When your business runs on data, even a few seconds of downtime can hurt. That's why this episode focuses on what keeps Oracle Database@AWS running when real-world problems strike. Hosts Lois Houston and Nikita Abraham are joined by Senior Principal Database Instructor Rashmi Panda, who takes us inside the systems that keep databases resilient through failures, maintenance, and growing workloads. Oracle Database@AWS Architect Professional: https://mylearn.oracle.com/ou/course/oracle-databaseaws-architect-professional/155574 Oracle University Learning Community: https://education.oracle.com/ou-community LinkedIn: https://www.linkedin.com/showcase/oracle-university/ X: https://x.com/Oracle_Edu Special thanks to Arijit Ghosh, Anna Hulkower, Kris-Ann Nansen, Radhika Banka, and the OU Studio Team for helping us create this episode. -------------------------------------------------- Episode Transcript:

00:00

Welcome to the Oracle University Podcast, the first stop on your cloud journey. During this series of informative podcasts, we'll bring you foundational training on the most popular Oracle technologies. Let's get started!

00:26

Lois: Hello and welcome to the Oracle University Podcast! I'm Lois Houston, Director of Communications and Adoption with Customer Success Services, and with me is Nikita Abraham, Team Lead: Editorial Services with Oracle University.

Nikita: Hi everyone! In our last episode, we explored the security and migration strengths of Oracle Database@AWS. Today, we're joined once again by Senior Principal Database Instructor Rashmi Panda to look at how the platform keeps your database available and resilient behind the scenes.

01:00

Lois: It's really great to have you with us, Rashmi. As many of you may know, keeping critical business applications running smoothly is essential for success. And that's why it's so important to have deployments that are highly resilient to unexpected failures, whether those failures are hardware-, software-, or network-related. With that in mind, Rashmi, could you tell us about the Oracle technologies that help keep the database available when those kinds of issues occur?

Rashmi: Databases deployed in Oracle Database@AWS are built on Oracle's Foundational High Availability Architecture. Oracle Real Application Cluster or Oracle RAC is an Active-Active architecture where multiple database instances are concurrently running on separate servers, all accessing the same physical database stored in a shared storage to simultaneously process various application workloads.

Even though each instance runs on a separate server, they collectively appear as a single unified database to the application. As the workload grows and demands additional computing capacity, then new nodes can be added to the cluster to spin up new database instances to support additional computing requirements. This enables you to scale out your database deployments without having to bring down your application and eliminates the need to replace existing servers with high-capacity ones, offering a more cost-effective solution.

02:19

Nikita: That's really interesting, Rashmi. It sounds like Oracle RAC offers both scalability and resilience for mission-critical applications. But of course, even the most robust systems require regular maintenance to keep them running at their best. So, how does planned maintenance affect performance?

Rashmi: Maintenance on databases can take a toll on your application uptime. Database maintenance activities typically include applying of database patches or performing updates. Along with the database updates, there may also be updates to the host operating system. These operations often demand significant downtime for the database, which consequently leads to slightly higher application downtime.

Oracle Real Application Cluster provides rolling patching and rolling upgrades feature, enabling patching and upgrades in a rolling fashion without bringing down the entire cluster that significantly reduces the application downtime.

03:10

Lois: And what happens when there's a hardware failure? How does Oracle keep things running smoothly in that situation?

Rashmi: In the event of an instance or a hardware failure, Oracle RAC ensures automatic service failover. This means that if one of the instance or node in the cluster goes down, the system transparently failovers the service to an available instance in the cluster, ensuring minimal disruption to your application.

This feature enhances the overall availability and resilience of your database.

03:39

Lois: That sounds like a powerful way to handle unexpected issues. But for businesses that need even greater resilience and can't afford any downtime, are there other Oracle solutions designed to address those needs?

Rashmi: Oracle Exadata is the maximum availability architecture database platform for Oracle databases. Core design principle of Oracle Exadata is built around redundancy, consisting of networking, power supplies, database, and storage servers and their components.

This robust architecture ensures protection against the failure of any individual component, effectively guaranteeing continuous database availability. The scale out architecture of Oracle Exadata allows you to start your deployment with two database servers and three storage servers, having different number of CPU cores and different sizes and types of storage to meet the current business needs.

04:26

Lois: And if a business suddenly finds demand growing, how does the system handle that? Is it able to keep up with increased needs without disruptions?

Rashmi: As the demand increases, the system can be easily expanded by adding more servers, ensuring that the performance and capacity grow with your business requirements. Exadata Database Service deployment in Oracle Database@AWS leverages this foundational technologies to provide high availability of database system. This is achieved by provisioning databases using Oracle Real Application Cluster, hosted on the redundant infrastructure provided by Oracle Exadata Infrastructure Platform.

