ABSTRACT
While numerous studies have examined the macro-level behavior of traffic in data center networks---overall flow sizes, destination variability, and TCP burstiness---little information is available on the behavior of data center traffic at packet-level timescales. Whereas one might assume that flows from different applications fairly share available link bandwidth, and that packets within a single flow are uniformly paced, the reality is more complex. To meet increasingly high link rates of 10 Gbps and beyond, batching is typically introduced across the network stack---at the application, middleware, OS, transport, and NIC layers. This batching results in short-term packet bursts, which have implications for the design and performance requirements of packet processing devices along the path, including middleboxes, SDN-enabled switches, and virtual machine hypervisors.
In this paper, we study the burst behavior of traffic emanating from a 10-Gbps end host across a variety of data center applications. We find that at 10--100 microsecond timescales, the traffic exhibits large bursts (i.e., 10s of packets in length). We further find that this level of burstiness is largely outside of application control, and independent of the high-level behavior of applications.
- 100Gb/s Ethernet Task Force. http://www.ieee802.org/3/ba/.Google Scholar
- A. Aggarwal, S. Savage, and T. E. Anderson. Understanding the Performance of TCP Pacing. In Proc. INFOCOM, 2000.Google ScholarCross Ref
- M. Alizadeh, A. Kabbani, T. Edsall, B. Prabhakar, A. Vahdat, and M. Yasuda. Less is More: Trading a little Bandwidth for Ultra-Low Latency in the Data Center. In Proc. NSDI, 2012. Google ScholarDigital Library
- M. Allman and E. Blanton. Notes on Burst Mitigation for Transport Protocols. SIGCOMM Comput. Commun. Rev., 35(2):53--60, Apr. 2005. Google ScholarDigital Library
- Apache Software Foundation. HDFS Architecture Guide. http://hadoop.apache.org/docs/hdfs/current/hdfs_design.html.Google Scholar
- T. Benson, A. Akella, and D. A. Maltz. Network Traffic Characteristics of Data Centers in the Wild. In Proc. IMC, 2010. Google ScholarDigital Library
- E. Blanton and M. Allman. On the Impact of Bursting on TCP Performance. In Proc. PAM, 2005. Google ScholarDigital Library
- J. Dean and S. Ghemawat. MapReduce: Simplified Data Processing on Large Clusters. In Proc. OSDI, 2004. Google ScholarDigital Library
- N. Farrington, G. Porter, S. Radhakrishnan, H. H. Bazzaz, V. Subramanya, Y. Fainman, G. Papen, and A. Vahdat. Helios: A Hybrid Electrical/Optical Switch Architecture for Modular Data Centers. In Proc. ACM SIGCOMM, Aug. 2010. Google ScholarDigital Library
- S. Ghemawat, H. Gobioff, and S.-T. Leung. The Google File System. In Proc. SOSP, 2003. Google ScholarDigital Library
- M. Ghobadi, Y. Cheng, A. Jain, and M. Mathis. Trickle: Rate Limiting YouTube Video Streaming. In Proc. ATC, 2012. Google ScholarDigital Library
- R. Jain and S. Routhier. Packet Trains--Measurements and a New Model for Computer Network Traffic. IEEE J.Sel. A. Commun., 4(6):986--995, Sept. 1986. Google ScholarDigital Library
- H. Jiang and C. Dovrolis. Source-level IP Packet Bursts: Causes and Effects. In Proc. IMC, 2003. Google ScholarDigital Library
- H. Jiang and C. Dovrolis. Why is the Internet Traffic Bursty in Short Time Scales? In Proc. SIGMETRICS, 2005. Google ScholarDigital Library
- M. Moshref, M. Yu, A. Sharma, and R. Govindan. Scalable rule management for data centers. In Proceedings of the 10th USENIX conference on Networked Systems Design and Implementation, nsdi'13, pages 157--170, Berkeley, CA, USA, 2013. USENIX Association. Google ScholarDigital Library
- R. N. Mysore, G. Porter, Subramanya, and A. Vahdat. FasTrak: Enabling Express Lanes in Multi-Tenant Data Centers. In Proc. CoNEXT, 2013. Google ScholarDigital Library
- ONF. Software-Defined Networking: The New Norm for Networks. https://www.opennetworking.org/.Google Scholar
- Open vSwitch: An Open Virtual Switch. http://openvswitch.org/.Google Scholar
- G. Porter, R. Strong, N. Farrington, A. Forencich, P.-C. Sun, T. Rosing, Y. Fainman, G. Papen, and A. Vahdat. Integrating Microsecond Circuit Switching into the Data Center. In Proc. SIGCOMM, 2013. Google ScholarDigital Library
- R. Prasad, M. Jain, and C. Dovrolis. Effects of Interrupt Coalescence on Network Measurements. In Proc. PAM, 2004.Google ScholarCross Ref
- A. Rao, A. Legout, Y.-s. Lim, D. Towsley, C. Barakat, and W. Dabbous. Network Characteristics of Video Streaming Traffic. In Proc. CoNEXT, 2011. Google ScholarDigital Library
- A. Rasmussen, G. Porter, M. Conley, H. V. Madhyastha, R. N. Mysore, A. Pucher, and A. Vahdat. TritonSort: A Balanced Large-Scale Sorting System. In Proc. NSDI, 2011. Google ScholarDigital Library
- V. Sekar, N. Egi, S. Ratnasamy, M. K. Reiter, , and G. Shi. Design and Implementation of a Consolidated Middlebox Architecture. In Proc. NSDI, 2012. Google ScholarDigital Library
- S. Sinha, S. Kandula, and D. Katabi. Harnessing TCPs Burstiness using Flowlet Switching. In Proc. HotNets, 2004.Google Scholar
- B. Vattikonda, G. Porter, A. Vahdat, and A. C. Snoeren. Practical TDMA for Datacenter Ethernet. In Proc. ACM EuroSys, 2012. Google ScholarDigital Library
- D. Wischik. Buffer Sizing Theory for Bursty TCP Flows. In Communications, 2006 International Zurich Seminar on, pages 98--101, 2006. Google ScholarDigital Library
- T. Yoshino, Y. Sugawara, K. Inagami, J. Tamatsukuri, M. Inaba, and K. Hiraki. Performance Optimization of TCP/IP over 10 Gigabit Ethernet by Precise Instrumentation. In Proc. SC, 2008. Google ScholarDigital Library
- L. Zhang, S. Shenker, and D. D. Clark. