Enhancing Fairness for Approximate Weighted Fair Queueing With a Single Queue
Authors: 
Wei Chen, Ye Tian, Xin Yu, Bowen Zheng, Xinming ZhangAuthors Info & Claims
Wei Chen, Ye Tian, Xin Yu, Bowen Zheng, Xinming ZhangAuthors Info & ClaimsPages 3901 - 3915
Abstract
Weighted fair queueing (WFQ) is an essential strategy for enforcing bandwidth guarantee and isolation in high-speed networks. Unfortunately, implementing the original WFQ packet scheduling algorithm on today’s commodity switch hardware is challenging due to the prohibitive complexity. Approximate WFQ packet schedulers, which work with the cheap and widely available First-In First-Out (FIFO) queues, have been proposed as an alternative in recent years. In this paper, we show that both the ideal and the approximate WFQ packet schedulers are unable to allocate bandwidths to TCP flows fairly, because of the bursty nature of the TCP traffic. Furthermore, we find that the representative approximate WFQ schedulers further degrade the scheduling fairness, due to their excessive packet drops. To address these issues, we present novel approximate WFQ packet scheduling algorithms in this paper. Our initial design, namely SQ-WFQ, imposes the minimum hardware requirement by using one single FIFO queue, and effectively reduces the excessive packet drops. Extended from SQ-WFQ, we propose the SQ-EWFQ packet scheduling algorithm. SQ-EWFQ inherits all the merits of SQ-WFQ, and is adaptive to the bursty TCP traffic by tolerating short-term packet bursts, while enforcing a long-term fairness among the TCP flows. We have implemented our proposed schedulers on commodity hardware programmable switches, and achieve line rate packet scheduling with them. Experiment results from a real-world testbed and large-scale simulations show that SQ-WFQ and SQ-EWFQ outperform the state-of-the-art approximate schedulers regarding the scheduling fairness, and SQ-EWFQ allocates bandwidths to TCP flows more fairly than SQ-WFQ and other existing solutions.
References
[1]
A. K. Parekh and R. G. Gallager, “A generalized processor sharing approach to flow control in integrated services networks: The single-node case,” IEEE/ACM Trans. Netw., vol. 1, no. 3, pp. 344–357, Jun. 1993.
[2]
H. J. Chao, Y.-R. Jenq, X. Guo, and C. H. Lam, “Design of packet-fair queuing schedulers using a RAM-based searching engine,” IEEE J. Sel. Areas Commun., vol. 17, no. 6, pp. 1105–1126, Jun. 1999.
[3]
A. Ioannou and M. G. H. Katevenis, “Pipelined heap (priority queue) management for advanced scheduling in high-speed networks,” IEEE/ACM Trans. Netw., vol. 15, no. 2, pp. 450–461, Apr. 2007.
[4]
A. Sivaraman et al., “Programmable packet scheduling at line rate,” in Proc. ACM SIGCOMM Conf., Florianopolis, Brazil, Aug. 2016, pp. 44–57.
[5]
V. Shrivastav, “Fast, scalable, and programmable packet scheduler in hardware,” in Proc. ACM Special Interest Group Data Commun., Beijing, China, Aug. 2019, pp. 367–379.
[6]
Z. Yu et al., “Programmable packet scheduling with a single queue,” in Proc. ACM SIGCOMM Conf., Aug. 2021, pp. 179–193.
[7]
N. K. Sharma et al., “Programmable calendar queues for high-speed packet scheduling,” in Proc. NSDI, Santa Clara, CA, USA, Feb. 2020, pp. 685–700.
[8]
A. G. Alcoz, A. Dietmuller, and L. Vanbever, “SP-PIFO: Approximating push-in first-out behaviors using strict-priority queues,” in Proc. NSDI, Santa Clara, CA, USA, Feb. 2020, pp. 59–76.
[9]
N. K. Sharma, M. Liu, K. Atreya, A. Krishnamurthy, and A. Sivaraman, “Approximating fair queueing on reconfigurable switches,” in Proc. NSDI, Renton, WA, USA, Apr. 2018, pp. 1–16.
[10]
P. Gao, A. Dalleggio, Y. Xu, and H. J. Chao, “Gearbox: A hierarchical packet scheduler for approximate weighted fair queuing,” in Proc. NSDI, Renton, WA, USA, Apr. 2022, pp. 551–565.