This deployment architecture provides the ability to scale compute and storage to growing resource demands without the need for downtime. You can scale up the number of enabled CPUs symmetrically in each node of the cluster when there is a need for higher processing power or you can scale out the infrastructure by adding more database and storage servers up to the Exadata Infrastructure model limit, which in itself is huge enough to support any large workloads.

The Exadata Database Service running on Oracle RAC instances enables any maintenance on individual nodes or patching of the database to be performed with zero or negligible downtime. The rolling feature allows patching one instance at a time, while services seamlessly failover to the available instance, ensuring that the application experienced little to no disruption during maintenance.

Oracle RAC, coupled with Oracle Exadata redundant infrastructure, protects the Database Service from any single point of failure. This fault-tolerant architecture features redundant networking and mirrored disk, enabling automatic failover in the event of a component failure. Additionally, if any node in the cluster fails, there is zero or negligible disruption to the dependent applications.

06:09

Nikita: That's really impressive, having such strong protection against failures and so little disruption, even during scaling and maintenance. But let's say a company wants those high-availability benefits in a fully managed environment, so they don't have to worry about maintaining the infrastructure themselves. Is there an option for that?

Rashmi: Similar to Oracle Exadata Database Service, Oracle Autonomous Database Service on dedicated infrastructure in Oracle Database@AWS also offers the same feature, with the key difference being that it's a fully managed service. This means customers have zero responsibility for maintaining and managing the Database Service.

This again, uses the same Oracle RAC technology and Oracle Exadata infrastructure to host the Database Service, where most of the activities of the database are fully automated, providing you a highly available database with extreme performance capability. It provides an elastic database deployment platform that can scale up storage and CPU online or can be enabled to autoscale storage and compute.

Maintenance activities on the database like database updates are performed automatically without customer intervention and without the need of downtime, ensuring seamless operation of applications.

07:20

Lois: Can we shift gears a bit, Rashmi? Let's talk about protecting data and recovering from the unexpected. What Oracle technologies help guard against data loss and support disaster recovery for databases?

Rashmi: Oracle Database Autonomous Recovery Service is a centralized backup management solution for Oracle Database services in Oracle Cloud Infrastructure.

It automatically takes backup of your Oracle databases and securely stores them in the cloud. It ensures seamless data protection and rapid recovery for your database. It is a fully managed solution that eliminates the need for any manual database backup management, freeing you from associated overhead.

It implements an incremental forever backup strategy, a highly efficient approach where only the changes since the last backup are identified and backed up. This approach drastically reduces the time and storage space needed for backup, as the size of the incremental changes is significantly lower than the full database backup.

08:17

Nikita: And what's the benefit of using this backup approach?

Rashmi: The benefit of this approach is that your backups are completed faster, with much lesser compute and network resources, while still guaranteeing the full recoverability of your database in the event of a failure. You can achieve zero data loss with this backup service by enabling the real-time protection option, while minimizing the data loss by recovering data up to the last subsecond.

It is highly recommended to enable this option for mission-critical databases that cannot tolerate any data loss, whether due to a ransomware attack or due to an unplanned outage. The protection policy can retain the protected database backups for a minimum of 14 days to a maximum of 95 days.

The recovery service requires and enforces the backups are encrypted. These backups are compressed and encrypted during the backup process. The integrity of the backups is continuously validated without placing a burden on the production database.

This ensures that the stored backup data is consistent and recoverable when needed. This protects against malicious user activity or any ransomware attack. With strict policy-based retention strategy, it prevents modification or deletion of backup data by malicious users.

09:30

Lois: Now, let's look at the next layer of protection. Rashmi, can you tell us about Oracle Active Data Guard?

Rashmi: Oracle Active Data Guard provides highly available data protection and disaster recovery for Enterprise Oracle Databases. It creates and manages one or more transactionally consistent standby copies of production database, which is the active primary.

The standby database is isolated from production environment located miles away in a distance data center, ensuring the standby remains protected and unaffected, even if the primary is impacted by a disaster.

In the event of a disaster or data corruption occurring at the primary, the standby can take over the role as new primary, thus allowing business to continue its operations uninterrupted. It keeps the standby database in sync with the production database by continuously applying change logs from production.

10:25

Do you want to stay ahead in today's fast-paced world? Check out our New Features courses for Oracle Fusion Cloud Applications. Each quarter brings new updates and hands-on training to keep your skills sharp and your knowledge current. Head over to mylearn.oracle.com to dive into the latest advancements!

