Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Jin, Yibo, Xu, Yixu, Chen, Yue, Wang, Chengbin, Wang, Tao, Huang, Jiaqi, Zhang, Rongfei, Dong, Yiming, Yan, Yuting, Cheng, Ke, Zhu, Yingjie, Wang, Shulan, Tang, Qianqian, Meng, Shuaishuai, Cheng, Guanxin, Wang, Ze, Miao, Shuyan, Wang, Ketao, Liu, Wen, Yang, Yifan, Zhang, Tong, Wang, Anran, Lu, Chengzhou, Dong, Tiantian, Zhang, Yongsheng, Wang, Zhe, Guo, Hefei, Liu, Hongjie, Lu, Wei, Zhang, Zhengyong
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2508.09035
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918123921211392
author Jin, Yibo
Xu, Yixu
Chen, Yue
Wang, Chengbin
Wang, Tao
Huang, Jiaqi
Zhang, Rongfei
Dong, Yiming
Yan, Yuting
Cheng, Ke
Zhu, Yingjie
Wang, Shulan
Tang, Qianqian
Meng, Shuaishuai
Cheng, Guanxin
Wang, Ze
Miao, Shuyan
Wang, Ketao
Liu, Wen
Yang, Yifan
Zhang, Tong
Wang, Anran
Lu, Chengzhou
Dong, Tiantian
Zhang, Yongsheng
Wang, Zhe
Guo, Hefei
Liu, Hongjie
Lu, Wei
Zhang, Zhengyong
author_facet Jin, Yibo
Xu, Yixu
Chen, Yue
Wang, Chengbin
Wang, Tao
Huang, Jiaqi
Zhang, Rongfei
Dong, Yiming
Yan, Yuting
Cheng, Ke
Zhu, Yingjie
Wang, Shulan
Tang, Qianqian
Meng, Shuaishuai
Cheng, Guanxin
Wang, Ze
Miao, Shuyan
Wang, Ketao
Liu, Wen
Yang, Yifan
Zhang, Tong
Wang, Anran
Lu, Chengzhou
Dong, Tiantian
Zhang, Yongsheng
Wang, Zhe
Guo, Hefei
Liu, Hongjie
Lu, Wei
Zhang, Zhengyong
contents Serving disaggregated large language models has been widely adopted in industrial practice for enhanced performance. However, too many tokens generated in decoding phase, i.e., occupying the resources for a long time, essentially hamper the cloud from achieving a higher throughput. Meanwhile, due to limited on-device resources, the time to first token (TTFT), i.e., the latency of prefill phase, increases dramatically with the growth on prompt length. In order to concur with such a bottleneck on resources, i.e., long occupation in cloud and limited on-device computing capacity, we propose to separate large language model between cloud and devices. That is, the cloud helps a portion of the content for each device, only in its prefill phase. Specifically, after receiving the first token from the cloud, decoupling with its own prefill, the device responds to the user immediately for a lower TTFT. Then, the following tokens from cloud are presented via a speed controller for smoothed TPOT (the time per output token), until the device catches up with the progress. On-device prefill is then amortized using received tokens while the resource usage in cloud is controlled. Moreover, during cloud prefill, the prompt can be refined, using those intermediate data already generated, to further speed up on-device inference. We implement such a scheme P/D-Device, and confirm its superiority over other alternatives. We further propose an algorithm to decide the best settings. Real-trace experiments show that TTFT decreases at least 60%, maximum TPOT is about tens of milliseconds, and cloud throughput increases by up to 15x.
format Preprint
id arxiv_https___arxiv_org_abs_2508_09035
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle P/D-Device: Disaggregated Large Language Model between Cloud and Devices
Jin, Yibo
Xu, Yixu
Chen, Yue
Wang, Chengbin
Wang, Tao
Huang, Jiaqi
Zhang, Rongfei
Dong, Yiming
Yan, Yuting
Cheng, Ke
Zhu, Yingjie
Wang, Shulan
Tang, Qianqian
Meng, Shuaishuai
Cheng, Guanxin
Wang, Ze
Miao, Shuyan
Wang, Ketao
Liu, Wen
Yang, Yifan
Zhang, Tong
Wang, Anran
Lu, Chengzhou
Dong, Tiantian
Zhang, Yongsheng
Wang, Zhe
Guo, Hefei
Liu, Hongjie
Lu, Wei
Zhang, Zhengyong
Distributed, Parallel, and Cluster Computing
Computation and Language
Machine Learning
Serving disaggregated large language models has been widely adopted in industrial practice for enhanced performance. However, too many tokens generated in decoding phase, i.e., occupying the resources for a long time, essentially hamper the cloud from achieving a higher throughput. Meanwhile, due to limited on-device resources, the time to first token (TTFT), i.e., the latency of prefill phase, increases dramatically with the growth on prompt length. In order to concur with such a bottleneck on resources, i.e., long occupation in cloud and limited on-device computing capacity, we propose to separate large language model between cloud and devices. That is, the cloud helps a portion of the content for each device, only in its prefill phase. Specifically, after receiving the first token from the cloud, decoupling with its own prefill, the device responds to the user immediately for a lower TTFT. Then, the following tokens from cloud are presented via a speed controller for smoothed TPOT (the time per output token), until the device catches up with the progress. On-device prefill is then amortized using received tokens while the resource usage in cloud is controlled. Moreover, during cloud prefill, the prompt can be refined, using those intermediate data already generated, to further speed up on-device inference. We implement such a scheme P/D-Device, and confirm its superiority over other alternatives. We further propose an algorithm to decide the best settings. Real-trace experiments show that TTFT decreases at least 60%, maximum TPOT is about tens of milliseconds, and cloud throughput increases by up to 15x.
title P/D-Device: Disaggregated Large Language Model between Cloud and Devices
topic Distributed, Parallel, and Cluster Computing
Computation and Language
Machine Learning
url https://arxiv.org/abs/2508.09035