SeqAss: Using Sequential Associative Caches to Mitigate Conflict-Based Cache Attacks With Reduced Cache Misses and Performance Overhead

SeqAss is a randomized cache structure using sequential associativity. Without relying on techniques intrusive to the traditional cache structure, such as cache skews, overprovided metadata space, and random replacement policy, SeqAss retains the set-associative structure and supports LRU. It achieves a defense as strong as Mirage. To be specific, SeqAss achieves around the same defense strength with Mirage in thwarting both Evict+Time and Prime+Probe attacks. Instead of raising cache miss rate, SeqAss actually reduces it by 11.4%, incurring the lowest cache miss rate in all existing randomized cache structures.