sarimahsan101/hpc-dataset

HPC数据集是一个专门设计的表格数据集，用于检测虚拟化环境中基于缓存的侧信道攻击。它通过高频采样（每200毫秒间隔）捕获低级别的硬件性能计数器（HPC）指标，以区分良性的虚拟机工作负载和试图通过共享缓存基础设施（例如Prime+Probe、Flush+Reload）泄露敏感数据的主动对抗攻击。通过整合不同的背景工作负载（空闲、web服务器、数据库查询）和工程化的时间特征，该数据集为训练和评估针对云和虚拟机管理程序安全的基于机器学习的入侵检测系统（IDS）提供了坚实的基础。The HPC Dataset is a specialized tabular dataset designed for the detection of cache-based side-channel attacks in virtualized environments. It captures low-level Hardware Performance Counter (HPC) metrics sampled at high frequency (200ms intervals) to distinguish between benign VM workloads and active adversarial attacks attempting to leak sensitive data via shared cache infrastructure (e.g., Prime+Probe, Flush+Reload). By incorporating varied background workloads (idle, web_server, db_queries) and engineered temporal features, this dataset provides a robust foundation for training and evaluating machine learning-based intrusion detection systems (IDS) tailored for cloud and hypervisor security.