Several AWS services are also pertinent to the acceleration of internal processes at Accolade including Amazon Kinesis, AWS Lambda, Amazon Elastic Container Service (ECS), AWS Fargate, Amazon CloudWatch, Amazon Simple Storage, and more.

With its open common schema, Elastic centralizes data from a variety of formats with ease. Accolade Software Engineer Jose Navarro elaborates: "We are now able to cross-compare data, like log data or metric data with application trace data. This was previously impossible when the data was isolated in different platforms. Before Elastic, our developers were required to check each data system individually. With Elastic, we have everything collected in a single place. They can go there, and they can see how any changes are affecting performance. They can also correlate any changes with the machine learning that’s provided in the platform. It's transforming our organization and making the lives of our developers so much easier.”

Elastic proved to be an ideal solution for Accolade. It simplified and accelerated observability through an aggregated, central viewing pane, while allowing Accolade to fully own and easily access all its data at any time. Navarro shares more on the value of Elastic in this context: “With our previous observability solution, our team had to jump through many hoops but with Elastic, it's very easy because you have access to everything. You own the data. It's your data and your service, your cluster, and you keep it as long as you want to and you delete it if you don't want it.”

Elastic also supports cost efficiency at Accolade by eliminating the pay-per-seat pricing structure required by its previous observability vendor.

The Elastic Observability solution is critical to operations at Accolade. All of the company’s core services and internally developed business applications are currently monitored by Elastic — about 400 services in all.

Elastic makes it easy for the Accolade team to compare and track real-time business impact. One of the core parts of the business is handling telephony. With Elastic, Accolade can measure, for instance, how much delay there is for a patient to wait for a doctor, or for a nurse to connect to the call. They can now measure the impact of performance improvements on this particular branch of the business.

Navarro speaks on Elastic’s contribution to team productivity: “Elastic has been a game changer. Although we are now dealing with more than twice the traffic we had last year, Elastic has freed up our engineers to develop new solutions and new services. With Elastic, we are able to do much more with the same resources.”

The team is utilizing the Elastic Stack (Elastic Agent, Elasticsearch, Logstash, and Kibana) for data collection, storage, aggregation, and visualization. They are also leveraging the OpenTelemetry Jenkins plugin to collect build data for CI/CD pipeline observability. Elastic’s commitment to open standards aligns well with Accolade as a multi-platform, multi-service organization. In addition, centralized management with the unified Elastic Agent has provided a streamlined experience for maintenance and monitoring of data within Kubernetes clusters.

Amazon Kinesis is a key component of the injection pipeline at Accolade. In the event of an unforeseen event that may cause a problem with a cluster, the cluster can easily be replicated elsewhere with 24 hours of data there ready to be ingested and evaluated.

“We're excited to start using Elastic’s AWS Lambda Application Performance Monitoring agent for our cloud-native applications. I especially like the fact that it is possible to see whether a particular invocation of the Lambda function involved a cold start directly in the trace waterfall chart. The availability of Lambda-specific metrics, such as cold start rate, at the service and transaction group levels are also very helpful.”

The integration between Elastic and AWS supports Accolade in maximizing the value of its data. This streamlined access to data insights help Accolade drive an enhanced performance and continued innovation from its tools and solutions. Discover how Elastic and AWS perform better together in this webinar.