27 February 2023 in IEEE Transactions on Computers
Oblivious RAM (ORAM) remains a bittersweet protection of memory access patterns because of its prohibitively high overhead. The root cause is that ORAM hides intended accesses among a sufficiently large number of dummy accesses. Most existing optimizations mitigate memory accesses using architectural enhancements (e.g., cache) yet few of them improve the efficiency of ORAM primitives per se. In this paper, we identify path-grained static scheduling as a fundamental ORAM performance bottleneck. We propose level-grained dynamic scheduling that directly optimizes ORAM primitives to boost efficiency. It enables ORAM to service more than one request per path and write paths batch wise. We can thus boost ORAM efficiency through handling queued requests as soon as possible and remove as many redundant accesses as possible. Since optimized memory accesses still target the same set of paths, dynamic scheduling preserves ORAM security. We implement dynamic scheduling through Hitchhiker ORAM. In comparison with the state-of-the-art primitive-optimized Fork Path ORAM, Hitchhiker ORAM yields 31.5% fewer memory accesses, 60.2% shorter latency, and 40.7% less energy consumption, being 2.5× faster. In comparison with the state-of-the-art architecture-optimized $\rho$ , Hitchhiker ORAM is 1.5× faster and the integrated version— $\rho$ -Hitchhiker ORAM is 2.0× faster.
22 February 2024 in IEEE Transactions on Information Forensics and Security
Conflict-based cache attacks can leak critical information from target programs. Accordingly, randomization-based cache designs have emerged as an efficient and LLC-favorable way to mitigate such attacks. However, later investigations have revealed several problems with these designs. Specifically, we identify limited randomness and coarse-grained protection as key issues of previous designs. To solve these issues, we propose Recast , a secure cache design with address-sensitive secret generation and tweakable index randomization. Our insight is that cache modules at different levels can work collaboratively to enhance their security. Address-sensitive secret generation in private caches generates a secret value for each address upon cache misses. The shared cache in Recast uses tweakable index randomization, where the cryptographic function uses the secret value from private caches as the input to calculate the cache set index. Therefore, Recast achieves fine-grained dynamic mapping. We implement Recast in the gem5 simulator. We use a micro-benchmark and a benchmark suite to showcase the security of Recast. Our performance evaluations on SPEC 2017 and PARSEC benchmarks show that Recast incurs 2.29% and 2.03% performance overhead. Moreover, Recast with the LRU replacement policy has only 0.51% and 1.04% performance overhead on the two benchmarks. Therefore, Recast provides higher security guarantees with minimal performance overhead.