IBM’s Planning Analytics version 12 introduces a new feature known as high availability. It enables you to manage multiple database replicas in parallel, thereby enhancing both performance and reliability.
Revelwood’s IBM Planning Analytics Practice Director Lee Lazarow explains high availability in one of our IBM Planning Analytics Tips & Tricks videos.
How High Availability Works
Load balancing and fault tolerance is at the core of high availability. When a user submits a request—be it updating a data point, executing a TurboIntegrator script, or conducting some other processing—the request is directed to one of the database replicas, known as the leader replica.
- 1. Role of the Leader Replica:
- The leader executes all required changes, performs calculations, and manages processing tasks.
- Once the leader completes these tasks successfully, it shares/sends the changes to all other replicas, ensuring that they are kept in sync.
- 2. Load Distribution:
- By distributing user requests across multiple replicas, the system mitigates the risk of a single heavy processing task impacting overall performance. This load spreading enhances the user experience, particularly in environments with fluctuating demands.
- 3. Active Management of Replicas:
- The engine takes an active role in monitoring and managing the health of each replica. If the leader becomes unavailable for any reason, another instance is automatically promoted to take over its responsibilities, ensuring continuity of service.
- 4. Dynamic Scaling:
- One of the standout features of high availability is the ability to dynamically adjust the number of replicas based on current demands—all without requiring downtime. This flexibility is particularly useful for organizations with varying usage patterns:
- During peak periods (e.g., month-end closings, budgeting cycles), you can scale up the number of replicas to accommodate increased user activity.
- Conversely, during quieter periods, you can scale down, optimizing resource usage and costs.
- One of the standout features of high availability is the ability to dynamically adjust the number of replicas based on current demands—all without requiring downtime. This flexibility is particularly useful for organizations with varying usage patterns:
Managing High Availability
Managing high availability in Planning Analytics is straightforward and user-friendly. Here’s a brief overview of how to configure and monitor this feature:
- 1. Accessing the Management Interface:
- Within the administration interface, you can easily enable or disable high availability by selecting the relevant checkbox.
- 2. Configuring Replicas:
- You can define the number of replicas you wish to create. As you increase the number of replicas, the system automatically adjusts memory usage based on your configuration. For instance:
- If the original database requires 4 GB of memory, adding a replica would require an additional 4 GB.
- You can define the number of replicas you wish to create. As you increase the number of replicas, the system automatically adjusts memory usage based on your configuration. For instance:
- 3. Real-time Monitoring:
- The system provides real-time feedback on memory usage and warns you if you attempt to allocate more replicas than your available resources can support. This proactive management ensures that you maintain optimal performance without overloading the system.
IBM Planning Analytics version 12’s high availability feature is an advancement in how organizations can manage their Planning Analytics environments. Users achieve greater efficiency, reliability and scalability by leveraging multiple replicas.
Revelwood is an IBM Gold Business Partner with 25+ years of experience designing, developing, implementing and maintaining IBM Planning Analytics environments. Revelwood has helped clients in all sizes across all industries optimize and grow their use of Planning Analytics. Revelwood’s Planning Analytics team consists of experienced PA experts, including a multi-year IBM Champion.
Stay up to date with PA – sign up for our weekly Planning Analytics Tips & Tricks newsletter, subscribe to our YouTube channel, and join our IBM Planning Analytics All-Stars group on LinkedIn.
Read more IBM Planning Analytics Tips & Tricks:
IBM Planning Analytics Tips & Tricks: ASCIIOUTPUT
IBM Planning Analytics Tips & Tricks: Upcoming Changes to Planning Analytics
IBM Planning Analytics Tips & Tricks: PAW Explorations and MDX
