IRQ Affinity: How Does Linux Distribute Hardware Interrupts Across CPU Cores to Improve Performance?

Introduction

Every network interface card (NIC), storage device, or hardware component connected to a server relies on Interrupt Requests (IRQs) to notify the CPU whenever an event requires immediate attention.

If all hardware interrupts are handled by a single CPU core, that core can become overloaded while the remaining cores remain underutilized. This imbalance can reduce system performance and increase latency.

To solve this problem, Linux provides a feature called IRQ Affinity, which allows administrators to distribute hardware interrupts across multiple CPU cores for better performance and scalability.

What Is IRQ Affinity?

IRQ Affinity is a Linux feature that determines which CPU core—or group of CPU cores—will handle the interrupts generated by a specific hardware device.

By controlling interrupt distribution, Linux can balance the workload across available processors, improving the performance of high-throughput devices such as network adapters and NVMe storage drives.

How Does IRQ Affinity Work?

The process works as follows:

1. A hardware device generates an Interrupt Request (IRQ).
2. The Linux kernel assigns that interrupt to a CPU core.
3. System administrators can modify the CPU affinity of specific IRQs, binding them to selected cores.
4. Interrupt processing is distributed more evenly, reducing CPU bottlenecks and improving overall performance.

This flexible approach enables better resource utilization on multi-core systems.

Benefits of IRQ Affinity

Improved Network Performance

Distributing network interrupts across multiple CPU cores prevents a single processor from becoming overloaded.

Lower Latency

Applications with strict latency requirements, such as databases and real-time services, benefit from more balanced interrupt handling.

Better CPU Utilization

IRQ Affinity spreads interrupt processing across available cores, improving load balancing and making better use of modern multi-core processors.

Enhanced Storage Performance

Fast storage devices such as NVMe SSDs generate large numbers of interrupts, making IRQ Affinity particularly valuable for storage-intensive workloads.

Where Is IRQ Affinity Commonly Used?

• Web servers
• Database servers
• Kubernetes clusters
• KVM virtualization
• Game servers
• High-performance storage systems

How Can You Monitor IRQ Affinity?

To view the current interrupt distribution, use:

cat /proc/interrupts

To inspect or configure IRQ affinity settings for individual devices, use the files located under:

/proc/irq/

These interfaces allow administrators to monitor and customize how interrupts are assigned to CPU cores.

Best Practices

• Distribute hardware interrupts across multiple CPU cores to avoid CPU bottlenecks.
• Combine IRQ Affinity with Receive Side Scaling (RSS) to maximize network performance.
• Monitor CPU utilization and interrupt distribution after making configuration changes.
• Benchmark workloads before applying IRQ Affinity settings in production.

FAQ

Do All Hardware Devices Benefit from IRQ Affinity?

Not necessarily. The greatest benefits are seen with devices that generate a large number of interrupts, such as high-speed network adapters, NVMe SSDs, and storage controllers.

Can Incorrect IRQ Affinity Settings Reduce Performance?

Yes. Poor interrupt distribution can overload certain CPU cores while leaving others underutilized. For this reason, IRQ Affinity settings should always be tested and monitored before deployment in production.

Conclusion

IRQ Affinity is a powerful Linux feature that distributes hardware interrupt processing across multiple CPU cores, reducing bottlenecks and improving system responsiveness. It is particularly valuable in high-performance environments with intensive network or storage workloads. When combined with technologies such as Receive Side Scaling (RSS) and proper CPU tuning, IRQ Affinity can significantly enhance the scalability and efficiency of modern Linux servers.