Abstract
The rapid growth of network traffic volume, link speed, and encryption has made real-time intrusion detection and prevention increasingly difficult in modern networks. Intrusion Detection Systems (IDS) and Intrusion Prevention Systems (IPS) are widely deployed to inspect traffic, identify and block malicious activity before reaching protected resources. However, conventional IDS/IPS architectures often rely on general-purpose CPUs and the traditional network stack for packet acquisition, packet inspection, flow tracking, cryptographic processing, and signature matching. As port speeds continue to outpace CPU performance, these systems face packet loss, increased latency, and missed detections, especially for encrypted traffic, stateful inspection, and inline prevention. To address these limitations, researchers and industry have proposed diverse software and hardware acceleration techniques. While prior surveys examine IDS accuracy, machine learning, and packet-processing acceleration in isolation, the literature lacks a view connecting acceleration technologies to the IDS/IPS functions they improve and the deployment trade-offs.This paper surveys acceleration techniques for IDS/IPS systems by presenting the fundamentals of detection methods, comparing different architectures, and discussing performance bottlenecks. It compares related surveys to identify the research gap, then introduces a taxonomy that maps acceleration techniques to core IDS/IPS functions. For each category, representative systems are analyzed in terms of performance, programmability, scalability, detection capabilities, and deployment implications. Finally, the survey discusses open challenges and future research directions to guide the design of scalable, high-performance network defenses.

This publication has 39 references indexed in Scilit: