English
Related papers

Related papers: TokenScale: Timely and Accurate Autoscaling for Di…

200 papers

Serving Large Language Models (LLMs) is a GPU-intensive task where traditional autoscalers fall short, particularly for modern Prefill-Decode (P/D) disaggregated architectures. This architectural shift, while powerful, introduces…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-08-28 Rongzhi Li , Ruogu Du , Zefang Chu , Sida Zhao , Chunlei Han , Zuocheng Shi , Yiwen Shao , Huanle Han , Long Huang , Zherui Liu , Shufan Liu

Prefill/decode disaggregation is increasingly adopted in LLM serving to improve the latency-throughput tradeoff and meet strict TTFT and TPOT SLOs. However, LLM inference remains energy-hungry: autoscaling alone is too coarse-grained to…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-04-07 Omar Basit , Yunzhao Liu , Z. Jonny Kong , Y. Charlie Hu

DistServe improves the performance of large language models (LLMs) serving by disaggregating the prefill and decoding computation. Existing LLM serving systems colocate the two phases and batch the computation of prefill and decoding across…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-06-07 Yinmin Zhong , Shengyu Liu , Junda Chen , Jianbo Hu , Yibo Zhu , Xuanzhe Liu , Xin Jin , Hao Zhang

Real-time LLM interactions demand streamed token generations, where text tokens are progressively generated and delivered to users while balancing two objectives: responsiveness (i.e., low time-to-first-token) and steady generation…

Machine Learning · Computer Science 2025-10-06 Junyi Chen , Chuheng Du , Renyuan Liu , Shuochao Yao , Dingtian Yan , Jiang Liao , Shengzhong Liu , Fan Wu , Guihai Chen

Monolithic serving with chunked prefill improves GPU utilization by batching prefill and decode together, but suffers from fine-grained phase interference. Engine-level prefill-decode (PD) disaggregation avoids interference but incurs…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-08-08 Xiaoxiang Shi , Colin Cai , Junjia Du , Zhihao Jia

Large Multimodal Models (LMMs) extend Large Language Models (LLMs) by handling diverse inputs such as images, audio, and video, but at the cost of adding a multimodal encoding stage that increases both computational and memory overhead.…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-07-01 Gursimran Singh , Xinglu Wang , Yifan Hu , Timothy Yu , Linzi Xing , Wei Jiang , Zhefeng Wang , Xiaolong Bai , Yi Li , Ying Xiong , Yong Zhang , Zhenan Fan

Pipeline parallelism has emerged as a predominant approach for deploying large language models (LLMs) across distributed nodes, owing to its lower communication overhead compared to tensor parallelism. While demonstrating high throughput in…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-05-29 Tianyu Guo , Xianwei Zhang , Jiangsu Du , Zhiguang Chen , Nong Xiao , Yutong Lu

To meet strict Service-Level Objectives (SLOs),contemporary Large Language Models (LLMs) decouple the prefill and decoding stages and place them on separate GPUs to mitigate the distinct bottlenecks inherent to each phase. However, the…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-03-10 Junhan Liao , Minxian Xu , Wanyi Zheng , Yan Wang , Kejiang Ye , Rajkumar Buyya , Chengzhong Xu

Two widely adopted techniques for LLM inference serving systems today are hybrid batching and disaggregated serving. A hybrid batch combines prefill and decode tokens of different requests in the same batch to improve resource utilization…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-01-21 Amna Masood , Pratishtha Gaur , Nuwan Jayasena

LLM inference must meet strict latency SLOs (e.g., 100 ms P99 time-between-tokens) while maximizing goodput. Yet, real-world variability in prompt and response lengths skews compute-intensive prefill and memory-bound decode phases, making…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-05-23 Chaoyi Ruan , Yinhe Chen , Dongqi Tian , Yandong Shi , Yongji Wu , Jialin Li , Cheng Li

As Large Language Models (LLMs) transition from research environments to production deployments, evaluating their performance against strict Service Level Objectives (SLOs) has become critical. However, current evaluation methodologies…

Artificial Intelligence · Computer Science 2026-05-27 Ashok Chandrasekar , Jason Kramberger

Modern LLM serving systems must sustain high throughput while meeting strict latency SLOs across two distinct inference phases: compute-intensive prefill and memory-bound decode phases. Existing approaches either (1) aggregate both phases…

Machine Learning · Computer Science 2025-11-10 Lei Gao , Chaoyi Jiang , Hossein Entezari Zarch , Daniel Wong , Murali Annavaram

Different from traditional Large Language Model (LLM) serving that colocates the prefill and decode stages on the same GPU, disaggregated serving dedicates distinct GPUs to prefill and decode workload. Once the prefill GPU completes its…

Performance · Computer Science 2026-01-15 Jiaxi Li , Yue Zhu , Eun Kyung Lee , Klara Nahrstedt

Large language model (LLM) serving is now limited by the key-value (KV) cache. During decode, each new token rereads prior KV state, so attention becomes a bandwidth- and capacity-heavy memory task. HBM-PIM helps by moving attention closer…

Hardware Architecture · Computer Science 2026-05-08 Zhuoran Li , Zhuohang Bian , Zihao Huang , Guangyu Sun , Yun Liang , Youwei Zhuo

Prefix caching is crucial to accelerate multi-turn interactions and requests with shared prefixes. At the cluster level, existing prefix caching systems are tightly coupled with request scheduling to optimize cache efficiency and…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-08-26 Bingyang Wu , Zili Zhang , Yinmin Zhong , Guanzhe Huang , Yibo Zhu , Xuanzhe Liu , Xin Jin

Efficient parallelization of Large Language Models (LLMs) with long sequences is essential but challenging due to their significant computational and memory demands, particularly stemming from communication bottlenecks in attention…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-12-31 Zongwu Wang , Fangxin Liu , Mingshuai Li , Li Jiang

Large Language Models (LLMs) are becoming the backbone of modern cloud services, yet their inference costs are dominated by GPU energy. Unlike traditional GPU workloads, LLM inference has two stages with different characteristics: the…

Performance · Computer Science 2025-08-25 Qunyou Liu , Darong Huang , Marina Zapater , David Atienza

Multimodal large language models (MLLMs) extend LLMs with visual understanding through a three-stage pipeline: multimodal preprocessing, vision encoding, and LLM inference. While these stages enhance capability, they introduce significant…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-12-22 Lingxiao Zhao , Haoran Zhou , Yuezhi Che , Dazhao Cheng

Large Language Models (LLMs) are increasingly deployed in both latency-sensitive online services and cost-sensitive offline workloads. Co-locating these workloads on shared serving instances can improve resource utilization, but directly…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-12-01 Siyu Wu , Zihan Tang , Yuting Zeng , Hui Chen , Guiguang Ding , Tongxuan Liu , Ke Zhang , Hailong Yang

Large language models (LLMs) are increasingly deployed in AI infrastructure, driving the need for high throughput, resource efficient serving systems. Disaggregated LLM serving, which separates prompt prefill from auto-regressive decode,…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-03-11 Yiyuan He , Minxian Xu , Jingfeng Wu , Jianmin Hu , Chong Ma , Min Shen , Le Chen , Chengzhong Xu , Lin Qu , Kejiang Ye
‹ Prev 1 2 3 10 Next ›