Cache-based side-channel attacks have achieved more attention along with the development of cloud computing technologies. However, current host-based mitigation methods either provide bad compatibility with current cloud infrastructure, or turn out too application-specific. Besides, they are defending blindly without any knowledge of ongoing attacks. In this work, we present CacheSCDefender, a framework that provides a (Virtual Machine Monitor) VMM-based comprehensive defense framework against all levels of cache attacks. In designing CacheSCDefender, we make three key contributions: (1) an attack-aware framework combining our novel dynamic remapping and traditional cache cleansing, which provides a comprehensive defense against all three cases of cache attacks that we identify in this paper; (2) a new defense method called dynamic remapping which is a developed version of random permutation and is able to deal with two cases of cache attacks; (3) formalization and quantification of security improvement and performance overhead of our defense, which can be applicable to other defense methods. We show that CacheSCDefender is practical for deployment in normal virtualized environment, while providing favorable security guarantee for virtual machines.
In seismic data processing, the formulation of the three-dimensional one-pass 45" implicit finitedifference migration schemes requires the solution of a huge number of very large pentadiagonal complex linear systems. Most of the eflort in using iterative methods to accurately solve the problem was not successful due to the bad condition numbers of the systems. In this paper, we present a parallel algorithm to solve the large sparse systems iteratively, based on the preconditioned conjugate gradient (PCG) method. First, b y reformation of the linear systems and choosing a special preconditioner, the PCG method is effective in solving these linear systems. By using a ununiform distribution of linear systems among all PES, a special global summation procedure and a heterogeneous computation technique, the parallel algorithm is quite eficient. Finally, we present some numerical results on both the CRAY-YMP and CRAY-T3D showing the eficiency and effectiveness of the new algorithm.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.