Related papers: Glider: A GPU Library Driver for Improved System S…

Full-graph training of graph neural networks (GNNs) is widely used as it enables direct validation of algorithmic improvements by preserving complete neighborhood information. However, it typically requires multiple GPUs or servers,…

While frontier large language models (LLMs) continue to push capability boundaries, their deployment remains confined to GPU-powered cloud infrastructure. We challenge this paradigm with SmallThinker, a family of LLMs natively designed -…

We introduce GRiD: a GPU-accelerated library for computing rigid body dynamics with analytical gradients. GRiD was designed to accelerate the nonlinear trajectory optimization subproblem used in state-of-the-art robotic planning, control,…

Adaptive finite elements combined with geometric multigrid solvers are one of the most efficient numerical methods for problems such as the instationary Navier-Stokes equations. Yet despite their efficiency, computations remain expensive…

This paper introduces a distributed, GPU-centric experience replay system, GEAR, designed to perform scalable reinforcement learning (RL) with large sequence models (such as transformers). With such models, existing systems such as Reverb…

Modern software systems heavily use C/C++ based libraries. Because of the weak memory model of C/C++, libraries may suffer from vulnerabilities which can expose the applications to potential attacks. For example, a very large number of…

Drivers have a special status among the developer community that sees them as mysterious and inaccessible. We think their extensive communication with the hardware and their need of high performance are the cause of this bad reputation.…

This paper introduces a new C++/CUDA library for GPU-accelerated stochastic optimization called MPPI-Generic. It provides implementations of Model Predictive Path Integral control, Tube-Model Predictive Path Integral Control, and Robust…

Tracers provide users with useful information about program executions. In this article, we propose a ``tracer driver''. From a single tracer, it provides a powerful front-end enabling multiple dynamic analysis tools to be easily…

Fine-tuning Large Language Models (LLMs) has become essential for domain adaptation, but its memory-intensive property exceeds the capabilities of most GPUs. To address this challenge and democratize LLM fine-tuning, we present SlideFormer,…

Continuous learning (CL) has emerged as one of the most popular deep learning paradigms deployed in modern cloud GPUs. Specifically, CL has the capability to continuously update the model parameters (through model retraining) and use the…

Graphics Processing Units (GPUs) are deployed on most present server, desktop, and even mobile platforms. Nowadays, a growing number of applications leverage the high parallelism offered by this architecture to speed-up general purpose…

Graphics Processing Units (GPUs) maintain a large register file to increase the thread level parallelism (TLP). To increase the TLP further, recent GPUs have increased the number of on-chip registers in every generation. However, with the…

The high GPU demand of ML training makes it hard to allocate large homogeneous clusters of high-end GPUs in a single availability zone. Leveraging heterogeneous GPUs available within and across zones can improve throughput at a reasonable…

With the increasing deployment of Large Language Models (LLMs) on mobile and edge platforms, securing them against model extraction attacks has become a pressing concern. However, protecting model privacy without sacrificing the performance…

GPUs are now used for a wide range of problems within HPC. However, making efficient use of the computational power available with multiple GPUs is challenging. The main challenges in achieving good performance are memory layout, affecting…

The number and variety of Internet-connected devices have grown enormously in the past few years, presenting new challenges to security and privacy. Research has shown that network adversaries can use traffic rate metadata from consumer IoT…

Even with generational improvements in DRAM technology, memory access latency still remains the major bottleneck for application accelerators, primarily due to limitations in memory interface IPs which cannot fully account for variations in…

On-device machine learning (ML) inference can enable the use of private user data on user devices without revealing them to remote servers. However, a pure on-device solution to private ML inference is impractical for many applications that…

The rapid advancements in autonomous driving have introduced increasingly complex, real-time GPU-bound tasks critical for reliable vehicle operation. However, the proprietary nature of these autonomous systems and closed-source GPU drivers…