云环境中基于cache负载实时定噪的同驻分析方法

Co-residency Detection Scheme Based on Cache Load and Real Time Noise Ascertainment in Cloud

云计算具有使用便捷、可按需定制服务、优化资源利用等特点,成为提供外包服务的主要计算模式。云环境中的虚拟机侧通道攻击是云计算的主要潜在威胁之一,同驻是云环境中侧通道攻击的前提。针对如何在多租户云环境下进行同驻检测,提出基于链式结构的Prime-Probe测量cache负载方法 MCLPPLS和针对云环境噪声复杂多变问题的实时噪声分析机制RTNAM。结合MCLPPLS与RTNAM提出一种新型的同驻检测分析方法。实验表明,该方法能减少突发噪声对同驻检测的干扰,有较高的同驻检测正确率及较低的同驻检测时耗,表现出良好的性能。

Cloud computing has the advantages of convenient use, designing customized service on need base, optimizing resource utilization etc. It has become the main computing model for outsourcing services. The side channel attack of vir- tual machines in the cloud environment is one of the main potential threats of cloud computing, and the co-residency is the premise of the side channel attack in the cloud environment. In view of how to carry out the co-residency detection in multi tenant cloud environment, this paper presented the measurement of cache load by Prime-Probe with linked struct （MCLPPLS） and real time noise ascertainment mechanism（RTNAM）. Based on MCLPPLS and RTNAM, we proposed a new method for the analysis of the co-residency detection. The experimental results show that the method can reduce the interference of the burst noise to the co-residency detection, and has higher true detection rate and lower detection time,which shows good performance.