Are you ready to learn best practices and available tools to optimize your Oracle HCM Cloud environment? As part of Quest Forum Digital Event: Innovation Week, Oracle’s own John Cafolla (VP, Oracle HCM Cloud Center of Excellence), Timothy Huth (Applications Developer, Oracle HCM Cloud Center of Excellence), and Gary Payton-McDowall (Principal Product Manager/Strategy, Oracle HCM Cloud Center of Excellence), presented how you can best optimize Oracle HCM Cloud solution.
To begin, it’s important to understand how HCM Cloud works. Oracle provides software and service infrastructure to deliver innovation, but customers manage their own functional and technical setups and data. Oracle provides appropriate instances tailored to your size and standard operating needs. These instances are based on a template from experiences with thousands of other customers and can be reviewed and adjusted for exceptions.
In many cases, customers make the system more complicated than necessary. Oracle offers performance tips to help customers effectively configure services to their own operations.
Customers can leverage best practices to adapt to specific needs with HCM configuration, Cloud Service operations, and Cloud infrastructure integration, performance, and monitoring.
Together, Oracle and customers can deliver:
- Functional Configuration of Standardized Business Processes
- Technical Configuration of Standardized Extracts, Interfaces, Business Rules, Analytics
- Operational Configuration of Standardized Online and Batch Processes for Payroll, Benefits, Time, Interfaces to third-party Payrolls, Extracts to Analytics/DW
In regard to provisions and monitoring, here’s how roles are broken down between you, the customer, and Oracle.
Oracle: Standard Service
- Provision the number of pods you request based on what is known about you (defaults)
- Pre-size based on general volume usage for similar customers
- Adapt sizing to your Operational Profile when appropriate
- Monitor for exceptional load and outages
- Publish usage info to your Cloud Portal
You: Adapt to Use
- Maintain instance management plan for implementation, stabilization, and optimization
- Confirm your standard operating profile and user procedures
- Request ops performance check or online load testing when appropriate
- Monitor for standard system loading and managed exceptions
- Maintain your own metrics and dashboard
On the subject of monitoring and action, you can monitor your own usage with RUEI Reports and Enterprise Manager Metric Reports. RUEI Reports include various page load performance metrics. EM Reports show ESS jobs and identify run times for these processes.
In deciding how to utilize your infrastructure, you need to know what’s running on your system. To understand this, ask yourself the following questions:
- Are your mission-critical jobs being hampered by ad hoc jobs outside of the run book?
- Need visibility into what is running or in the queue before you submit that critical job?
- Need to know why jobs are taking longer than expected? (are they in a queue?)
- Need help identifying who or what is running outside of the schedule?
- Need this information for all your HCM applications? (Payroll, Benefits, Absence, Compensation, Time, etc.)
- Need to know if your environment is at or beyond capacity?
You can use the Active Processes Dashboard in OTBI for a real-time view of the load on your system. Additionally, you can extract data with HCM Extract. This expert-level tool requires training and experience, so Oracle now offers templates to help customers extract effectively.
When considering performance for operations and deployment, you can batch processing with ESS. Examples of ESS jobs include HDL, HCM Extracts, Payroll, Time Processing, Absence, Keyword Search, and any scheduled or batch process. All HCM ESS jobs use the same pool of resources.
During Batch Processing, the application invokes ESS Batch job. ESS then spawns jobs that call back into the application code to complete the work. For some tasks, application code or ESS calls into BI engine. This is important because any batch job will take resources from the global pool. When planning your daily workload, you should take all activities into account. Lastly, during performance base-lining, you should start by testing jobs in isolation to make sure nothing else happens on the system to affect your baseline numbers.
6 Steps for Oracle HCM Cloud Optimization Standards
To keep it simple, you’ll want to snap back to standards again and again. Here’s a checklist to follow:
- Check the basics first.
- Configuration: use of sql inside app features
- Service: number of pods, production sizing, browser & settings
- Operations: operational run books, performance testing, online load testing, regression testing
- Use standard configuration features.
- Avoid complexity of custom solutions with SQL ‘code.’
- Avoid complex expression language rules in the UI to control LOVs, entries, and defaults based on multiple parameters or context from the user role.
- If you must…then test for performance impact and check new features with each update
- Check security profiles, number of roles, and roles per user, especially for reporting.
- Check logging options for batch processes and the scope of auditing.
- Monitor as part of your standard operating procedures.