Related papers: FlexLink: Boosting your NVLink Bandwidth by 27% wi…

High-Performance Computing (HPC) and Artificial Intelligence (AI) workloads typically demand substantial memory bandwidth and, to a degree, memory capacity. CXL memory expansion modules, also known as CXL "type-3" devices, enable…

Large Language Models (LLMs) face challenges for on-device inference due to high memory demands. Traditional methods to reduce memory usage often compromise performance and lack adaptability. We propose FlexInfer, an optimized offloading…

We propose an architecture, called NVQLink, for connecting high-performance computing (HPC) resources to the control system of a quantum processing unit (QPU) to accelerate workloads necessary to the operation of the QPU. We aim to support…

Multi-GPU systems are becoming increasingly important in highperformance computing (HPC) and cloud infrastructure, providing acceleration for data-intensive applications, including machine learning workloads. These systems consist of…

Supporting ultra-reliable and low-latency communications (URLLC) is one of the major goals for the fifth-generation cellular networks. Since spectrum usage efficiency is always a concern, and large bandwidth is required for ensuring…

In the last decade, specific-purpose computing and storage devices, such as GPUs, TPUs, or high-speed storage, have been incorporated into server nodes of Supercomputers and Data centers. The development of high-bandwidth memory (HBM)…

This paper proposes ScalePool, a novel cluster architecture designed to interconnect numerous accelerators using unified hardware interconnects rather than traditional long-distance networking. ScalePool integrates Accelerator-Centric Links…

Massive interconnection has sparked people's envisioning for next-generation ultra-reliable and low-latency communications (xURLLC), prompting the design of customized next-generation advanced transceivers (NGAT). Rate-splitting multiple…

NCCL is the de facto standard for collective GPU communication in large-scale distributed training, relying heavily on plugins to customize runtime behavior. However, these plugins execute as unverified native code within NCCL's address…

The advent of massive ultra-reliable and low-latency communications (mURLLC) has introduced a critical class of time- and reliability-sensitive services within next-generation wireless networks. This shift has attracted significant research…

Large Language Models (LLMs) have become the new foundation for many applications, reshaping human society like a storm. Disaggregated inference, which separates prefill and decode stages, is a promising approach to improving hardware…

Large Language Models (LLMs) have achieved remarkable success in various fields, but their training and finetuning require massive computation and memory, necessitating parallelism which introduces heavy communication overheads. Driven by…

The rapid advancements in artificial intelligence (AI), particularly the Large Language Models (LLMs), have profoundly affected our daily work and communication forms. However, it is still a challenge to deploy LLMs on resource-constrained…

Large language models represent a groundbreaking shift in generative AI. Yet, these advances come with a significant challenge: the high cost of model serving. To mitigate these costs, consumer-grade GPUs emerge as a more affordable…

Machine learning models are increasingly being trained across multiple GPUs and servers. In this setting, data is transferred between GPUs using communication collectives such as AlltoAll and AllReduce, which can become a significant…

The trend toward specialized processing devices such as TPUs, DPUs, GPUs, and FPGAs has exposed the weaknesses of PCIe in interconnecting these devices and their hosts. Several attempts have been proposed to improve, augment, or downright…

The ever-increasing sizes of large language models necessitate distributed solutions for fast inference that exploit multi-dimensional parallelism, where computational loads are split across various accelerators such as GPU clusters.…

The recent huge advance of Large Language Models (LLMs) is mainly driven by the increase in the number of parameters. This has led to substantial memory capacity requirements, necessitating the use of dozens of GPUs just to meet the…

The rapid advancement of large language models (LLMs) has exacerbated the memory bottleneck due to the widening gap between model parameter scaling and hardware capabilities. While post-training quantization techniques effectively reduce…

Recent efforts to improve the performance of neural network (NN) accelerators that meet today's application requirements have given rise to a new trend of logic-based NN inference relying on fixed-function combinational logic (FFCL). This…