AWS Cloud Database Users Detail Applications

Rather than creating a single database service for all needs, over its 15 year history, AWS has launched multiple cloud database services with the goal of delivering the feature set that organizations want. .

In a session on Database Leadership at the AWS re: Invent 2021 conference on December 1, Jeff Carter, Vice President of Relational Databases at AWS, detailed the new capabilities and outlined the overall strategy for cloud databases.

AWS introduced new data lake capabilities this week with its Lake formation service to improve governance and management of data lakes. AWS also launched its new DevOps Guru for RDS service, which provides a machine-learning-based capability to Amazon Relational Database Service optimization.

AWS is also emphasizing the topic of database migration with the launch of its AWS Database Migration Server Fleet Advisor system.

During this time, Sundar shankar, Head of Service and Cloud Reliability Engineering (SRE) at online insurance marketplace provider eHealth, detailed the migration process to AWS cloud databases.

How eHealth Migrated to AWS Cloud Databases

Shankar said eHealth migrated to AWS cloud databases in 2021 in a process that took 11 months of effort. He noted that before the migration, his team had expected it to take several years to migrate.

Sundar Shankar, Head of Cloud and SRE at eHealth, explained how his organization migrated to AWS cloud databases during a leadership session at AWS re: Invent.

The migration of the eHealth cloud database involved a three-phase strategy.

In the first phase, eHealth sought to “lift and move” its on-premises MySQL and PostgreSQL database deployments to AWS. The goal of the initial phase was to identify potential gaps and then start formulating a plan to optimize the databases.

If you are wondering why we chose these two databases [Aurora and DocumentDB], they’re already best-in-class, decouple storage and compute, provide automated recovery and failover, automated backups, and security.

Sundar shankarHead of Cloud Engineering and Service Reliability, eHealth

In the second phase, eHealth began to consolidate its databases by moving to Amazonian Aurora for its MySQL and PostgreSQL needs. Shankar noted that eHealth also uses MongoDB and migrated to Document database on AWS for this database.

“If you are wondering why we chose these two databases [Aurora and DocumentDB], they’re already best-in-class, decoupling storage and compute, providing automated restore and automated failover backups as well as security, ”Shankar said.

The final phase of the migration focused on ensuring high availability and resilience. Shankar explained that during the third phase, eHealth developed its global replication capabilities for cloud database deployments, which was essential for eHealth to complete the global migration in 11 months.

How HBO Max Uses AWS Cloud Database Services

In the same leadership session, Rob ford, director of software engineering at HBO Max, explained how the online streaming service uses multiple AWS cloud databases for its operations.

Ford noted that a key challenge for HBO Max is the ability to handle traffic demands as the number of users and the amount of content increase. The automation and scalability options provided by the AWS cloud database have been a real plus for HBO Max, he said.

“We believe in the discipline of database reliability engineering, which focuses on heavy investments in automation,” said Ford.

Ford said HBO Max chose different AWS cloud database services to meet different needs. For its purchasing power, which allows the consumer to purchase subscriptions and content, HBO Max uses Amazon RDS for PostgreSQL.

“The purchasing database is where we capture direct consumer purchasing information and it’s our crown jewel,” said Ford. “This data can be accessed by the HBO Max app in a number of ways, making indexing support critical, so the native PostgreSQL functionality was ideal for this use case.”

Meanwhile, HBO Max uses the Amazon ElastiCache service for its device database.

The Device Database is where HBO Max monitors and manages the status of devices connected to its network. The streaming provider then uses the DynamoDB database for what Ford has called the “pickup” service. He explained that the pickup allows users to pick up where they left off after watching part of a series or episode.

This year “has been a good year for HBO Max,” Ford said. “In just a few months, we went from operating in one country to several dozen. I think a lot of this is due to the capabilities that come with Amazon databases.

The AWS conference session was held live in Las Vegas and replayed online.

Maria H. Underwood