[11]
Intel Tofino Series. Accessed: Nov. 30, 2022. [Online]. Available: https://intel.com/content/www/us/en/products/details/network-io/programmable-ethernet-switch/tofino-series.html
[12]
P4_16 Portable Switch Architecture (PSA), P4.org Archit. Work. Group, Dec. 2022. [Online]. Available: https://staging.p4.org/p4-spec/docs/PSA-v1.2.html
[13]
A. Demers, S. Keshav, and S. Shenker, “Analysis and simulation of a fair queueing algorithm,” in Proc. Commun. Architectures Protocols, Austin, TX, USA, Aug. 1989, pp. 1–12.
[14]
M. Alizadeh et al., “PFabric: Minimal near-optimal datacenter transport,” in Proc. ACM SIGCOMM Conf. SIGCOMM, Hong Kong, Aug. 2013, pp. 435–446.
[15]
W. Bai, L. Chen, K. Chen, D. Han, C. Tian, and H. Wang, “Information-agnostic flow scheduling for commodity data centers,” in Proc. NSDI, Oakland, CA, USA, May 2015, pp. 455–468.
[16]
V. Dukic, S. A. Jyothi, B. Karlas, M. Owaida, C. Zhang, and A. Singla, “Is advance knowledge of flow sizes a plausible assumption?,” in Proc. NSDI, Boston, MA, USA, Feb. 2015, pp. 565–580.
[17]
P. Goyal, H. M. Vin, and H. Cheng, “Start-time fair queueing: A scheduling algorithm for integrated services packet switching networks,” IEEE/ACM Trans. Netw., vol. 5, no. 5, pp. 690–704, May 1997.
[18]
L. E. Schrage and L. W. Miller, “The queue M/G/1 with the shortest remaining processing time discipline,” Oper. Res., vol. 14, no. 4, pp. 670–684, Aug. 1966.
[19]
Z. Wang, J. Ye, D. Lin, Y. Chen, and J. C. S. Lui, “Approximate and deployable shortest remaining processing time scheduler,” IEEE/ACM Trans. Netw., vol. 30, no. 3, pp. 1368–1381, Jun. 2022.
[20]
C. L. Liu and J. W. Layland, “Scheduling algorithms for multiprogramming in a hard-real-time environment,” J. ACM, vol. 20, no. 1, pp. 46–61, Jan. 1973.
[21]
P. E. McKenney, “Stochastic fairness queueing,” in Proc. IEEE INFOCOM, San Francisco, CA, USA, Jun. 1990, pp. 733–740.
[22]
M. Shreedhar and G. Varghese, “Efficient fair queuing using deficit round-robin,” IEEE/ACM Trans. Netw., vol. 4, no. 3, pp. 375–385, Jun. 1996.
[23]
J. Liu, J. Huang, Z. Li, Y. Li, J. Wang, and T. He, “Achieving per-flow fairness and high utilization with limited priority queues in data center,” IEEE/ACM Trans. Netw., vol. 30, no. 5, pp. 2374–2387, Oct. 2022.
[24]
R. Mahajan, S. Floyd, and D. Wetherall, “Controlling high-bandwidth flows at the congested router,” in Proc. 9th Int. Conf. Netw. Protocols. ICNP, Riverside, CA, USA, 2001, pp. 192–201.
[25]
R. Pan, L. Breslau, B. Prabhakar, and S. Shenker, “Approximate fairness through differential dropping,” SIGCOMM CCR, vol. 33, no. 2, pp. 23–39, 2003.
[26]
I. Stoica, S. Shenker, and H. Zhang, “Core-stateless fair queueing: A scalable architecture to approximate fair bandwidth allocations in high-speed networks,” IEEE/ACM Trans. Netw., vol. 11, no. 1, pp. 33–46, Feb. 2003.
[27]
Z. Yu, J. Wu, V. Braverman, I. Stoica, and X. Jin, “Twenty years after: Hierarchical core-stateless fair queueing,” in Proc. NSDI, Boston, MA, USA, Apr. 2021, pp. 29–45.
[28]
R. MacDavid, X. Chen, and J. Rexford, “Scalable real-time bandwidth fairness in switches,” IEEE/ACM Trans. Netw., vol. 32, pp. 1423–1434, 2024.
