Programmed I/O vs interrupt-driven I/O in Operating Systems - Performance Comparison

Time Complexity: Programmed I/O vs interrupt-driven I/O

O(n)

Understanding Time Complexity

We want to understand how the time taken by Programmed I/O and interrupt-driven I/O changes as the amount of data or number of I/O operations increases.

How does the way the CPU handles input/output affect the overall execution time?

Scenario Under Consideration

Analyze the time complexity of these two I/O methods.

// Programmed I/O example
while (data_not_ready()) {
  check_device_status();
}
read_data();

// Interrupt-driven I/O example
// CPU does other work
// When device ready, interrupt triggers
void interrupt_handler() {
  read_data();
}
    

This code shows how Programmed I/O waits actively for data, while interrupt-driven I/O lets the CPU do other work until data is ready.

Identify Repeating Operations

Look at what repeats in each method.

• Programmed I/O primary operation: Continuously checking device status in a loop.
• Programmed I/O how many times: Repeats until data is ready, which depends on device speed.
• Interrupt-driven I/O primary operation: CPU runs other tasks; device signals when ready via interrupt.
• Interrupt-driven I/O how many times: Interrupt handler runs only when data is ready, not repeatedly checking.

How Execution Grows With Input

Consider how time spent waiting or handling I/O grows as more data arrives.

Input Size (n)	Programmed I/O Approx. Operations	Interrupt-driven I/O Approx. Operations
10	Many repeated checks per data item, total high	10 interrupt calls, CPU free otherwise
100	Much more repeated checking, grows quickly	100 interrupt calls, CPU mostly free
1000	Very large number of checks, CPU mostly busy waiting	1000 interrupt calls, CPU efficiently used

Pattern observation: Programmed I/O time grows quickly because CPU waits actively; interrupt-driven I/O scales better as CPU does other work between interrupts.

Final Time Complexity

Time Complexity: O(n)

This means the time spent handling I/O grows linearly with the number of data items, but programmed I/O wastes CPU time checking repeatedly, while interrupt-driven I/O uses CPU time more efficiently.

Common Mistake

[X] Wrong: "Programmed I/O is always faster because it checks immediately for data."

[OK] Correct: Constant checking wastes CPU time and slows down other tasks, making overall performance worse as data grows.

Interview Connect

Understanding how different I/O methods affect CPU time helps you explain system efficiency clearly and shows you can think about resource use in real systems.

Self-Check

"What if the device sends data in large bursts instead of one item at a time? How would that affect the time complexity of interrupt-driven I/O?"

Practice

1. Which of the following best describes programmed I/O?

easy

A. CPU ignores the device until a scheduled time

B. Device sends a signal to CPU only when it needs attention

C. CPU and device communicate using direct memory access

D. CPU continuously checks the device status until it is ready

Solution

1. Step 1: Understand programmed I/O behavior

   Programmed I/O means the CPU actively checks the device repeatedly to see if it is ready.

2. Step 2: Compare options with definition

   CPU continuously checks the device status until it is ready matches this behavior exactly, while others describe interrupt or other methods.

3. Final Answer:

   CPU continuously checks the device status until it is ready -> Option D

4. Quick Check:

   Programmed I/O = CPU polls device [OK]

Hint: Programmed I/O means CPU keeps checking device status [OK]

Common Mistakes:

• Confusing programmed I/O with interrupt-driven I/O
• Thinking device signals CPU in programmed I/O
• Mixing up direct memory access with programmed I/O

2. Which of the following is the correct statement about interrupt-driven I/O?

easy

A. Device interrupts CPU only when it needs service

B. CPU waits in a loop checking device status

C. CPU ignores device signals completely

D. CPU uses programmed polling to check device

Solution

1. Step 1: Recall interrupt-driven I/O definition

   Interrupt-driven I/O means the device sends a signal (interrupt) to the CPU when it needs attention.

2. Step 2: Match options to definition

   Device interrupts CPU only when it needs service correctly states that the device interrupts the CPU only when service is needed.

3. Final Answer:

   Device interrupts CPU only when it needs service -> Option A

4. Quick Check:

   Interrupt-driven I/O = device signals CPU [OK]

Hint: Interrupt-driven means device signals CPU when ready [OK]

Common Mistakes:

• Choosing options describing CPU polling
• Assuming CPU ignores device signals
• Confusing programmed I/O with interrupt-driven I/O

3. Consider this scenario: A CPU uses programmed I/O to read data from a device. What is the main effect on CPU usage?

medium

A. CPU can perform other tasks while waiting for device

B. CPU is busy checking device status and cannot do other work

C. CPU ignores the device until data is ready

D. CPU receives an interrupt when device is ready

Solution

1. Step 1: Understand CPU behavior in programmed I/O

   In programmed I/O, CPU repeatedly checks device status, which keeps it busy and unable to do other tasks efficiently.

2. Step 2: Analyze options for CPU usage

   CPU is busy checking device status and cannot do other work correctly states CPU is busy polling and cannot perform other work effectively.

3. Final Answer:

   CPU is busy checking device status and cannot do other work -> Option B

4. Quick Check:

   Programmed I/O = CPU busy polling device [OK]

Hint: Programmed I/O keeps CPU busy waiting, no multitasking [OK]

Common Mistakes:

• Thinking CPU can multitask during programmed I/O
• Confusing interrupt-driven I/O effects with programmed I/O
• Assuming CPU ignores device in programmed I/O

4. A system uses interrupt-driven I/O but the CPU never responds to device interrupts. What is the likely problem?

medium

A. Interrupts are disabled or masked in the CPU

B. CPU is using programmed I/O instead

C. Device is not sending any interrupts

D. CPU is stuck in a loop polling the device

Solution

1. Step 1: Identify why CPU ignores interrupts

   If CPU never responds to interrupts, a common cause is that interrupts are disabled or masked, preventing CPU from handling them.

2. Step 2: Evaluate other options

   Interrupts are disabled or masked in the CPU correctly identifies this cause. Other options either contradict the interrupt-driven setup or describe different issues.

3. Final Answer:

   Interrupts are disabled or masked in the CPU -> Option A

4. Quick Check:

   CPU ignoring interrupts = interrupts disabled [OK]

Hint: CPU ignoring interrupts usually means interrupts are disabled [OK]

Common Mistakes:

• Assuming device is not sending interrupts
• Confusing programmed I/O with interrupt-driven I/O
• Thinking CPU polls device in interrupt-driven mode

5. A system designer wants to maximize CPU efficiency while handling multiple I/O devices. Which approach is best and why?

hard

A. Use programmed I/O because CPU controls all device checks

B. Use programmed I/O with longer polling intervals to reduce CPU load

C. Use interrupt-driven I/O so CPU works on other tasks until interrupted

D. Use neither; rely on manual device checks by the user

Solution

1. Step 1: Analyze CPU efficiency needs

   Maximizing CPU efficiency means CPU should not waste time waiting or polling devices unnecessarily.

2. Step 2: Compare I/O methods for efficiency

   Interrupt-driven I/O allows CPU to do other work and only respond when device signals, improving efficiency.

3. Step 3: Evaluate other options

   Programmed I/O wastes CPU cycles; manual checks are impractical. Longer polling reduces load but still wastes CPU time.

4. Final Answer:

   Use interrupt-driven I/O so CPU works on other tasks until interrupted -> Option C

5. Quick Check:

   Interrupt-driven I/O maximizes CPU efficiency [OK]

Hint: Interrupt-driven I/O lets CPU multitask efficiently [OK]

Common Mistakes:

• Choosing programmed I/O for efficiency
• Ignoring CPU multitasking benefits
• Thinking manual checks improve performance