数据中心差异化时延满足率优化调度
作者:
基金项目:

天津市教委科研计划(2019KJ094)


Optimization Scheduling of Differential Delay Satisfaction Rate in Data Center
Author:
  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [32]
  • |
  • 相似文献
  • |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    光交换网络数据传输时根据数据性质不同, 用户对时延要求也有所不同, 如何在保证光交换调度效率的同时满足差异化时延需求, 是决定网络性能的一个重要因素. 目前针对光网络调度的研究主要基于逐个时隙或基于分组进行调度. 前者没有考虑重配置开销的问题, 无法处理大规模数据交换, 后者忽略了不同延迟以及QoS保证的需要. 为了解决数据中心光交换数据时延需求不同的问题, 本文提出两种新的调度算法SDF (stringent delay first)和m-SDF (m-order stringent delay first), 将不同数据包的差异化时延需求、配置顺序、重配置开销和加速比作为考虑因素, 在流量调度时采用贪心策略, 每次选择对时延最为敏感的数据包进行优先调度以满足时延需求. 所提算法在保证投递率的前提下, 能最大程度满足更多数据包的传输时延. 仿真实验表明两个算法具有较高的时延满足率, 证明了调度算法的有效性.

    Abstract:

    Users have differential delay requirements for data transmission in optical switching networks according to the nature of the data. In addition, it is an important factor to determine the network performance that how to satisfy the differential delay requirement while ensuring the scheduling efficiency of optical switching. At present, research on optical network scheduling is mainly based on single slot time or groups. The former does not consider the reconfi-guration overhead and thus cannot handle large-scale data exchange, while the latter ignores the need for different delays and QoS guarantees. In order to solve the problem of different data delay requirements for optical switching in data centers, this study proposes two new scheduling algorithms including stringent delay first (SDF) and m-order stringent delay first (m-SDF). The study also takes differential delay requirements, configuration order, reconfiguration overhead, and acceleration ratio of different data packets into account and adopts a greedy strategy in data scheduling. Furthermore, the study chooses a data packet that is the most sensitive to delay each time for priority scheduling, so as to meet the delay requirement. The proposed algorithms can maximize the transmission delay of more data packets under the premise of guaranteeing the delivery rate. The simulation results show that the two algorithms have a high delay satisfaction rate, which proves the effectiveness of the scheduling algorithms.

