Related papers: UCCL-EP: Portable Expert-Parallel Communication

Mixture-of-Experts (MoE) has been gaining popularity due to its successful adaptation to large language models (LLMs). In this work, we introduce Privacy-preserving Collaborative Mixture-of-Experts (PC-MoE), which leverages the sparsity of…

The NVIDIA Collective Communication Library (NCCL) is a critical software layer enabling high-performance collectives on large-scale GPU clusters. Despite being open source with a documented API, its internal design remains largely opaque.…

Token-choice Mixture-of-Experts (TC-MoE) routes each token to a fixed number of experts, limiting dynamic computation allocation and requiring auxiliary losses to maintain load balance. We propose Expert Threshold (ET) routing, where each…

Unequal error protection (UEP) coding that enables differentiated reliability levels within a transmitted message is essential for modern communication systems. Autoencoder (AE)-based code designs have shown promise in the context of…

To alleviate hardware scarcity in training large deep neural networks (DNNs), particularly large language models (LLMs), we present FusionLLM, a decentralized training system designed and implemented for training DNNs using geo-distributed…

High parallel framework has been proved to be very suitable for graph processing. There are various work to optimize the implementation in FPGAs, a pipeline parallel device. The key to make use of the parallel performance of FPGAs is to…

General Purpose Graphics Processing Unit (GPGPU) computing plays a transformative role in deep learning and machine learning by leveraging the computational advantages of parallel processing. Through the power of Compute Unified Device…

Efficient large-scale inference of transformer-based large language models (LLMs) remains a fundamental systems challenge, frequently requiring multi-GPU parallelism to meet stringent latency and throughput targets. Conventional tensor…

Mixture-of-experts (MoE) has been extensively employed to scale large language models to trillion-plus parameters while maintaining a fixed computational cost. The development of large MoE models in the distributed scenario encounters the…

This manuscript has been submitted to Transactions on Communications on September 7, 2017; revised on January 10, 2018 and March 27, 2018; and accepted on April 25, 2018 We propose a novel filter-type equalizer to improve the solution of…

It is commonly assumed that the end-to-end networking performance of edge offloading is purely dictated by that of the network connectivity between end devices and edge computing facilities, where ongoing innovation in 5G/6G networking can…

With the increasing adoption of AI applications such as large language models and computer vision AI, the computational demands on AI inference systems are continuously rising, making the enhancement of task processing capacity using…

OpenCL is a standard for parallel programming of heterogeneous systems. The benefits of a common programming standard are clear; multiple vendors can provide support for application descriptions written according to the standard, thus…

We present cudaclaw, a CUDA-based high performance data-parallel framework for the solution of multidimensional hyperbolic partial differential equation (PDE) systems, equations describing wave motion. cudaclaw allows computational…

Emergency Wireless Communication (EWC) networks adopt the User Datagram Protocol (UDP) to transmit scene images in real time for quickly assessing the extent of the damage. However, existing UDP-based EWC exhibits suboptimal performance…

Multi-Domain Continual Learning (MDCL) acquires knowledge from sequential tasks with shifting class sets and distribution. Despite the Parameter-Efficient Fine-Tuning (PEFT) methods can adapt for this dual heterogeneity, they still suffer…

In large language models like the Generative Pre-trained Transformer, the Mixture of Experts paradigm has emerged as a powerful technique for enhancing model expressiveness and accuracy. However, deploying GPT MoE models for parallel…

RDMA has been widely adopted for high-speed datacenter networks. However, native RDMA merely supports one-to-one reliable connection, which mismatches various applications with group communication patterns (e.g., one-to-many). While there…

In recent years, Mixture-of-Experts (MoE) has emerged as an effective approach for enhancing the capacity of deep neural network (DNN) with sub-linear computational costs. However, storing all experts on GPUs incurs significant memory…

The Brain-Computer Interface (BCI) enables direct brain-to-device communication, with the Steady-State Visual Evoked Potential (SSVEP) paradigm favored for its stability and high accuracy across various fields. In SSVEP BCI systems,…