Operating Oracle Database@AWS - Management, Monitoring and Troubleshooting
This follow-up episode focuses on the operational aspects of Oracle Database@AWS. We cover: - Day-to-day management tasks for ODB networks, Exadata infrastructure and VM clusters - Monitoring capabilities using CloudWatch metrics and EventBridge - Security and compliance features including IAM integration - Migration
In the second part of this series, we are diving deep into the operational aspects of Oracle Database at AWS. The episode provides a comprehensive walkthrough of provisioning, management, and monitoring capabilities for this marketplace offering that brings Oracle Exadata and Autonomous Database services into AWS data centers. Joe demonstrates the complete provisioning workflow, from creating ODB networks with specified CIDR ranges through deploying Exadata infrastructure and VM clusters, emphasizing that VM cluster creation is the most time-consuming step, taking 4-5 hours for a two-node cluster.
The show covers extensive connectivity options and AWS service integrations that make Oracle Database at AWS feel like a native AWS service. Connectivity can be established through ODB peering for direct VPC connections, Transit Gateway for multi-account and multi-region scenarios, or Cloud Wan for global AWS footprint access. The service integrates seamlessly with CloudTrail for API logging and auditing, CloudWatch for performance metrics at VM cluster, tablespace, and database levels, and Amazon S3 for automated backups that can remain within AWS to meet compliance requirements. Additionally, the hosts highlight powerful integrations like zero-ETL replication to Amazon Redshift for analytics, Bedrock integration for AI capabilities, and the DBMS_CLOUD package for direct S3 file access from Oracle databases.
Database creation and management differ between the two offerings: Exadata databases support versions 19c and 23 AI with SSH access to underlying VMs for direct management, while Autonomous databases require creating an Autonomous Container Database first and operate as fully managed services without SSH access. The episode also covers migration strategies, distinguishing between physical migration options (RMAN, Data Guard) and logical migration tools (GoldenGate, AWS DMS, Data Pump) for Exadata databases, while Autonomous databases only support logical migration approaches. The Zero Downtime Migration (ZDM) tool can orchestrate both migration types for minimal disruption.
Key Highlights:
- Provisioning workflow includes creating ODB networks, deploying Exadata infrastructure in specific availability zones, and creating VM clusters (4-5 hours for two-node clusters)
- Database creation options include Exadata databases (19c or 23 AI) with SSH access and Autonomous databases (Data Warehouse or Transaction Processing) as fully managed services
- Multiple connectivity patterns via ODB peering, Transit Gateway, Cloud Wan, and VPC Lattice for secure access across accounts and regions
- DNS resolution configured through Route 53 outbound resolver endpoints for seamless name resolution
- Comprehensive AWS integrations with CloudTrail, CloudWatch, S3, Redshift (zero-ETL), and Bedrock for AI capabilities
- Flexible backup destinations including Amazon S3 (recommended for AWS compliance), OCI Object Storage, and Autonomous Recovery Service
- Scaling capabilities supporting up to 32 DB servers and 64 storage servers as demand grows
- Oracle Database 23 AI features including vector store and vector search capabilities for generative AI applications
- Migration tool options including physical methods (RMAN, Data Guard) for Exadata and logical methods (GoldenGate, DMS, Data Pump) for both, orchestrated by ZDM
- DBMS_CLOUD package enables direct access to S3 files from Oracle databases without intermediate ETL processes
Hosts of the show 🎤
Ibrahim Emara - Database Specialist SA @AWS
Guests 🎤
Jobin Joseph - Database Specialist SA @AWS
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article