Intro
A leading hotel company in Croatia - Maistra, turned to us to help them move their Edmond platform to the cloud. With no prior AWS experience, they relied on us to implement the quality solution and simplify the deployment of their platform for new clients.
Edmond is a revenue management system combining property and price management systems with machine learning algorithms based on extensive historical data.
To deploy Edmond for a new client, Maistra needed to define platform requirements, send it to the client’s IT staff and depend on their ability to satisfy requirements. This process was complicated and took a lot of time. To solve these challenges, we helped Maistra adopt the cloud, made Edmond’s deployment fully automated, and made Edmund available in the Software as a Service model.
The Problem
Maistra’s Edmond was a system initially developed as an in-house solution but eventually ended as a full-blown commercial product. In the beginning, Edmond was installed on the client’s on-premise infrastructure. Even though Edmond is a well-built production-ready system, its scalability, security, and stability depend on the client’s ability to satisfy the system’s requirements.
Moving Edmond to the cloud and making it available in the Software as a Service model was a way Maistra wanted to solve infrastructure management problems.
Why AWS cloud?
Maistra had no previous knowledge about AWS cloud, so they approached us to help them set up Edmond on AWS. We examined their needs and processes and offered a technical solution based on more than 10 AWS services.
AWS’s scalability eliminates the need to wait for the client to set up infrastructure for Edmond. AWS’s flexible pricing helps anticipate Edmond’s consumption and communicate it to the end client.
Scalability and Performance
With AWS tools such as Auto Scaling and Elastic Load Balancing, it is easy to scale up or down based on the demand to ensure high performance at every moment.
Cost-Effectiveness
With AWS, clients only pay for the actual resource consumption without long-term contracts or up-front commitments.
Reliability
AWS cloud runs on a scalable, reliable, and secure global computing infrastructure that ensures that critical services run reliably.
Security
AWS utilizes an end-to-end approach to secure and harden the infrastructure, including physical, operational, and software measures.
Solution Description
Serverless
The complete solution is serverless with heavy usage of a couple of components. The solution is now more efficient, flexible, scalable, and easier to deploy.
Machine learning
Managed Airflow service and Amazon Athena save long-term data on S3; this way, we introduced a practically unlimited retention rate for history data. With a lot of historical data, the ML algorithm can predict and calculate accommodation prices much better.
Database
With the implementation of Amazon Aurora, Maistra developed complex business dashboards. The speed and efficiency of Amazon Aurora enable data filtering in real-time, which was not possible before.
Deployment Procedure
CI/CD is implemented with AWS’s Code Pipeline and Code Build tools; the output of the CI/CD procedure is the Docker image stored in the Elastic Container Registry. The Docker image is later used to spawn Fargate tasks.
AWS Well Architected Framework
In the implementation of this solution, we used AWS Well-Architected Framework. The AWS Well-Architected Framework describes key concepts, design principles, and architectural best practices for designing and running workloads in the cloud. It is built around six pillars - operational excellence, security, reliability, performance efficiency, cost optimization, and sustainability.
We know quality matters, and there is no room for errors. Critical IT systems companies depend on must be implemented following best practices and industry standards to ensure everything works as expected - all the time.
Solution Services
- AWS Site-to-Site VPN - VPN tunnels to hotel chains.
- Amazon EC2 - compute capacity for workloads.
- Elastic Load Balancing - distribution of incoming application traffic across multiple targets.
- S3 - Amazon’s simple storage service.
- AWS Fargate - serverless, pay-as-you-go compute engine.
- CodePipeline, CodeDeploy, and CodeBuild - deployment automation tools.
- AWS Aurora PostgreSQL - relational database management system.
- AWS WAF - web application firewall.
- Amazon Elastic Container Registry (ECR) - fully managed container registry for high-performance hosting.
- Cloud Formation - infrastructure as a code solution.
