Inferring Coflow Size Mechanism Based on ELM in Data Center Network

doi:10.3778/j.issn.1673-9418.2003024

Abstract

Abstract:

In recent years, Coflow scheduling has become a research hotspot in data center network. However, it is difficult for existing non-clairvoyant Coflow schedulers to infer the task information quickly. Therefore, small tasks cannot be scheduled in time, making it fail to minimize the average task completion time. Data center network requires effective inferring model to improve the accuracy and sensitivity in inferring Coflow size. This paper proposes a machine learning based Coflow size inferring model (MLcoflow), which utilizes an extreme learning machine (ELM) to establish Coflow size inferring model to minimize training error, and uses the incomplete infor-mation in training to increase the sensitivity. Experiment results show that the accurate score and sensitivity of ELM method are 19.8% and 10.2% higher than other algorithms on average, respectively. This paper compares several schedulers by simulation. MLcoflow-based scheduler reduces the average task completion time by up to 20.1%.

Key words: data center, Coflow size, Coflow scheduling, inferring model, extreme learning machine (ELM)

摘要：

近年来研究流簇（Coflow）为单位的调度策略成为改进数据中心网络的新热点。然而现有的信息未知流簇调度器难以快速地推理任务级信息，导致小任务不能被及时调度，以及平均任务完成时间无法最小化。因此数据中心网络需要更加高效的推理模型提升流簇大小判断的准确性和敏感性。提出了一种基于机器学习的流簇大小推理模型（MLcoflow），利用极限学习机（ELM）以最小训练误差为求解目标建立推理模型，并且使用不完全信息建模以提升敏感度。实验证明与其他算法相比，ELM方法的准确性评分平均高出19.8%，敏感度平均高出10.2%。通过仿真模拟对比了几种调度器，基于MLcoflow的调度器将平均任务完成时间降低了20.1%。

关键词: 数据中心, 流簇大小, 流簇调度, 推理模型, 极限学习机（ELM）

YE Jin, XIE Ziqi, XIAO Qingyu, SONG Ling, LI Xiaohuan. Inferring Coflow Size Mechanism Based on ELM in Data Center Network[J]. Journal of Frontiers of Computer Science and Technology, 2021, 15(2): 261-269.

叶进, 谢紫琪, 肖庆宇, 宋玲, 李晓欢. 数据中心网络中基于ELM的流簇大小推理机制[J]. 计算机科学与探索, 2021, 15(2): 261-269.

References

[1] CHOWDHURY M, ZAHARIA M, MA J, et al. Managing data transfers in computer clusters with orchestra[J]. ACM SIGCOMM Computer Communication Review, 2011, 41(4): 98-109.
[2] AL-FARES M, LOUKISSAS A, VAHDAT A. A scalable, commodity data center network architecture[C]//Proceedings of the ACM SIGCOMM 2008 Conference on Applications, Seattle, Aug 17-22, 2008. New York: ACM, 2008: 63-74.
[3] CHOWDHURY M, STOICA I. Coflow: a networking abstraction for cluster applications[C]//Proceedings of the 11th ACM Workshop on Hot Topics in Networks, Redmond, Oct 29-30, 2012. New York: ACM, 2012: 31-36.
[4] CHOWDHURY M, STOICA I. Efficient Coflow scheduling without prior knowledge[J]. ACM SIGCOMM Computer Communication Review, 2015, 45(5): 393-406.
[5] WANG S, ZHANG J, HUANG T, et al. Leveraging multiple Coflow attributes for information-agnostic Coflow scheduling [C]//Proceedings of the 2017 IEEE International Conference on Communications, Paris, May 21-25, 2017. Piscataway: IEEE, 2017: 1-6.
[6] JAJOO A, HU Y C, LIN X. Your Coflow has many flows: sampling them for fun and speed[C]//Proceedings of the 2019 USENIX Annual Technical Conference, Renton, Jul 10-12, 2019. Berkeley: USENIX Association, 2019: 833-848.
[7] CHOWDHURY M, ZHONG Y, STOICA I. Efficient coflow scheduling with varys[J]. ACM SIGCOMM Computer Communication Review, 2014, 44(4): 443-454.
[8] AGARWAL S, RAJAKRISHNAN S, NARAYAN A, et al. Sincronia: near-optimal network design for coflows[C]//Proceedings of the 2018 Conference of the ACM Special Interest Group on Data Communication, Budapest, Aug 20-25, 2018. New York: ACM, 2018: 16-29.
[9] HU Z Y, LI D S, LI Z Y. Recent advances in datacenter flow scheduling[J]. Journal of Computer Research and Development, 2018, 55(9): 96-106.
胡智尧, 李东升, 李紫阳. 数据中心网络流调度技术前沿进展[J]. 计算机研究与发展, 2018, 55(9): 96-106.
[10] LIU S, HUANG J, ZHOU Y, et al. Task-aware TCP in data center networks[J]. IEEE/ACM Transactions on Networking, 2019, 27(1): 389-404.
[11] POUPART P, CHEN Z, JAINI P, et al. Online flow size prediction for improved network routing[C]//Proceedings of the IEEE 24th International Conference on Network Protocols, Singapore, Nov 8-11, 2016. Washington: IEEE Computer Society, 2016: 1-6.
[12] DUKI? V, JYOTHI S A, KARLA? B, et al. Is advance knowledge of flow sizes a plausible assumption?[C]//Proceedings of the 16th USENIX Symposium on Networked Systems Design and Implementation, Boston, Feb 26-28, 2019. Berkeley: USENIX Association, 2019: 565-580.
[13] YE K, ANTON FEENSTRA K, HERINGA J, et al. Multi-RELIEF: a method to recognize specificity determining residues from multiple sequence alignments using a machine-learning approach for feature weighting[J]. Bioinformatics, 2007, 24(1): 18-25.
[14] HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine: a new learning scheme of feedforward neural networks[J]. Neural networks, 2004, 2: 985-990.
[15] ZHANG F P, CHEN L, ZHANG J J. Measurement and performance bottleneck analysis method for large-scale complex networks[J]. Journal of Frontiers of Computer Science and Technology, 2017, 11(2): 262-270.
张飞朋, 陈琳, 张京京. 面向大规模复杂网络测量和性能瓶颈分析方法[J]. 计算机科学与探索, 2017, 11(2): 262-270.
[16] WU X, KUMAR V, QUINLAN J R, et al. Top 10 algorithms in data mining[J]. Knowledge and Information Systems, 2008, 14(1): 1-37.
[17] CHEN T, GUESTRIN C. Xgboost: a scalable tree Boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, Aug 13-17, 2016. New York: ACM, 2016: 785-794.
[18] Coflow-Benchmark[EB/OL]. (2015-08-06)[2020-01-06]. https://github.com/Coflow/Coflow-benchmark.
[19] YANG Y M, LIU X. A re-examination of text categorization methods[C]//Proceedings of the 22nd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, Berkeley, Aug 15-19, 1999. New York: ACM, 1999: 42-49.
[20] ALIZADEH M, EDSALL T. On the data path performance of leaf-spine datacenter fabrics[C]//Proceedings of the 2013 IEEE 21st Annual Symposium on High-Performance Interconnects, Santa Clara, Aug 1-23, 2013: 71-74.