4 papers
Amplifying Latency and Cost of LLM Inference via Rowhammer Attack
Mrityunjay Shukla, Shubham Roy, Biswabandan Panda
Large language models (LLMs) achieve high inference efficiency using optimized GPU kernels generated by just-in-time (JIT) compilation frameworks such as Triton. Triton compiles GP…
The Avatar Cache: Enabling On-Demand Security with Morphable Cache Architecture
Anubhav Bhatla, Navneet Navneet, Moinuddin Qureshi +1
The sharing of the last-level cache (LLC) among multiple cores makes it vulnerable to cross-core conflict- and occupancy-based attacks. Despite extensive prior work, modern process…
Micro BTB: A High Performance and Lightweight Last-Level Branch Target Buffer for Servers
Vishal Gupta, Biswabandan Panda
High-performance branch target buffers (BTBs) and the L1I cache are key to high-performance front-end. Modern branch predictors are highly accurate, but with an increase in code fo…
Introducing Fast and Secure Deterministic Stash Free Write Only Oblivious RAMs for Demand Paging in Keystone
Mriganka Shekhar Chakravarty, Biswabandan Panda
Keystone is a trusted execution environment, based on RISC-V architecture. It divides the memory into a secure Keystone private memory and an unsecure non-Keystone memory, and allo…