English
Related papers

Related papers: Revisiting Parameter Server in LLM Post-Training

200 papers

Memory-to-memory data streaming is essential for modern scientific workflows that require near real-time data analysis, experimental steering, and informed decision-making during experiment execution. It eliminates the latency bottlenecks…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-09-10 Anjus George , Michael J. Brim , Christopher Zimmer , Tyler J. Skluzacek , A. J. Ruckman , Gustav R. Jansen , Sarp Oral

Although LLM training is typically centralized with high-bandwidth interconnects and large compute budgets, emerging methods target communication-constrained training in distributed environments. The model trade-offs introduced by this…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-11-19 Alexander Acker , Soeren Becker , Sasho Nedelkoski , Dominik Scheinert , Odej Kao , Philipp Wiesner

The growth of wireless communication technologies has been producing the intense demand for high-speed, efficient, reliable voice & data communication. As a result, third generation partnership project (3GPP) has implemented next generation…

Networking and Internet Architecture · Computer Science 2014-03-14 Md. Munjure Mowla , Md. Yeakub Ali , Rifat Ahmmed Aoni

In this paper, we study the optimal design for simultaneous wireless information and power transfer (SWIPT) in downlink multiuser orthogonal frequency division multiplexing (OFDM) systems. For information transmission, we consider two types…

Information Theory · Computer Science 2014-02-14 Xun Zhou , Rui Zhang , Chin Keong Ho

The Future wireless communication systems face the challenging task of simultaneously providing high quality of service (QoS) and broadband data transmission, while also minimizing power consumption, latency, and system complexity. Although…

Information Theory · Computer Science 2024-10-03 Amina Darghouthi , Abdelhakim Khlifi , Hmaied Shaiek , Fatma Ben Salah , Belgacem Chibani

Bulk synchronous parallel (BSP) is the de-facto paradigm for distributed DNN training in today's production clusters. However, due to the global synchronization nature, its performance can be significantly influenced by network bottlenecks…

Machine Learning · Computer Science 2022-01-14 Weiyan Wang , Cengguang Zhang , Liu Yang , Kai Chen , Kun Tan

The sixth generation (6G) of wireless networks introduces integrated sensing and communication (ISAC), a technology in which communication and sensing functionalities are inextricably linked, sharing resources across time, frequency, space,…

Signal Processing · Electrical Eng. & Systems 2024-06-28 Silvia Mura , Dario Tagliaferri , Marouan Mizmizi , Umberto Spagnolini , Athina Petropulu

Multiuser resource allocation has recently been recognized as an effective methodology for enhancing the power and spectrum efficiency in OFDM (orthogonal frequency division multiplexing) systems. It is, however, not directly applicable to…

Numerical Analysis · Mathematics 2025-10-20 Ying Jun Zhang

The Data Distribution Service (DDS) is an Object Management Group (OMG) standard for high-performance and real-time systems. DDS is a data-centric middleware based on the publish-subscribe communication pattern and is used in many…

Performance · Computer Science 2022-09-20 Kaleem Peeroo , Peter Popov , Vladimir Stankovic

Large Language Models (LLMs) have become indispensable in numerous real-world applications. However, fine-tuning these models at scale, especially in federated settings where data privacy and communication efficiency are critical, presents…

Machine Learning · Computer Science 2025-06-10 Yao Shu , Wenyang Hu , See-Kiong Ng , Bryan Kian Hsiang Low , Fei Richard Yu

Parameter updating is an important stage in parallelism-based distributed deep learning. Synchronous methods are widely used in distributed training the Deep Neural Networks (DNNs). To reduce the communication and synchronization overhead…

Machine Learning · Computer Science 2020-09-09 Qing Ye , Yuxuan Han , Yanan sun , JIancheng Lv

Over-the-air (OTA) computation has recently emerged as a communication-efficient Federated Learning (FL) paradigm to train machine learning models over wireless networks. However, its performance is limited by the device with the worst SNR,…

Machine Learning · Computer Science 2024-02-05 Muhammad Faraz Ul Abrar , Nicolò Michelusi

Model aggregation, the process that updates model parameters, is an important step for model convergence in distributed deep learning (DDL). However, the parameter server (PS), a popular paradigm of performing model aggregation, causes CPU…

Distributed, Parallel, and Cluster Computing · Computer Science 2022-04-08 Juncheng Gu , Mosharaf Chowdhury , Kang G. Shin , Aditya Akella

Most existing work on adaptive allocation of subcarriers and power in multiuser orthogonal frequency division multiplexing (OFDM) systems has focused on homogeneous traffic consisting solely of either delay-constrained data (guaranteed…

Networking and Internet Architecture · Computer Science 2016-11-15 Meixia Tao , Ying-Chang Liang , Fan Zhang

Task-oriented communication is an emerging paradigm for next-generation communication networks, which extracts and transmits task-relevant information, instead of raw data, for downstream applications. Most existing deep learning (DL)-based…

Signal Processing · Electrical Eng. & Systems 2024-02-07 Hongru Li , Wentao Yu , Hengtao He , Jiawei Shao , Shenghui Song , Jun Zhang , Khaled B. Letaief

The challenge of communication-efficient distributed optimization has attracted attention in recent years. In this paper, a communication efficient algorithm, called ordering-based alternating direction method of multipliers (OADMM) is…

Machine Learning · Computer Science 2022-02-08 Yicheng Chen , Rick S. Blum , Brian M. Sadler

This paper introduces OSC (Orchestrating Cognitive Synergy), a knowledge-aware adaptive collaboration framework designed to enhance cognitive synergy in multi-agent systems with large language models. While prior work has advanced agent…

Artificial Intelligence · Computer Science 2025-09-08 Jusheng Zhang , Yijia Fan , Kaitong Cai , Xiaofei Sun , Keze Wang

Flooding protocols based on concurrent transmissions are regarded as the most reliable way to collect or disseminate data across a multi-hop low-power wireless mesh network. Recent works have shown that such protocols are effective for…

Networking and Internet Architecture · Computer Science 2022-08-30 Michael Baddeley , Yevgen Gyl , Markus Schuss , Xiaoyuan Ma , Carlo Alberto Boano

The demand for high-speed mobile wireless communications is rapidly growing. Orthogonal Frequency Division Multiplexing (OFDM) technology promises to be a key technique for achieving the high data capacity and spectral efficiency…

Information Theory · Computer Science 2013-03-19 I. Pramanik , M. A. F. M. Rashidul Hasan , Rubaiyat Yasmin , M. Sakir Hossain , Ahmed Kamal S. K

Resource allocation in Integrated Sensing and Communication (ISAC) systems is critical for balancing communication and sensing performance. This paper introduces a novel dynamic power allocation strategy for Orthogonal Frequency Division…

Signal Processing · Electrical Eng. & Systems 2025-02-13 Ali Al Khansa , Giyyarpuram Madhusudan , Guillaume Larue , Louis-Adrien Dufrene