Migration of a high-load online service DOC.ONLINE to AWS

Alexander Abgaryan
CEO, Senior DevOps
Migration of a high-load online service DOC.ONLINE to AWS

Migration of a high-load online service to AWS

DOC.online is a rapidly growing, high-load online service that allows to book appointments with medical professionals.

It successfully operates in several countries, including Ukraine, Kazakhstan, Uzbekistan, Moldova, and Georgia.

The service also provides the opportunity to book a medical test appointment, and to connect with drug stores for ordering medications.

Doc.online is cooperating with more than
  • 1 700 clinics
  • 2 000 drug stores
  • 500 medical labs

There are more than 11 000 doctors available in the database, and about 300 000 patient feedback entries.

Over 2 000 appointments are booked every day using the service

It's not a new project, it has been around for many years operating in Ukraine, and we like IT support company has been happy to provide DevOps services to the company.

However, Doc.online started its rapid development in 2020, and added a number of new features to the service, a convenient mobile app, expanded its geography to other countries, and is continuing to grow.

Because of all these changes, more server resources were required to support its perfect functioning.

The problems that arose regarding the hosting, with the growth of the service:

  1. Not able to add server resources rapidly as needed
  2. Not able to reach high uptime and quick service response to the growing number of user requests with the current system architecture
  3. Not able to build the needed architecture using the current web hosting
  1. The decisions that were made:
  2. Choose AWS as a new hosting
  3. Build highly-scalable architecture for the project and perform the migration

What we did:

1. Designing and building the architecture

For each separate country service of doc.online, we have built a highly-scalable architecture on AWS. The architecture was created with microservices. The system allows to rapidly gain resources when needed, and is scailing automatically depending on the user activity load.

2. Ensuring security and fault-tolerance

We made sure the architecture is secure from hacker and DDoS attacks, and is fault-tolerant. We set up a professional monitoring to see and eliminate the upcoming problems.

3. Cost optimization

We were building the system with cost optimization in mind, and afterwords adjusted the system even more, so it could bring more profit to our clients.

4. Testing

We tested the system thoroughly

5. Performing the migrations with zero downtime

The services were moved to the new AWS architecture without the downtime.

The result:

Using highly-scalable cloud architecture on AWS, the service is operating smoothly now, showing short response time even in the hours of high user activity.