Related papers: Spy in the GPU-box: Covert and Side Channel Attack…

The last decade has seen a shift in the computer systems industry where heterogeneous computing has become prevalent. Graphics Processing Units (GPUs) are now present in supercomputers to mobile phones and tablets. GPUs are used for…

Covert channels can be utilized to secretly deliver information from high privileged processes to low privileged processes in the context of a high-assurance computing system. In this case study, we investigate the possibility of covert…

GPUs in High-Performance Computing systems remain under-utilised due to the unavailability of schedulers that can safely schedule multiple applications to share the same GPU. The research reported in this paper is motivated to improve the…

Despite the remarkable progress of diffusion models in image generation, recent studies reveal their vulnerability to backdoor attacks via covert visual or textual triggers. Although evolving defense mechanisms can detect most existing…

Backdoor attack is a powerful attack algorithm to deep learning model. Recently, GNN's vulnerability to backdoor attack has been proved especially on graph classification task. In this paper, we propose the first backdoor detection and…

Using GPUs as general-purpose processors has revolutionized parallel computing by offering, for a large and growing set of algorithms, massive data-parallelization on desktop machines. An obstacle to widespread adoption, however, is the…

Timing-based side and covert channels in processor caches continue to be a threat to modern computers. This work shows for the first time a systematic, large-scale analysis of Arm devices and the detailed results of attacks the processors…

GPUs exploit a high degree of thread-level parallelism to hide long-latency stalls. Due to the heterogeneous compute requirements of different applications, there is a growing need to share the GPU across multiple applications in…

This work introduces a GPU storage expansion solution utilizing CXL, featuring a novel GPU system design with multiple CXL root ports for integrating diverse storage media (DRAMs and/or SSDs). We developed and siliconized a custom CXL…

Deep Neural Networks (DNNs) have become ubiquitous due to their performance on prediction and classification problems. However, they face a variety of threats as their usage spreads. Model extraction attacks, which steal DNNs, endanger…

Among the many possible approaches for the parallelization of self-organizing networks, and in particular of growing self-organizing networks, perhaps the most common one is producing an optimized, parallel implementation of the standard…

In this paper, we present that security threats coming with existing GPU memory management strategy are overlooked, which opens a back door for adversaries to freely break the memory isolation: they enable adversaries without any privilege…

Full batch training of Graph Convolutional Network (GCN) models is not feasible on a single GPU for large graphs containing tens of millions of vertices or more. Recent work has shown that, for the graphs used in the machine learning…

Graph Neural Networks (GNNs) have boosted the performance for many graph-related tasks. Despite the great success, recent studies have shown that GNNs are highly vulnerable to adversarial attacks, where adversaries can mislead the GNNs'…

Video and image streaming on edge devices requires low latency. To address this, Neural Networks (NNs) are widely used, and prior work mainly focuses on accelerating them with single hardware units such as Graphics Processing Units (GPUs),…

Protected-module architectures (PMAs) have been proposed to provide strong isolation guarantees, even on top of a compromised system. Unfortunately, Intel SGX -- the only publicly available high-end PMA -- has been shown to only provide…

A Multigrid Full Approximation Storage algorithm for solving Deep Residual Networks is developed to enable neural network parallelized layer-wise training and concurrent computational kernel execution on GPUs. This work demonstrates a 10.2x…

Deep learning training at scale is resource-intensive and time-consuming, often running across hundreds or thousands of GPUs for weeks or months. Efficient checkpointing is crucial for running these workloads, especially in multi-tenant…

NVIDIA Multi-Process Service (MPS) enables fine-grained GPU sharing by allowing multiple processes to execute concurrently on the same GPU, making it an important mechanism for improving GPU utilization. However, MPS has weak fault…

Power side-channel attacks are a very effective cryptanalysis technique that can infer secret keys of security ICs by monitoring the power consumption. Since the emergence of practical attacks in the late 90s, they have been a major threat…