Page 441 - 《软件学报》2024年第4期

李彤 等: 传输控制中的确认机制研究 2019

广阔的发挥空间. 例如, 数据中心网络中超高带宽和超低时延, 要求确认机制能够实现及时反馈, 贸然减少 ACK 数
目可能导致性能恶化; 而无线局域网中频谱作为宝贵资源, 应该尽量让给数据报文的传输, 因此减少 ACK 数目将
显著提升性能. 另一方面, 数据中心网络和无线局域网, 相比广域网而言, 通常同时支持端节点和中间交换节点的
定制化, 这种灵活性为确认机制的设计提供了更多的可能性.
另一方面, 内核协议栈最大的问题在于它的更新迭代太慢, 与操作系统绑定, 调试和问题定位非常复杂, 无法
适应网络条件动态变化、新型业务需求多样化的现状. 而用户态协议栈很好地解决了这个问题. 当前最热的用户
态协议框架为 QUIC 协议. 各个大厂都在积极布局, 例如, Microsoft 的 MsQuic [48] , Facebook 的 mvfst [49] , 华为的
hQUIC [50] , 阿里的 XLINK [51] 等, 可以说是百花齐放, 百家争鸣的状态, QUIC 作为 HTTP 3.0 的协议底座, 未来也拥
有无限可能. 可以预见, 依托用户态协议栈这个“罗马广场”, 针对不同的网络环境和差异化的应用需求, 按需确认
机制的设计理念, 将有力推动下一代传输协议的发展.

References:
[1] Jacobson V. Congestion avoidance and control. ACM SIGCOMM Computer Communication Review, 1988, 18(4): 314–329. [doi: 10.
1145/52325.52356]
[2] Luo WM, Lin C, Yan BP. A survey of congestion control in the Internet. Chinese Journal of Computers, 2001, 24(1): 1–18 (in Chinese
with English abstract). [doi: 10.3321/j.issn:0254-4164.2001.01.001]
[3] Wu H, Yu ZH, Cheng G, Hu XY. Encrypted video recognition in large-scale fingerprint database. Ruan Jian Xue Bao/Journal of
Software, 2021, 32(10): 3310–3330 (in Chinese with English abstract). http://www.jos.org.cn/1000-9825/6025.htm [doi: 10.13328/j.cnki.
jos.006025]
[4] Wang JX, Gong H, Chen JE. A cooperant congestion control protocol in high bandwidth-delay product networks. Ruan Jian Xue
Bao/Journal of Software, 2008, 19(1): 125–135 (in Chinese with English abstract). http://www.jos.org.cn/1000-9825/19/125.htm [doi: 10.
3724/SP.J.1001.2008.00125]
[5] Xu CB, Xian YJ, Tang CW, Yang SZ. An active congestion control mechanism for transmission control protocol. Ruan Jian Xue
Bao/Journal of Software, 2008, 19(6): 1533–1545 (in Chinese with English abstract). http://www.jos.org.cn/1000-9825/19/1533.htm [doi:
10.3724/SP.J.1001.2008.01533]
[6] Zhang YM, Xu WQ, Huang J, Wang YM, Shu T, Liu LG. Optimal cross-layer power control and congestion control providing energy
saving for ad hoc networks. Ruan Jian Xue Bao/Journal of Software, 2013, 24(4): 900–914 (in Chinese with English abstract). http://www.
jos.org.cn/1000-9825/4317.htm [doi: 10.3724/SP.J.1001.2013.04317]
[7] Ren FY, Lin C, Liu WD. Congestion control in IP network. Chinese Journal of Computers, 2003, 26(9): 1025–1034 (in Chinese with
English abstract). [doi: 10.3321/j.issn:0254-4164.2003.09.001]
[8] Du XL, Xu K, Li T, Zheng K, Fu ST, Shen M. Traffic control for data center network: State of the art and future research. Chinese
Journal of Computers, 2021, 44(7): 1287–1309 (in Chinese with English abstract). [doi: 10.11897/SP.J.1016.2021.01287]
[9] Li D, Chen GH, Ren FY, Jiang CL, Xu MW. Data center network research progress and trends. Chinese Journal of Computers, 2014,
37(2): 259–274 (in Chinese with English abstract). [doi: 10.3724/SP.J.1016.2014.00259]
[10] Xu K, Zhu M, Lin C. Internet architecture evaluation models, mechanisms and methods. Chinese Journal of Computers, 2012, 35(10):
1985–2006 [doi: 10.3724/SP.J.1016.2012.01985]
[11] Wan K, Luo XF, Jiang Y, Xu K. The flow-oriented scheduling algorithms in SDN system. Chinese Journal of Computers, 2016, 39(6):
1208–1223 (in Chinese with English abstract). [doi: 10.11897/SP.J.1016.2016.01208]
[12] Zhang YC, Xu K, Wang HY, Li Q, Li T, Cao X. Going fast and fair: Latency optimization for cloud-based service chains. IEEE Network,
2018, 32(2): 138–143. [doi: 10.1109/mnet.2017.1700275]
[13] Li T, Liang J, Ding Y, et al. On design and performance of offline finding network. In: Proc. of the 2023 IEEE INFOCOM. 2023. 1–10.
[14] Floyd S, Kohler E. RFC 4341: Profile for datagram congestion control protocol (DCCP) Congestion Control ID 2: TCP-like Congestion
Control. 2006. https://www.rfc-editor.org/rfc/rfc4341.html
[15] Langley A, Riddoch A, Wilk A, Vicente A, Krasic C, Zhang D, Yang F, Kouranov F, Swett I, Iyengar J, Bailey J, Dorfman J, Roskind J,
Kulik J, Westin P, Tenneti R, Shade R, Hamilton R, Vasiliev V, Chang WT, Shi ZY. The QUIC transport protocol: Design and Internet-
scale deployment. In: Proc. of the 2017 Conf. of the ACM Special Interest Group on Data Communication. Los Angeles: ACM, 2017.
183–196. [doi: 10.1145/3098822.3098842]
[16] Palmer M, Appel M, Spiteri K, Chandrasekaran B, Feldmann A, Sitaraman R. The subtle art of not worrying about losses: Optimizing