AI, Data Science & Cloud Certificates from Google, IBM & Meta — 40% Off
One plan covers every Professional Certificate on Coursera. 40% off Coursera Plus Annual.
Unlock All Certificates
Explore a comprehensive systematization of knowledge (SoK) conference presentation that analyzes the security landscape of randomized cache designs against side-channel attacks. Examine how researchers from the Indian Institute of Technology Bombay investigate five critical microarchitectural modifications—skewing, extra invalid tags, high associativity, replacement policy, and remapping—used in modern secure cache designs to defend against conflict-based attacks on Last-Level Caches (LLC). Learn about the systematic evaluation of these "security knobs" and discover why no single modification provides complete protection, while only specific combinations offer effective security guarantees. Understand the evolution of cache security over the past decade, including defenses against flush-based, conflict-based, and occupancy-based attacks, and gain insights into the previously unexplored design space of randomized LLC architectures. Discover the research findings that reveal which combinations of security modifications actually provide meaningful protection and which offer minimal security benefits, presented by the Distinguished Artifact Award winners at USENIX Security '25.
