From a8b5daf1cad42d5847521a1c53231071fcd79091 Mon Sep 17 00:00:00 2001 From: Reese Levine Date: Tue, 3 Dec 2024 11:33:58 -0800 Subject: [PATCH] Update 2024fa.md --- content/lsd-seminar/2024fa.md | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/content/lsd-seminar/2024fa.md b/content/lsd-seminar/2024fa.md index 1c71215..d09f6c4 100644 --- a/content/lsd-seminar/2024fa.md +++ b/content/lsd-seminar/2024fa.md @@ -27,7 +27,7 @@ Talks will be advertised on the [ucsc-lsd-seminar-announce](https://groups.googl | [Nov 8](#nov-8) | Bhakti Shah | Proof Visualization for Graphical Structures | | [Nov 15](#nov-15) | Federico Mora | Automated Reasoning About Distributed Systems | | [Nov 22](#nov-22) | Haining Tong | Towards Unified Analysis of GPU Consistency | -| [Dec 6](#dec-6) | Yanan Guo | _TBD_ | +| [Dec 6](#dec-6) | Yanan Guo | Behind the Pixels: Unveiling GPU Hardware and Software Security Flaws | --- @@ -131,11 +131,13 @@ In this talk, I will present three automated reasoning techniques that can help **Speaker:** Yanan Guo -**Title:** _TBD_ +**Title:** Behind the Pixels: Unveiling GPU Hardware and Software Security Flaws -**Abstract:** _TBD_ +**Abstract:** Today's computing systems face significant security challenges. While security researchers have extensively studied CPU vulnerabilities, GPUs–an increasingly essential component of modern computing systems–have received much less attention. +In this talk, I will present two of my recent studies that help fill in this blank. In the first part, I will demonstrate how hardware designs in modern GPUs can enable practical and robust cache side channels. In the second part, I will discuss our findings on GPU memory management systems and their implications for GPU memory safety. Finally, I will briefly introduce countermeasures to these vulnerabilities and outline directions for future research in GPU security. -**Bio:** _TBD_ + +**Bio:** Yanan Guo joins us as an Assistant Professor. Her research interests lie in computer architecture and cybersecurity, with a goal of building secure, high-performance computing systems. Specifically, she studies microarchitectural side channels, memory exploitation, machine learning security, as well as GPU systems and architectures. She has multiple CPU/GPU side-channel attacks. Her study on prefetch-based side channels was nominated for the Top Picks in Hardware and Embedded Security 2023. ---