Research on hardware infrastructure innovation for large language model of telecom operators and RDMA traffic control technology

doi:10.12267/j.issn.2096-5931.2024.02.005

Abstract

Abstract:

Based on the main modes of hardware infrastructure innovation for large language model in the industry, this paper explores the considerations for route selection by telecom operators in this field. This paper presents a congestion control algorithm that supports NO-PFC and does not require the configuration of switches. The algorithm is designed with considerations for the actual networking environment and service scenario. To achieve low latency, the Round-Trip Time (RTT) is employed as the congestion sensing signal, enabling effective regulation of the switch queue length.

Key words: RDMA congestion control, hardware infrastructure innovation for large language model, telecom operator data center network

CLC Number:

TP30
F124

CHE Biyao, ZHANG Yonghang, LIAO Yi, TANG Jian, FAN Xiaoping, ZHAO Jizhuang, LU Gang. Research on hardware infrastructure innovation for large language model of telecom operators and RDMA traffic control technology[J]. Information and Communications Technology and Policy, 2024, 50(2): 26-32.

Add to citation manager EndNote|Ris|BibTeX

URL:

http://ictp.caict.ac.cn/EN/10.12267/j.issn.2096-5931.2024.02.005

http://ictp.caict.ac.cn/EN/Y2024/V50/I2/26

Figures/Tables 1

References 10

[1]	张佳欣. 德媒: 芯片之争, 中国绝非无能为力[N]. 科技日报, 2021-04-09(004).
[2]	ZHU Y, ZHANG M, ERAN H, et al. Congestion control for large-scale RDMA deployments[J]. ACM SIGCOMM Computer Communication Review, 2015, 45(5):523-536. DOI:10.1145/2829988.2787484.
[3]	MITTAL R, LAM V T, DUKKIPATI N, et al. TIMELY: RTT-based congestion control for the datacenter[C]// Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication. New York: ACM, 2015:537-550. DOI:10.1145/2785956.2787510.
[4]	KUMAR G, DUKKIPATI N, JANG K, et al. Swift: delay is simple and effective for congestion control in the datacenter[C]// SIGCOMM’20:Annual Conference of the ACM Special Interest Group on Data Communication on the Applications, Technologies, Architectures, and Protocols for Computer Communication. New York: ACM, 2020:514-528. DOI:10.1145/3387514.3406591.
[5]	LI Y, MIAO R, LIU H, et al. HPCC: high precision congestion control[C]// Proceedings of the ACM Special Interest Group on Data Communication. New York: ACM, 2019:44-58. DOI:10.1145/3341302.3342085.
[6]	BASAT R B, RAMANATHAN S, LI Y, et al. PINT: probabilistic in-band network telemetry[C]. Proceedings of the Annual Conference of the ACM Special Interest Group on Data Communication. New York: ACM, 2020:662-680.
[7]	何春志. 腾讯星脉高性能计算网络:为AI大模型构筑网络底座[EB/OL]. (2023-03-06)[2023-12-20]. https://cloud.tencent.com/developer/article/2234084.
[8]	SHALEV L, AYOUB H, BSHARA N, et al. A cloud-optimized transport protocol for elastic and scalable HPC[J]. IEEE Computer Society, 2020(6):67-73. DOI:10.1109/MM.2020.3016891.
[9]	GIBSON D, HARIHARAN H, LANCE E, et al. Aquila: a unified, low-latency fabric for datacenter networks[C]// Proceedings of 19th USENIX Symposium on Networked Systems Design and Implementation. Seattle: NSDI, 2022:1249-1266.
[10]	WU X G. Reducing job completion time in AI/ML clusters[EB/OL]. (2022-06-09)[2023-12-20]. https://www.broadcom.com/blog/reducing-job-completion-time-in-ai-ml-clusters.

[1]	WANG Hongyu, QIU Ben, WANG Shaopeng. Research on development path of computing and network convergence city [J]. Information and Communications Technology and Policy, 2024, 50(2): 12-17.
[2]	YANG Shuo, WANG Yue, ZHANG Yixing. Research on definition and key technologies of green computility [J]. Information and Communications Technology and Policy, 2024, 50(2): 18-25.
[3]	GUO Liang. Challenge analysis of computing power development in China and suggestions [J]. Information and Communications Technology and Policy, 2024, 50(2): 2-6.
[4]	CHEN Xiaopeng, XU Kexin, LIANG Yudong. Research on AI promoting green energy conservation in data centers [J]. Information and Communications Technology and Policy, 2024, 50(2): 33-39.
[5]	WU Shuai, HAN Zhendong, WANG Jia, CAI Hang. Research on computing power network technology and practice from the perspective of operators [J]. Information and Communications Technology and Policy, 2024, 50(2): 40-45.
[6]	WEN Xiaozhen, CHANG Jinfeng, WU Meixi. Analysis of current situation and trend of comprehensive computing power development [J]. Information and Communications Technology and Policy, 2024, 50(2): 7-11.
[7]	SONG Tianle, LIN Chenhao, GAO Shuxin, ZHAO Junyi, ZHOU Yajie, JI Yingshuai, YANG Minghui, SHEN Chao. Continuous identity authentication of mobile intelligent terminal users via interactive behavior [J]. Information and Communications Technology and Policy, 2024, 50(1): 19-31.
[8]	WANG Yanhong, KONG Ling, FU Yanyan, LI Jingdian. Analysis and suggestions on the standardization path of privacy computing technology [J]. Information and Communications Technology and Policy, 2024, 50(1): 32-36.
[9]	ZHU Yue. Practical difficulties and solutions of differential privacy for personal information protection [J]. Information and Communications Technology and Policy, 2024, 50(1): 37-44.
[10]	HU Guohua, WANG Zhendong, LI Qiang. Research on challenges and countermeasures of data security in new smart cities [J]. Information and Communications Technology and Policy, 2023, 49(8): 31-37.
[11]	SONG Kai, QU Leilei, YANG Mengke. Research on governance strategy of generative Artificial Intelligence [J]. Information and Communications Technology and Policy, 2023, 49(7): 83-88.
[12]	JIA Binbin, WANG Zhongqing, FANG Wei. Thoughts on improving rate of large number modular multiplication in RSA algorithm [J]. Information and Communications Technology and Policy, 2023, 49(6): 84-90.
[13]	HUANG Xiaojie, SONG Wenxuan, ZHANG Hengsheng, XU Fangmin, ZHAO Chenglin. Challenge and design of computing power network for new AIGC computing services [J]. Information and Communications Technology and Policy, 2023, 49(6): 10-16.
[14]	GUO Liang. Overview of data center development [J]. Information and Communications Technology and Policy, 2023, 49(5): 2-8.
[15]	YAN Ximin, ZHANG Linlin. Discussion on security management of data sharing in telecommunications and Internet industries [J]. Information and Communications Technology and Policy, 2023, 49(5): 73-79.