TalentLMS is a web-based SaaS LMS solution offered by Epignosis, a company that specializes in e-learning software. The solution is used by more than 70k organizations across many industries around the globe, with millions of daily users.
Stackmasters helps Epignosis’ teams to manage the infrastructure of their 3 distinct product lines.
Problem Definition
The TalentLMS team, following its strategic direction to move to Amazon Web Services (AWS) to better support smooth scaling, required a migration of its whole infrastructure. The cloud instances under scope were numbered to a few hundreds, initially residing in Rackspace cloud.
To further optimize their business workflow, they considered the potential benefits that a richer cloud platform like AWS would enable.
Moreover, a significant boost on deployments times also required faster on-boarding process for each new client, from a few days to a few minutes.
Another significant concern was better and more detailed monitoring, that could bring into attention service performance insights.
Proposed Solution & Architecture
Stackmasters planned and implemented the migration of TalentLMS production infrastructure to AWS, working along with their teams. The goal was to efficiently architect and build a more robust platform for current and future needs.
The core workload platform is based upon Amazon Elastic Compute Cloud (Amazon EC2) instances, as that is the place where the most application logic resides by software design. Different Amazon EC2 instance types were selected based on performance metrics available, taking into account the role of the software components that will be hosted in the solution. Elastic Load Balancing was selected to provide public facing access to the end-users.
For the storage layer Amazon RDS, Amazon S3 and Amazon EFS are utilized in the solution.
Every EC2 instance is placed inside an Amazon Virtual Private Cloud (Amazon VPC) that spans across different Availability Zones, with access provided to the teams through AWS Client VPN services.
AWS organizations were utilized to provide isolated access to multiple accounts among the different product teams for developers and operators.
The solution is completed with Route 53 to manage public DNS records, Amazon CloudFront to distribute static web assets and AWS Key Management Service (AWS KMS) to store secrets used in the platform.
The deployment was automated with Ansible, which handles the whole environment. Everything from AWS resources provisioning and allocation to the application stack services deployment and configuration management are orchestrated with the Stackmasters Ansible-based toolkit.
Results & Outcomes
The migration to the AWS platform was delivered ahead of time, following a validation of the solution and user acceptance test phase from TalentLMS teams. The end-user experienced no downtime!
The deployment of new resources ready to extend the platform is now at least 5x faster.
The performance boost for the whole platform is more than 2x time higher, as a result of the rich instance types available in AWS.
Overall, the reliability of the hosting system for the workload under scope has increased, which allowed for the business to achieve faster scaling with smoother user experience.