Dispatching priority in mainframe systems (like IBM Z) is the ranking assigned to a unit of work (task, thread, or address space) that determines when and how often it gets CPU time.

👉 In simple terms:
It decides “who runs next on the CPU—and for how long.”

🔷 1. What “Dispatching” Means

• The OS scheduler selects a task from a queue
• Assigns it to a CPU (dispatches it)
• Runs it for a time slice

👉 Dispatching priority controls the order of selection

🔷 2. Dispatching Priority vs Importance

These are related but different:

• Importance (WLM) → Business priority (e.g., banking > batch)
• Dispatching Priority → Technical execution priority

👉 IBM Workload Manager converts business goals into dispatching priorities.

🔷 3. How It Works in z/OS

Each unit of work gets a dynamic dispatching priority value.

Higher priority:

• Gets CPU sooner
• Runs more frequently

Lower priority:

• Waits longer
• Gets CPU less often

🔷 4. Dynamic Adjustment (Key Feature)

Dispatching priority is not fixed.

WLM continuously adjusts it based on:

• Response time goals
• CPU usage
• Queue delays
• Workload importance

👉 Example:

• If a critical transaction is slow → priority increases
• If a batch job is ahead → priority decreases

🔷 5. Priority Queues

The system maintains multiple queues:

• High-priority queue
• Medium-priority queue
• Low-priority queue

Dispatcher always picks from:
👉 Highest non-empty queue

🔷 6. Time Slicing and Fairness

Even high-priority tasks:

• Run for a limited time slice
• Then yield CPU

👉 Prevents:

• Starvation of lower-priority work
• System imbalance

🔷 7. Integration with CPU Scheduling

Dispatching priority influences:

• Which logical processor runs the task
• How often it gets CPU
• How long it stays runnable

👉 Works with PR/SM and hardware scheduling

🔷 8. Special Cases

🔹 Real-Time / Critical Work

• Very high dispatching priority
• Near-immediate CPU access

🔹 Background Jobs

• Lower priority
• Run when system is less busy

🔷 9. Sysplex-Level Impact

In IBM Parallel Sysplex:

• Dispatching priority works with:
  • Workload routing
  • System balancing

👉 Ensures:

• Best system handles the work

🔷 🔥 Simple Example

Imagine 3 tasks:

CPU will run:
👉 A → A → B → A → C (depending on system state)

🔷 🔥 Analogy

Think of a hospital:

• Emergency patients → immediate treatment (high priority)
• Regular patients → wait their turn
• Routine checkups → lowest priority

👉 Dispatching priority ensures critical cases are handled first

🔷 🚀 Bottom Line

Dispatching priority:

✔ Determines which task gets CPU time first
✔ Is dynamically adjusted by WLM
✔ Aligns system behavior with business goals
✔ Ensures critical workloads meet performance targets
✔ Maintains fairness and system stability