Related papers: UB-Mesh: a Hierarchically Localized nD-FullMesh Da…

The rapid evolution of large language models (LLMs), driven by growing parameter scales, adoption of mixture-of-experts (MoE) architectures, and expanding context lengths, imposes unprecedented demands on AI infrastructure. Traditional AI…

The recent progress made in large language models (LLMs) has brought tremendous application prospects to the world. The growing model size demands LLM training on multiple GPUs, while data parallelism is the most popular distributed…

AI deployment increasingly resembles a pipeline of data transformation, fine-tuning, and agent interactions rather than a monolithic LLM job; recent examples include RLHF/RLAIF training and agentic workflows. To cope with this shift, we…

We study the problem of routing queries to large language models (LLMs) under cost, GPU resources, and concurrency constraints. Prior per-query routing methods often fail to control batch-level cost, especially under non-uniform or…

A key challenge in on-chip interconnect design is to scale up bandwidth while maintaining low latency and high area efficiency. 2D-meshes scale with low wiring area and congestion overhead; however, their end-to-end latency increases with…

In the evolving landscape of transportation systems, integrating Large Language Models (LLMs) offers a promising frontier for advancing intelligent decision-making across various applications. This paper introduces a novel 3-dimensional…

Non-Uniform Memory Access (NUMA) architecture imposes numerous performance challenges to today's cloud workloads. Due to the complexity and the massive scale of modern warehouse-scale computers (WSCs), a lot of efforts need to be done to…

Recent advancements in Multimodal Large Language Models (MLLMs) underscore the significance of scalable models and data to boost performance, yet this often incurs substantial computational costs. Although the Mixture of Experts (MoE)…

In this paper, we propose the MultiLevel Variational MultiScale (ML-VMS) method, a novel approach that seamlessly integrates a multilevel mesh strategy into the Variational Multiscale (VMS) framework. A key feature of the ML-VMS method is…

Network topology is critical for efficient parameter synchronization in distributed learning over networks. However, most existing studies do not account for bandwidth limitations in network topology design. In this paper, we propose a…

Achieving an end-to-end low-latency for computations offloading, in Mobile Edge Computing (MEC) systems, is still a critical design problem. This is because the offloading of computational tasks via the MEC servers entails the use of uplink…

Large Language Models (LLMs) have demonstrated remarkable language understanding and generation capabilities. However, training, deploying, and accessing these models pose notable challenges, including resource-intensive demands, extended…

Innovation in the physical layer of communication systems has traditionally been achieved by breaking down the transceivers into sets of processing blocks, each optimized independently based on mathematical models. Conversely, deep learning…

Deep neural networks (DNNs) have shown superior performances on various multimodal learning problems. However, it often requires huge efforts to adapt DNNs to individual multimodal tasks by manually engineering unimodal features and…

This paper proposes an efficient neural network (NN) architecture design methodology called Chameleon that honors given resource constraints. Instead of developing new building blocks or using computationally-intensive reinforcement…

Unmanned aerial vehicle (UAV)-assisted multi-access edge computing (MEC) has become one promising solution for energy-constrained devices to meet the computation demand and the stringent delay requirement. In this work, we investigate a…

Data analytics systems commonly utilize in-memory query processing techniques to achieve better throughput and lower latency. Modern computers increasingly rely on Non-Uniform Memory Access (NUMA) architectures in order to achieve…

Large language model (LLM) decoding is a major inference bottleneck because its low arithmetic intensity makes performance highly sensitive to memory bandwidth. 3D-stacked near-memory processing (NMP) provides substantially higher local…

Nowadays, the explosion of unstructured data presents immense analytical value. Leveraging the remarkable capability of large language models (LLMs) in extracting attributes of structured tables from unstructured data, researchers are…

Modern Deep Learning (DL) models have grown to sizes requiring massive clusters of specialized, high-end nodes to train. Designing such clusters to maximize both performance and utilization--to amortize their steep cost--is a challenging…