Related papers: OnePiece: A Large-Scale Distributed Inference Syst…

Deep learning models can take weeks to train on a single GPU-equipped machine, necessitating scaling out DL training to a GPU-cluster. However, current distributed DL implementations can scale poorly due to substantial parameter…

The continued growth of the computational capability of throughput processors has made throughput processors the platform of choice for a wide variety of high performance computing applications. Graphics Processing Units (GPUs) are a prime…

As AI workloads drive increasing memory requirements, domain-specific accelerators need higher-density on-chip memory beyond what current SRAM scaling trends can provide. Simultaneously, the vast amounts of short-lived data in these…

Advances in GPU compute throughput and memory capacity brings significant opportunities to a wide range of workloads. However, efficiently utilizing these resources remains challenging, particularly because diverse application…

We present the design and implementation of a RAG-based AI system benchmarking (RAGPerf) framework for characterizing the system behaviors of RAG pipelines. To facilitate detailed profiling and fine-grained performance analysis, RAGPerf…

We introduce relative and absolute position matching (RAPM), a diffusion distillation method resulting in high quality generation that can be trained efficiently on a single GPU. Recent diffusion distillation research has achieved excellent…

GPUs are critical for compute-intensive applications, yet emerging workloads such as recommender systems, graph analytics, and data analytics often exceed GPU memory capacity. Existing solutions allow GPUs to use CPU DRAM or SSDs as…

In past years, the OpenAI's Scaling-Laws shows the amazing intelligence with the next-token prediction paradigm in neural language modeling, which pointing out a free-lunch way to enhance the model performance by scaling the model…

Large Language Models (LLMs) with the Mixture-of-Experts (MoE) architectures have shown promising performance on various tasks. However, due to the huge model sizes, running them in resource-constrained environments where the GPU memory is…

AI-Generated Content (AIGC), as a novel manner of providing Metaverse services in the forthcoming Internet paradigm, can resolve the obstacles of immersion requirements. Concurrently, edge computing, as an evolutionary paradigm of computing…

Due to massive computational demands of large generative models, AI-Generated Content (AIGC) can organize collaborative Mobile AIGC Service Providers (MASPs) at network edges to provide ubiquitous and customized content generation for…

The paper provides a unified co-design of 1) a programming and execution model that allows spawning tasks from within the vertex data at runtime, 2) language constructs for \textit{actions} that send work to where the data resides,…

With the rapid advancement of artificial intelligence (AI), generative AI (GenAI) has emerged as a transformative tool, enabling customized and personalized AI-generated content (AIGC) services. However, GenAI models with billions of…

During the past few years, interest in convolutional neural networks (CNNs) has risen constantly, thanks to their excellent performance on a wide range of recognition and classification tasks. However, they suffer from the high level of…

There is a growing need for distributed graph processing systems that are capable of gracefully scaling to very large graph datasets. Unfortunately, this challenge has not been easily met due to the intense memory pressure imposed by…

Leveraging one-sided RDMA for applications that replicate small data objects can be surprisingly difficult: such uses amplify any protocol overheads. Spindle is a set of optimization techniques for systematically tackling this class of…

Large language models (LLMs) are widely used but expensive to run, especially as inference workloads grow. To lower costs, maximizing the request batch size by managing GPU memory efficiently is crucial. While PagedAttention has recently…

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…

In this paper, we propose LoopLynx, a scalable dataflow architecture for efficient LLM inference that optimizes FPGA usage through a hybrid spatial-temporal design. The design of LoopLynx incorporates a hybrid temporal-spatial architecture,…

The rise of Large Language Models (LLM) has increased the need for scalable, high-performance inference systems, yet most existing frameworks assume homogeneous, resource-rich hardware, often unrealistic in academic, or resource-constrained…