As follow up to the Exadata as Code post, today I’m going to focus on one of the latest features added to our automation: PDB Snapshot Cloning.
PDB snapshot cloning is one of the best development options to offers in a CI/CD project. In an Exadata environment there are special requirements to implement before start using this technology: Sparse Grid Disks and Sparse ASM Disk Group ( description and step-by-step example is available here).
On Exadata, the PDB Snapshot beneficial of all Smart features including offload capabilities, with in addition space- and time-efficient provisioning.
After this brief introduction let’s see how the PDB Snapshot Cloning has been implemented on our Exadata as Code automation.
Exadata Sparse Storage Automation
Select the Oracle Container Database (CDB) where the PDB Snapshot Cloning should be activated, and with One-clickprovisioning the Sparse Grid disks and Sparse ASM disk group are created.
After the initial provisioning, the automation starts monitoring the space usage, automatically resizing it (increasing/decreasing) when necessary.
PDB Snapshot Cloning Automation
Same principle applies to the different PDB actions: One-click Provisioning/Decommissioning of the PDB Test Master and PDB Snapshot.
Those features are exposed via UI or API to the application developers, making them autonomous on the management of such space efficient database environment.
It is very exciting for me to share this post, because what I’m going to describe here is not the final, but the current and intermediate result achieved by the Trivadis team on the development and implementation of what I call “Exadata as Code” project.
In the Cloud and automation era, this is the Trivadis answer to increase efficiency, time-to-market and quality on the most challenging Exadata projects. Trivadis has been working hard to achieve this level of automation, covering most of the recurrent activities on the platform, and this is not all, as any CI/CD development, it is getting every day better, enriched by new features and fixes, which simplify the lifecycle management of such platforms.
Few months ago, I posted a blog entitled Bulk Exadata Patching, where it is shown how to improve the Exadata patching automation, but despite the improvements, it was still having a number of manual interactions. Now we have reached a much better level of automation, with one-click action to perform cumbersome tasks like: Infrastructure and Database Provisioning/Decommissioning, Patching, and many other operational tasks.
How Exadata as Code works
The concept is quite simple, all developments are made available in the form of Ansible playbooks and incapsulated inside Jenkins; this brings the following advantages:
User friendly interface
Enhanced Security, recording auditing information and logging job executions
Exadata Administration Workflow
Oracle Database Administration Workflow
One takeaway from this experience, automation is the only option to stay competitive and deliver high quality service.
Special thanks go to all Trivadis’ colleagues working everyday so passionately on the project #BetterTogether.