English
Related papers

Related papers: DELTA: Dynamically Optimizing GPU Memory beyond Te…

200 papers

The rapid growth of large language model (LLM) services imposes increasing demands on distributed GPU inference infrastructure. Most existing scheduling systems follow a reactive paradigm, relying solely on the current system state to make…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-09-17 Chengze Du , Zhiwei Yu , Heng Xu , Haojie Wang , Bo liu , Jialong Li

In the evolving landscape of neural network models, one prominent challenge stand out: the significant memory overheads associated with training expansive models. Addressing this challenge, this study delves deep into the Rotated Tensor…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-11-06 Cheng Luo , Tianle Zhong , Geoffrey Fox

Deep learning (DL) jobs use multi-dimensional parallelism, i.e. combining data, model, and pipeline parallelism, to use large GPU clusters efficiently. Long-running jobs may experience changes to their GPU allocation: (i) resource…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-09-27 Marcel Wagenländer , Guo Li , Bo Zhao , Luo Mai , Peter Pietzuch

Deep learning emerges as an important new resource-intensive workload and has been successfully applied in computer vision, speech, natural language processing, and so on. Distributed deep learning is becoming a necessity to cope with…

Distributed, Parallel, and Cluster Computing · Computer Science 2018-05-23 Jilong Xue , Youshan Miao , Cheng Chen , Ming Wu , Lintao Zhang , Lidong Zhou

Training deep learning (DL) models has become a dominant workload in data-centers and improving resource utilization is a key goal of DL cluster schedulers. In order to do this, schedulers typically incorporate placement policies that…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-08-08 Song Bian , Saurabh Agarwal , Md. Tareq Mahmood , Shivaram Venkataraman

Increasingly complex and diverse deep neural network (DNN) models necessitate distributing the execution across multiple devices for training and inference tasks, and also require carefully planned schedules for performance. However,…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-11-28 Zhiqi Lin , Youshan Miao , Guanbin Xu , Cheng Li , Olli Saarikivi , Saeed Maleki , Fan Yang

Task scheduling is a well-studied problem in the context of optimizing the Quality of Service (QoS) of cloud computing environments. In order to sustain the rapid growth of computational demands, one of the most important QoS metrics for…

Distributed, Parallel, and Cluster Computing · Computer Science 2022-05-24 Shreshth Tuli , Giuliano Casale , Nicholas R. Jennings

The widespread use of Deep Neural Networks (DNNs) is limited by high computational demands, especially in constrained environments. GPUs, though effective accelerators, often face underutilization and rely on coarse-grained scheduling. This…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-04-15 Amir Fakhim Babaei , Thidapat Chantem

Large language models (LLMs) are remarked by their substantial computational requirements. To mitigate the cost, researchers develop specialized CUDA kernels, which often fuse several tensor operations to maximize the utilization of GPUs as…

Hardware Architecture · Computer Science 2025-01-15 Guoliang He , Eiko Yoneki

Deep learning (DL) frameworks take advantage of GPUs to improve the speed of DL inference and training. Ideally, DL frameworks should be able to fully utilize the computation power of GPUs such that the running time depends on the amount of…

Machine Learning · Computer Science 2020-12-07 Woosuk Kwon , Gyeong-In Yu , Eunji Jeong , Byung-Gon Chun

Deep learning is slowly, but steadily, hitting a memory bottleneck. While the tensor computation in top-of-the-line GPUs increased by 32x over the last five years, the total available memory only grew by 2.5x. This prevents researchers from…

Machine Learning · Computer Science 2021-04-06 Aashaka Shah , Chao-Yuan Wu , Jayashree Mohan , Vijay Chidambaram , Philipp Krähenbühl

Memory-based Temporal Graph Neural Networks are powerful tools in dynamic graph representation learning and have demonstrated superior performance in many real-world applications. However, their node memory favors smaller batch sizes to…

Machine Learning · Computer Science 2023-07-18 Hongkuan Zhou , Da Zheng , Xiang Song , George Karypis , Viktor Prasanna

Deep learning (DL) compilers rely on cost models and auto-tuning to optimize tensor programs for target hardware. However, existing approaches depend on large offline datasets, incurring high collection costs and offering suboptimal…

Machine Learning · Computer Science 2026-04-15 Chaoyao Shen , Linfeng Jiang , Yixian Shen , Tao Xu , Guoqing Li , Anuj Pathania , Andy D. Pimentel , Meng Zhang

With the fast development of deep neural networks (DNNs), many real-world applications are adopting multiple models to conduct compound tasks, such as co-running classification, detection, and segmentation models on autonomous vehicles.…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-11-30 Fuxun Yu , Shawn Bray , Di Wang , Longfei Shangguan , Xulong Tang , Chenchen Liu , Xiang Chen

Hardware accelerators such as GPUs are required for real-time, low-latency inference with Deep Neural Networks (DNN). However, due to the inherent limits to the parallelism they can exploit, DNNs often under-utilize the capacity of today's…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-04-27 Aditya Dhakal , Sameer G. Kulkarni , K. K. Ramakrishnan

Recent years have seen an increase in the development of large deep learning (DL) models, which makes training efficiency crucial. Common practice is struggling with the trade-off between usability and performance. On one hand, DL…

Machine Learning · Computer Science 2023-12-27 Hongzheng Chen , Cody Hao Yu , Shuai Zheng , Zhen Zhang , Zhiru Zhang , Yida Wang

Communication scheduling aims to reduce communication bottlenecks in data parallel training (DP) by maximizing the overlap between computation and communication. However, existing schemes fall short due to three main issues: (1) hard data…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-03-24 Lin Meng , Yuzhong Sun

The Nvidia GPU architecture has introduced new computing elements such as the \textit{tensor cores}, which are special processing units dedicated to perform fast matrix-multiply-accumulate (MMA) operations and accelerate \textit{Deep…

Distributed, Parallel, and Cluster Computing · Computer Science 2019-03-12 Roberto Carrasco , Raimundo Vega , Cristóbal A. Navarro

Training large AI models such as LLMs and DLRMs costs massive GPUs and computing time. The high training cost has become only affordable to big tech companies, meanwhile also causing increasing concerns about the environmental impact. This…

Machine Learning · Computer Science 2024-12-03 Zi Yang , Ziyue Liu , Samridhi Choudhary , Xinfeng Xie , Cao Gao , Siegfried Kunzmann , Zheng Zhang

In recent years, the development of specialized edge computing devices has significantly increased, driven by the growing demand for AI models. These devices, such as the NVIDIA Jetson series, must efficiently handle increased data…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-06-03 Ashiyana Abdul Majeed , Mahmoud Meribout