In brief

  • Memory Band-Aid: A Principled Rowhammer Defense-in-Depth introduces per-thread and per-bank rate limits for DRAM accesses.
  • MEVisor: High-Throughput MEV Discovery in DEXs with GPU Parallelism compiles an MEV bot into a GPU application, searching for profit in parallel.
  • MIRAGE: Private, Mobility-based Routing for Censorship Evasion uses a district-based routing scheme based on high-level mobility patterns.

Memory Band-Aid: A Principled Rowhammer Defense-in-Depth

  • Memory Band-Aid introduces per-thread and per-bank rate limits for DRAM accesses in the memory controller, ensuring that the minimum number of row activations for Rowhammer bit flips cannot be reached.
  • Tested on 2 Intel and 2 AMD systems, a full implementation would have a low overhead of 0% to 9.4% on Phoronix macro-benchmarks.
  • In a micro-benchmark to cause DRAM pressure, it causes a slowdown by a factor of 1 to 5.1.
  • Carina Fiedler, Jonas Juffinger, Sudheendra Raghav Neela, Martin Heckel, Martin Schwarzl, Daniel Gruss. “Memory Band-Aid: A Principled Rowhammer Defense-in-Depth.” NDSS — https://www.ndss-symposium.org/ndss-paper/memory-band-aid-a-principled-rowhammer-defense-in-depth/

MEVisor: High-Throughput MEV Discovery in DEXs with GPU Parallelism

  • MeVisor compiles an MEV bot into a GPU application and launches thousands of GPU threads to search for Maximal Extractable Value (MEV) profit in parallel.
  • Evaluated on 3,941 real MEV cases from Ethereum, MeVisor achieves 3.3M-5.1M transactions per second, outperforming the CPU baseline by 100,000x.
  • In a large-scale study of Q1 2025 data, MeVisor estimates MEV opportunities yielding at most $1.1 million in profit.
  • Weimin Chen, Xiapu Luo. “MEVisor: High-Throughput MEV Discovery in DEXs with GPU Parallelism.” NDSS — https://www.ndss-symposium.org/ndss-paper/mevisor-high-throughput-mev-discovery-in-dexs-with-gpu-parallelism/

MIRAGE: Private, Mobility-based Routing for Censorship Evasion

  • MIRAGE is a privacy-preserving mobility-based messaging system designed for censorship-resistant communication, using a district-based routing scheme that probabilistically forwards messages based on high-level mobility patterns.
  • MIRAGE protects users’ mobility patterns with local differential privacy, ensuring participation does not reveal an individual’s location history.
  • Evaluated using real-world pedestrian and taxi trajectories, MIRAGE reduces message overhead compared to epidemic routing and outperforms probabilistic flooding in delivery rate.
  • Zachary Ratliff, Ruoxing (David) Yang, Avery Bai, Harel Berger, Micah Sherr, James Mickens. “MIRAGE: Private, Mobility-based Routing for Censorship Evasion.” NDSS — https://www.ndss-symposium.org/ndss-paper/mirage-private-mobility-based-routing-for-censorship-evasion/

Also published