English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

The deep learning revolution has been enabled in large part by GPUs, and morerecently accelerators, which make it possible to carry out computationallydemanding training and inference in acceptable times. As the size of machinelearning networks and workloads continues to increase, multi-GPU machines haveemerged as an important platform offered on High Performance Computing andcloud data centers. As these machines are shared between multiple users, itbecomes increasingly important to protect applications against potentialattacks. In this paper, we explore the vulnerability of Nvidia's DGX multi-GPUmachines to covert and side channel attacks. These machines consist of a numberof discrete GPUs that are interconnected through a combination of custominterconnect (NVLink) and PCIe connections. We reverse engineer the cachehierarchy and show that it is possible for an attacker on one GPU to causecontention on the L2 cache of another GPU. We use this observation to firstdevelop a covert channel attack across two GPUs, achieving the best bandwidthof 3.95 MB/s. We also develop a prime and probe attack on a remote GPU allowingan attacker to recover the cache hit and miss behavior of another workload.This basic capability can be used in any number of side channel attacks: wedemonstrate a proof of concept attack that fingerprints the application runningon the remote GPU, with high accuracy. Our work establishes for the first timethe vulnerability of these machines to microarchitectural attacks, and we hopethat it guides future research to improve their security.

Spy in the GPU-box: Covert and Side Channel Attacks on Multi-GPU Systems