Reducing Server and Database Costs by 50% for an Insurance Broker using AWS
Expertise Delivered
Industry
Client Brief
Our client, HealthPro Insurance, is a premier Medicare insurance broker, ranked among the top 5 in the United States by Healthcare.gov. Their comprehensive offerings include specialized support for health insurance, Dental, Vision, and long-term life insurance, ensuring their clients receive the best possible care and coverage.
HealthPro provides valuable subsidies to employers, individual consumers, and nonprofits. They aim to enhance the accessibility and provision of insurance for all US citizens.
Project Scope
Our client managed their website and peripheral services on legacy processes and infrastructure. Their unoptimized processes and abrupt resource investments made them pay through their nose.
They wanted to introduce automation with DevOps practices to optimize their system’s performance, flexibility, efficiency, and data security. HealthPro’s project scope revolves around the following attributes.
- Distribute server load and processes across multiple domains.
- Switch to private cloud from public to prevent cyber attacks like DDoS, resulting in data breaches.
- Creating two different environments for production and staging can mitigate the risks of compromised user experiences when experimenting with or upgrading systems.
The Challenge
Due to hosting all services in a single public cloud and relying solely on credential protection, our client faced server overloading and security vulnerabilities. This also increased the platform’s instability, posing limitations with hosting the production and staging environments. The above issues compromised the performance of their services, increased costs, and created data risk, hindering their accessibility and scalability.
Another prime challenge was switching all their domains and sub-domains to AWS and implementing autoscaling to optimize costs.
Our client’s services used MariaDP as a database, and a significant challenge was to plan a systematic switch from MariaDP to MySQL while keeping their data intact.
Why Maruti Techlabs?
One of our consultants, who was acquainted with our client's challenges and with our expertise, advocated for our services. Subsequently, the consultant and the client's Chief of Sales, who served as the primary point of contact, designed a questionnaire to evaluate our approach.
After thoroughly understanding the client's expectations through the questionnaire, our team of experts brainstormed various solutions. We developed a detailed workflow outlining the implementation of these solutions, focusing on enhancing security, improving stability, and optimizing costs.
Our proven DevOps expertise in server administration and maintenance, implementing and maintaining CI/CD pipelines, cloud migration, integrating DevSecOps, and more, supported by the consultant's confidence in our capabilities, secured the project for us. The proposed workflow and solutions demonstrated our ability to meet the client's needs effectively, ultimately earning their trust and business.
Solution
A phase-wise approach was chosen to implement the required solutions to ensure uninterrupted services for the client's customers. Here's how we implemented the perfect tech infrastructure and optimized costs for our client:
1. Isolating Staging and Production Environment
Firstly, we isolated the whole(one) unit into two different environments for ‘Production’ and ‘Staging.’ This helped us offer a seamless customer experience in production, allowing our client to experiment with new features, load and regression testing, and introducing new upgrades with security in staging.
2. Transitioning from Public to Private Cloud
Our next step was building the staging environment by creating a new Virtual Private Cloud (VPC) with private and public subnets.
3. Creating Data Back-Up
We then transferred the database from the old VPC’s public subnet to the new VPC’s private subnet. We did this by taking a snapshot or backing up the data, creating a copy in the new VPC, and restoring it to the private subnet.
Additionally, to allow the server to access the database seamlessly, we assigned roles to the EC2 server and upgraded its database security rules.
4. Cost Optimization
As a final step, we optimized costs for our client by implementing automatic scaling with AWS’s Aurora Serverless. Aurora Serverless offers the convenience of auto-scaling CPU and memory usage based on traffic. For instance, if our clients don’t have weekly traffic, they do not need to pay for that period.
We achieved this by migrating from MariaDB to MySQL, as Aurora supports both database engines. We then transferred the migrated data to Aurora Serverless.
Your product idea. Our engineering prowess & seamless execution.
Communication and Collaboration
We conducted a one-week discovery phase during which our team members communicated extensively to understand their current solution's limitations and expectations concerning their desired solution.
We allocated two team members, one DevOps manager and one project manager, to execute this project’s specifics end-to-end. For the entirety of the project, our primary point of contact was our client’s ‘Chief of Sales.’
Here are the platforms we majorly use for communication:
- Slack - Quick communication,
- Google Meet - Daily updates and queries, roll-outs
- Email - Official channel to share contracts, weekly updates, and more.
Technology Stack
Our Development Process
We follow Agile, Lean, & DevOps best practices to create a superior prototype that brings your users’ ideas to fruition through collaboration & rapid execution. Our top priority is quick reaction time & accessibility.
We really want to be your extended team, so apart from the regular meetings, you can be sure that each of our team members is one phone call, email, or message away.
