About the Customer Ello is an edtech company that combines generative AI, machine learning, education and psychology to create an app-based virtual reading coach for children.
Children read books to the app, and based on their reading level and interests, the AI-powered app provides appropriate material for them to improve their reading skills. Ello is using patent-pending speech recognition and adaptive learning technology to learn speech recognition, which is especially difficult with children.
The Customer Challenge Ello engaged Loka to perform the assessment of workloads to be migrated and modernized to AWS cloud. The scope includes the migration of environments such as machine learning models to Amazon SageMaker with training jobs, GKE kubernetes cluster, Virtual Machines, Memorystore Redis, Cloud Storage data migration, Firebase, Bigquery, Cloud Run, Postgresql Database, MongoDB, App Engine, Cloud Scheduler and Github Runners.
The main drivers for this migration and modernization are: 1. Scalability. To address their increasing need for GPU resources, Ello currently lacks the necessary capacity on Google Cloud Platform (GCP). 2. Availability. To ensure reliable service delivery to their users, Ello must maintain a high level of availability and meet uptime service level agreements (SLAs).
Loka started its partnership with Ello with a MAP proposal. Ello was looking to move from GCP to AWS because they needed increased GPU capacity for their ML models, particularly for training. We identified several needs for Ello in the discovery phase, including control over resource creation and access, implementing infrastructure as code, optimizing costs for compute workloads and storage, acquiring additional GPU capacity, scaling ML models and ensuring security.
Loka concentrated on migrating Ello's architecture from GCP to AWS, emphasizing security, monitoring, GPU capacity and best practices for Ello's environment.
Our migration strategy was to re-platform all GCP services with those that were similar in AWS, refactoring some applications toward new services in AWS to improve scalability, deployment and management.
Test Plan: We implemented functional, integration, performance and security tests, ensuring the validity of the migration process.
DevOps Optimization: All workloads were constructed using Infrastructure as Code (IaC) and followed Ello's CI/CD (Continuous Integration/Continuous Deployment) and SDLC (Software Development Life Cycle) policies.
Cutover and Rollback Plan: We put a safe cutover and rollback plan in place for relevant production workloads in scope or parallel operation of environments until appropriately tested and validated.
DR (Disaster Recovery) and BCP (Business Continuity Plan): The key was to ensure the implementation of Disaster Recovery for all AWS services and Business Continuity Plans.