[29]
L. Yu, J. Sonchack, and V. Liu, “Cebinae: Scalable in-network fairness augmentation,” in Proc. ACM SIGCOMM Conf., Amsterdam, Netherlands, Aug. 2022, pp. 219–232.
[30]
B. Stephens, A. Singhvi, A. Akella, and M. Swift, “Titan: Fair packet scheduling for commodity multiqueue NICs,” in Proc. ATC, Santa Clara, CA, USA, Jul. 2017, pp. 431–444.
[31]
B. Stephens, A. Akella, and M. Swift, “Loom: Flexible and efficient NIC packet scheduling,” in Proc. NSDI, Boston, MA, USA, Feb. 2019, pp. 33–46.
[32]
A. Saeed et al., “Eiffel: Efficient and flexible software packet scheduling,” in Proc. NSDI, Boston, MA, USA, Feb. 2019, pp. 17–31.
[33]
T. Flach et al., “Reducing web latency: The virtue of gentle aggression,” in Proc. ACM SIGCOMM Conf. SIGCOMM, Hong Kong, Aug. 2013, pp. 159–170.
[34]
NetBench. Accessed: Feb. 20, 2023. [Online]. Available: https://github.com/ndal-eth/
[35]
S. Kassing, A. Valadarsky, G. Shahaf, M. Schapira, and A. Singla, “Beyond fat-trees without antennae, mirrors, and disco-balls,” in Proc. Conf. ACM Special Interest Group Data Commun., Los Angeles, CA, USA, Aug. 2017, pp. 281–294.
[36]
Iperf3. Accessed: Feb. 20, 2023. [Online]. Available: https://iperf.fr/
[37]
T. Henderson, S. Floyd, A. Gurtov, and Y. Nishida, The NewReno Modification to TCP's Fast Recovery Algorithm, document RFC 6582, 2004.
[38]
TC (Linux). Accessed: Feb. 20, 2022. [Online]. Available: https://en.wikipedia.org/wiki/Tc_
[39]
D. Leith and R. Shorten. (2005). H-TCP: TCP Congestion Control for High Bandwidth-delay Product Paths. [Online]. Available: https://datatracker.ietf.org/doc/pdf/draft-leith-tcp-htcp-00.html
[40]
I. Rhee, L. Xu, S. Ha, A. Zimmermann, L. Eggert, and R. Scheffenegger, CUBIC for Fast Long-distance Networks, document RFC 8312, 2018.
[41]
L. Xu, K. Harfoush, and I. Rhee, “Binary increase congestion control (BIC) for fast long-distance networks,” in Proc. IEEE INFOCOM, Hong Kong, Mar. 2004, pp. 2514–2524.
[42]
M. Alizadeh et al., “Data center TCP (DCTCP),” in Proc. ACM SIGCOMM Conf., New Delhi, India, Aug. 2010, pp. 63–74.
Recommendations
Weighted start-time fair queuing: a new start-time fair queuing for elastic traffic
SpringSim '09: Proceedings of the 2009 Spring Simulation MulticonferenceIn this paper, we present a new Start-time Fair Queuing (SFQ) algorithm called Weighted Start-time Fair Queuing (WSFQ) which is more efficient and achieves better fairness than SFQ in the presence of small and huge elastic traffic flows. WSFQ scheduler, ...
Core-stateless fair rate estimation fair queuing: Research Articles
Core-stateless mechanisms, such as core-stateless fair queuing (CSFQ), reduce the complexity of fair queuing, which usually need to maintain states, manage buffers, and perform flow scheduling on a per-flow basis. However, they require executing label ...
Inversion impact of approximate PIFO to Start-Time Fair Queueing
AbstractProgrammable packet schedulers have emerged in the context of Software-Defined Networking (SDN) and diverse service requirements in today’s networks. With a programmable packet scheduler, network operators can apply different scheduling ...
Comments
Information & Contributors
Information
Published In
1063-6692 © 2024 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See https://www.ieee.org/publications/rights/index.html for more information.
Publisher
IEEE Press
Publication History
Published: 14 May 2024
Published in TON Volume 32, Issue 5
Qualifiers
- Research-article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 10Total Downloads
- Downloads (Last 12 months)10
- Downloads (Last 6 weeks)3
Reflects downloads up to 13 Jan 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in