10:45

Nikita: Welcome back! Rashmi, how does Oracle Active Data Guard operate in practice?

Rashmi: It uses the knowledge of Oracle Database block format to continuously validate physical blocks or logical intrablock corruption during redo transport and change apply. With automatic block repair feature, whenever any corrupt block is detected in the primary or the standby database, then it is automatically repaired by transferring a good copy of the block from other destination that holds it. This is handled transparently without any error being reported in the application.

It enables you to upload the read-only workloads and backup operations to the standby database, reducing the load on the production database. You can achieve zero data loss at any distance by configuring a special synchronization mechanism known as parsing.

File systems form the attack surface for ransomware. Since Active Data Guard replicates the data at the memory level, any ransomware attack on the primary database will never be replicated to the standby database. This allows for a safe failover to the standby without any data loss, and shielding the database from effects of the attack.

You can enable automatic failover of the primary database to a chosen standby database without any manual intervention by configuring a Data Guard Broker. The Data Guard Broker continuously monitors the primary database and automatically performs a failover to the standby when the predefined failover conditions are met. Active Data Guard enables you to perform database maintenance or database software upgrades with almost zero or minimal downtime.

12:18

Lois: And how does disaster recovery work for Exadata Database Service in Oracle Database@AWS?

Rashmi: Exadata Database Service, by design, are already protected against local failures by use of technologies like Oracle RAC and Oracle Exadata.

Now, by deploying Exadata Database Service across multiple availability zones in an AWS region, it can ensure that your database services remain resilient to site failures. It leverages Oracle Active Data Guard to create standby in a separate availability zone such that if the primary availability zone is affected, then all application traffic can be routed to the database services in the secondary availability zone, restoring business continuity of the application back to normal.

Through continuous validation of the data blocks at both the primary and the standby database, any potential corruption is detected and prevented. This ensures data integrity and protection across the entire database service.

By leveraging zero data loss Autonomous Recovery Service, the database ensures that the backup remains secure and unaffected by ransomware. This enables rapid restoration of clean, uncompromised data in the event of an attack.

Periodic patching and upgrades are performed online in a rolling fashion with little to no impact on the application uptime using a combination of Oracle RAC and Oracle Active Data Guard technologies. For all resource-intensive workloads like database backup or generating monthly reports, which are read-only in nature, they can be uploaded to the standby, reducing the load on the production database.

In the cross-availability zone DR setup, you have the flexibility to configure Active Data Guard to use either the AWS network or the OCI network for keeping database redo logs to the standby database.

Choosing which network to use for the traffic is entirely at the enterprise discretion. However, both are Oracle maximum availability–compliant and the setup is pretty simple. If the network traffic being used is OCI network or AWS network, then respective cloud provider is responsible for ensuring the reliability.

You have to take into account the different charges that each cloud provider may have. And you can provision multiple standby databases using the console. Optionally, you may set up a broker manually to enable automatic failover capability.

14:30

Nikita: We just covered cross-availability-zone protection. But what if an entire AWS region goes down?

Rashmi: This is where we can provide an additional level of protection by provisioning cross-region disaster recovery for your Exadata Database Service in Oracle Database@AWS.

This deployment protects your database against regional disasters. You can provision another DR environment in a different AWS region that supports Oracle Database@AWS. This deployment, together with the cross-availability zone deployment, complements your highly available and protected database service deployment in Oracle Database@AWS.

Under the hood, it uses the same Oracle Database technologies that include Oracle Active Data Guard, OCI Autonomous Recovery Service, Oracle Exadata, Oracle RAC to provide the same capabilities as in case of cross-availability zone deployment.

Here too, you have the flexibility to configure Oracle Active Data Guard to use either AWS network or OCI network for shipping database redo logs to the standby. And for the network traffic options, the feature remains the same, except a small difference with respect to chargeback.

When using OCI Network for cross-region deployment, there is no charge for the first 10 TB of data transfer per month. Beyond that, standard OCI charges would apply. When using AWS network, you may refer to AWS charging sheet for the cross-region traffic.

15:49

Nikita: Thank you so much, Rashmi, for this insightful episode.

Lois: Yes, thank you! And if you want to dive deeper into the topics we covered today, go to mylearn.oracle.com and search for the Oracle Database@AWS Architect Professional course. Until next time, this is Lois Houston…

Nikita: And Nikita Abraham, signing off!

16:13

That's all for this episode of the Oracle University Podcast. If you enjoyed listening, please click Subscribe to get all the latest episodes. We'd also love it if you would take a moment to rate and review us on your podcast app. See you again on the next episode of the Oracle University Podcast.