Oracle Apps DBA: Keeping E-Business Suite Alive · Module 11 · Performance: finding the actual problem
Statistics: the commonest real cause
Lesson 119 of 184 · 3 min
What they are, why EBS has its own way of gathering them, and THE PLAN THAT CHANGED OVERNIGHT BECAUSE NOBODY HAD. Statistics describe the data — how many rows, how values are distributed — and the optimiser chooses a plan from them. So the plan is only as good as the description, and a description that no longer matches the data produces a confidently wrong plan. WHICH IS WHY THIS IS THE COMMONEST REAL CAUSE OF A PERFORMANCE COMPLAINT IN EBS. A table that had four hundred rows when statistics were last gathered now has two million. The optimiser still believes four hundred, chooses a plan that would be excellent for four hundred, and applies it to two million — which is Lesson 7's estimate mismatch, seen from its cause. THE OVERNIGHT CHANGE IS THE SIGNATURE, AND IT IS WHAT MAKES THIS DIAGNOSABLE FROM THE COMPLAINT ALONE. A report that…
The full lesson is part of the course
The video, the complete written lesson and the module quiz are included in Oracle Apps DBA: Keeping E-Business Suite Alive, with a certificate on completion and a fourteen-day refund window.
In this module: Module 11 · Performance: finding the actual problem
- 1Measure before you tuneFree preview
- 2"The system is slow": turning that into a question
- 3Where time actually goes
- 4AWR and ASH, used properly
- 5A baseline is what makes a report meaningful
- 6Finding the expensive SQL
- 7Execution plans, enough to act
- 8Statistics: the commonest real cause
- 9Locking and blocking
- 10When it is the application, not the database
- 11What breaks: four performance investigations that went wrong
- 12Lab briefing · Diagnose three slow systems