DeSFAM: An Adaptive eBPF and AI-Driven Framework for Securing Cloud Containers in Real Time

Containerized applications offer lightweight and scalable deployment but remain exposed to security risks due to a shared kernel. We present DeSFAM (Dynamic eBPF-driven Syscall Filtering and Anomaly Mitigation), a real-time security framework that enforces least-privilege syscall usage and detects b...

Full description

Saved in:
Bibliographic Details
Main Authors: Sehar Zehra, Hassan Jamil Syed, Fahad Samad, Ummay Faseeha
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11095719/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849237958873317376
author Sehar Zehra
Hassan Jamil Syed
Fahad Samad
Ummay Faseeha
author_facet Sehar Zehra
Hassan Jamil Syed
Fahad Samad
Ummay Faseeha
author_sort Sehar Zehra
collection DOAJ
description Containerized applications offer lightweight and scalable deployment but remain exposed to security risks due to a shared kernel. We present DeSFAM (Dynamic eBPF-driven Syscall Filtering and Anomaly Mitigation), a real-time security framework that enforces least-privilege syscall usage and detects behavioral anomalies. DeSFAM integrates: 1) hybrid syscall profiling through static analysis and dynamic eBPF tracing; 2) SyscallAD (System call Anomaly Detection), a low-latency anomaly detector combining Variational Autoencoder (VAE) and Isolation Forest (iForest); 3) contextual risk scoring based on MITRE ATT&CK mappings and CVE correlations; and 4) adaptive syscall enforcement using eBPF maps and LSM hooks. Evaluations using the DongTing dataset and real-world CVE attack scenarios show DeSFAM achieves 94% precision, 90% recall, sub-millisecond enforcement latency, and less than 1% performance overhead. DeSFAM effectively blocks privilege escalation, container escape attempts, and syscall injection attacks in modern container environments.
format Article
id doaj-art-b6e7083533a0467bbb90cbe6e69c62e9
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-b6e7083533a0467bbb90cbe6e69c62e92025-08-20T04:01:48ZengIEEEIEEE Access2169-35362025-01-011313920313922410.1109/ACCESS.2025.359219211095719DeSFAM: An Adaptive eBPF and AI-Driven Framework for Securing Cloud Containers in Real TimeSehar Zehra0https://orcid.org/0009-0007-7595-1221Hassan Jamil Syed1https://orcid.org/0000-0002-1834-1810Fahad Samad2https://orcid.org/0000-0003-3833-2644Ummay Faseeha3https://orcid.org/0009-0000-5276-1504Department of Computer Science, FAST National University of Computer and Emerging Sciences, Karachi, PakistanAsia Pacific University of Technology and Innovation (APU), Kuala Lumpur, MalaysiaDepartment of Cyber Security, FAST National University of Computer and Emerging Sciences, Karachi, PakistanDepartment of Computer Science, FAST National University of Computer and Emerging Sciences, Karachi, PakistanContainerized applications offer lightweight and scalable deployment but remain exposed to security risks due to a shared kernel. We present DeSFAM (Dynamic eBPF-driven Syscall Filtering and Anomaly Mitigation), a real-time security framework that enforces least-privilege syscall usage and detects behavioral anomalies. DeSFAM integrates: 1) hybrid syscall profiling through static analysis and dynamic eBPF tracing; 2) SyscallAD (System call Anomaly Detection), a low-latency anomaly detector combining Variational Autoencoder (VAE) and Isolation Forest (iForest); 3) contextual risk scoring based on MITRE ATT&CK mappings and CVE correlations; and 4) adaptive syscall enforcement using eBPF maps and LSM hooks. Evaluations using the DongTing dataset and real-world CVE attack scenarios show DeSFAM achieves 94% precision, 90% recall, sub-millisecond enforcement latency, and less than 1% performance overhead. DeSFAM effectively blocks privilege escalation, container escape attempts, and syscall injection attacks in modern container environments.https://ieeexplore.ieee.org/document/11095719/Anomaly detectioncloud containerscontainer securityeBPF monitoringreal-time threat detectionsyscall filtering
spellingShingle Sehar Zehra
Hassan Jamil Syed
Fahad Samad
Ummay Faseeha
DeSFAM: An Adaptive eBPF and AI-Driven Framework for Securing Cloud Containers in Real Time
IEEE Access
Anomaly detection
cloud containers
container security
eBPF monitoring
real-time threat detection
syscall filtering
title DeSFAM: An Adaptive eBPF and AI-Driven Framework for Securing Cloud Containers in Real Time
title_full DeSFAM: An Adaptive eBPF and AI-Driven Framework for Securing Cloud Containers in Real Time
title_fullStr DeSFAM: An Adaptive eBPF and AI-Driven Framework for Securing Cloud Containers in Real Time
title_full_unstemmed DeSFAM: An Adaptive eBPF and AI-Driven Framework for Securing Cloud Containers in Real Time
title_short DeSFAM: An Adaptive eBPF and AI-Driven Framework for Securing Cloud Containers in Real Time
title_sort desfam an adaptive ebpf and ai driven framework for securing cloud containers in real time
topic Anomaly detection
cloud containers
container security
eBPF monitoring
real-time threat detection
syscall filtering
url https://ieeexplore.ieee.org/document/11095719/
work_keys_str_mv AT seharzehra desfamanadaptiveebpfandaidrivenframeworkforsecuringcloudcontainersinrealtime
AT hassanjamilsyed desfamanadaptiveebpfandaidrivenframeworkforsecuringcloudcontainersinrealtime
AT fahadsamad desfamanadaptiveebpfandaidrivenframeworkforsecuringcloudcontainersinrealtime
AT ummayfaseeha desfamanadaptiveebpfandaidrivenframeworkforsecuringcloudcontainersinrealtime