Oracle University reduces administration costs by 30% with Autonomous Database
Two Oracle University applications, including one that migrated from PostgreSQL, rely on Autonomous Database to improve performance, scalability, and security.
“Oracle University uses Oracle Autonomous Database for its Guided Learning and MyLearn commercial applications to improve performance, scalability, and security while delivering zero downtime and automated database maintenance. We have reduced administration costs by 30%.”
Business challenges
Oracle University helps customers build proficiency with Oracle Cloud Infrastructure (OCI) and SaaS applications. Oracle Guided Learning is integrated with Oracle Cloud Applications, providing up-to-date guidance and knowledge in the user’s workflow.
Oracle MyLearn offers subscription-based, personalized training on a range of Oracle solutions, including Oracle Cloud Infrastructure (OCI), Oracle Cloud Applications, and database and industry solutions.
Oracle Guided Learning incorporated technology that was acquired by Oracle in 2018, and migrated from AWS. The underlying database used on AWS was PostgreSQL, which is available on OCI. However, as part of the migration, the development team also wanted to address some PostgreSQL shortcomings. In particular, performance optimization was poor, while building an HA setup required substantial expertise. The required additional compute resources forced downtime while users had to manage the underlying block storage that PostgreSQL used. The resulting administrative overhead and required expertise was simply too high.
Why Oracle chose Autonomous Database
The Oracle University development team selected Oracle Autonomous Database for transaction processing and to replace PostgreSQL for Oracle Guided Learning. The solution supports developers who create new guides as well as end users who interact with those guides from host applications.
In addition, Autonomous Database for analytics and data warehousing supports administrative users who can look at the entire learning environment to understand how it’s being used.
Finally, Oracle MyLearn was a new application first built on Autonomous Database for transaction processing and mixed workloads.
Developers anticipated reduced maintenance costs due to several factors. First, Oracle Autonomous Database automatically manages storage and allows additional storage or compute resources on demand without needing specialized expertise or taking the database down. Backups, performance tuning, patching, and updates are all handled automatically, further reducing operational costs.
Autonomous Database comes with security policies and settings already in place, which reduced the work to keep data secure, and also decreased the risk of errors.
Autonomous Database also simplified the process of integrating with object storage, which is an integral part of the overall solution.
Results
Today Oracle Guided Learning and Oracle MyLearn are implemented using a lakehouse architecture. About 8 TB of application code, user data, and metadata are stored in Oracle Autonomous Database, while 10 TB of images, multimedia content, and log data are kept in object storage.
The complete solution includes other services, including use of Elasticsearch for indexing data to speed searches and Oracle Analytics Cloud for reporting.
Behind the scenes, developers also use OCI Logging, OCI Logging Analytics, Oracle Application Performance Management, OCI Monitoring, OCI Certificates, OCI Vault, and OCI Email Delivery.
Setting up an HA configuration, which used to be a two-week operation, now takes just a few minutes with Autonomous Data Guard. Furthermore, the work can be done by employees who don’t have database expertise, and there is no ongoing maintenance—providing further savings.
Adding storage capacity is now a simple API call, requiring no downtime, no manual storage management, and no special expertise. Adding more compute resources if required is equally simple. Developers don’t use autoscaling in their production environment, but have found it useful for new versions or first installs, when the exact performance characteristics are not well known.
Built-in security capabilities saved the development team many hours in developing scripts to secure the database, and there is no ongoing work needed to maintain the scripts and configuration information.
Autonomous Database offers simple, direct access to object storage. This has enabled the team to automate database refreshes, which now take only 2 to 3 hours, rather than a whole week.
Overall, administering Autonomous Database requires significantly less time and expertise. The development team estimates this reduces overall administration costs by at least 30%.