    参考文献
    [1] Lewis GA. Cloud computing. Computer, 2017, 50(5): 8–9. [doi: 10.1109/MC.2017.141
    [2] 宁士勇. 虚拟化云计算数据中心资源节能调度算法研究. 计算机应用研究, 2021, 38(4): 1088–1091. [doi: 10.19734/j.issn.1001-3695.2020.02.0037
    [3] Schleicher JM, Vögler M, Dustdar S, et al. Enabling a smart city application ecosystem: Requirements and architectural aspects. IEEE Internet Computing, 2016, 20(2): 58–65. [doi: 10.1109/MIC.2016.39
    [4] Lewis FL. Wireless sensor networks. In: Cook DJ, Das SK, eds. Smart Environments: Technologies, Protocols, and Applications. Hoboken: John Wiley, 2005. 11–46.
    [5] Fuerst C, Schmid S, Suresh L, et al. Kraken: Online and elastic resource reservations for cloud datacenters. IEEE/ACM Transactions on Networking, 2018, 26(1): 422–435. [doi: 10.1109/TNET.2017.2782006
    [6] Alhaidari F, Balharith TZ. Enhanced round-robin algorithm in the cloud computing environment for optimal task scheduling. Computers, 2021, 10(5): 63. [doi: 10.3390/computers10050063
    [7] Qing Y. Data center network architecture design for cloud computing. Converter, 2021, 2021(1): 23–29
    [8] Maelah R, Al Lami MFF, Ghassan G. Usefulness of management accounting information in decision making among SMEs: The moderating role of cloud computing. Asia-Pasific Management Accounting Journal, 2021, 16(1): 59–92.
    [9] Zhang ZX, Liu FZ, Yao B. Cloud computing data center system, gateway, server, and packet processing method: US, 20210320872. 2021-06-25.
    [10] Zhan ZH, Liu XF, Gong YJ, et al. Cloud computing resource scheduling and a survey of its evolutionary approaches. ACM Computing Surveys, 2015, 47(4): 63
    [11] CTI论坛. 思科全球云指数: 预测和方法2015–2020年. http://ec.ctiforum.com/jishu/qiye/qiyetongxinjishu/yunjisuan/baipishu/503007.html (2017-02-10).
    [12] Towles B, Dally WJ. Guaranteed scheduling for switches with configuration overhead. IEEE/ACM Transactions on Networking, 2003, 11(5): 835–847. [doi: 10.1109/TNET.2003.818190
    [13] Wu B, Yeung KL, Hamdi M, et al. Minimizing internal speedup for performance guaranteed switches with optical fabrics. IEEE/ACM Transactions on Networking, 2009, 17(2): 632–645. [doi: 10.1109/TNET.2008.926501
    [14] Kulkarni J, Lee E, Singh M. Minimum Birkhoff-von Neumann decomposition. Proceedings of the 19th International Conference on Integer Programming and Combinatorial Optimization. Waterloo: Springer, 2017. 343–354.
    [15] Benzi M, Uçar B. Preconditioning techniques based on the Birkhoff-von Neumann decomposition. Computational Methods in Applied Mathematics, 2017, 17(2): 201–215. [doi: 10.1515/cmam-2016-0040
    [16] Chang CS, Chen WJ, Huang HY. Birkhoff-von Neumann input buffered crossbar switches. Proceedings of the 2000 Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies. Tel Aviv: IEEE, 2000. 1614–1623.
    [17] Li J, Ansari N. Enhanced Birkhoff-von Neumann decomposition algorithm for input queued switches. IEE Proceedings-Communications, 2001, 148(6): 339–342. [doi: 10.1049/ip-com:20010618
    [18] Wu B, Yeung KL. NXG05–6: Minimum delay scheduling in scalable hybrid electronic/optical packet switches. Proceedings of the 2006 IEEE Globecom. San Francisco: IEEE, 2006. 1–5.
    [19] van Hautegem K, Rogiest W, Bruneel H. Fill the void: Improved scheduling for optical switching. Proceedings of the 2015 27th International Teletraffic Congress. Ghent: IEEE, 2015. 82–88.
    [20] Jhunjhunwala PR, Maguluri ST. Low-complexity switch scheduling algorithms: Delay optimality in heavy traffic. IEEE/ACM Transactions on Networking, 2022, 30(1): 464–473. [doi: 10.1109/TNET.2021.3116606
    [21] Jie X, Yeung KL. Iterative multicast scheduling algorithm for input-queued switch with variable packet size. Proceedings of the 2017 30th IEEE Canadian Conference on Electrical and Computer Engineering. Windsor: IEEE, 2017. 1–4.
    [22] Zhu KK, Shen GB, Jiang Y, et al. Differentiated transmission based on traffic classification with deep learning in datacenter. Proceedings of the 2020 IFIP Networking Conference. Paris: IEEE, 2020. 599–603.
    [23] Varshney S, Singh S. A survey on resource scheduling algorithms in cloud computing. International Journal of Applied Engineering Research, 2018, 13(9): 6839–6845
    [24] 姚晓魁. 时间调度网络流量规划技术研究[硕士学位论文]. 成都: 电子科技大学, 2017.
    [25] 唐旭, 王飞, 李彤, 等. 数据中心实时交换系统的研究与实现. 计算机科学, 2017, 44(6A): 459–462, 490. [doi: 10.11896/j.issn.1002-137X.2017.6A.103
    [26] 王耀民, 王霞, 董易, 等. 面向云数据中心的多业务差异化流量管理优化策略. 通信学报, 2019, 40(11): 45–56. [doi: 10.11959/j.issn.1000-436x.2019215
    [27] Zhang BX, Wan XL, Luo JZ, et al. A nearly optimal packet scheduling algorithm for input queued switches with deadline guarantees. IEEE Transactions on Computers, 2015, 64(6): 1548–1563
    [28] Anderson T, Owicki S, Saxe J, et al. High speed switch scheduling for local area networks. ACM Transaction on Computer Systems, 1993, 11(4): 319–352.
    [29] Kar K, Stiliadis D, Lakshman TV, et al. Scheduling algorithms for optical packet fabrics. IEEE Journal on Selected Areas in Communications, 2003, 21(7): 1143–1155. [doi: 10.1109/JSAC.2003.815911
    [30] Agranat AJ. Electroholographic wavelength selective crossconnect. Proceedings of the 1999 Digest of the LEOS Summer Topical Meetings: Nanostructures and Quantum Dots/WDM Components/VCSELs and Microcavaties/RF Photonics for CATV and HFC Systems. San Diego: IEEE, 1999. II61–II62.
    [31] Wu B, Yeung KL. Traffic scheduling in non-blocking optical packet switches with minimum delay. Proceedings of IEEE Global Telecommunications Conference. Saint Louis: IEEE, 2005. 2041–2045.
    [32] Munkres J. Algorithms for the assignment and transportation problems. Journal of the Society for Industrial and Applied Mathematics, 1957, 5(1): 32–38. [doi: 10.1137/0105003
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

李耀芳,刘国瑞,洪姣.数据中心差异化时延满足率优化调度.计算机系统应用,2023,32(4):77-85

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2022-09-20
  • 最后修改日期:2022-10-19
  • 在线发布日期: 2023-02-24
文章二维码
您是第11418752位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京海淀区中关村南四街4号 中科院软件园区 7号楼305房间,邮政编码:100190
电话:010-62661041 传真: Email:csa (a) iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号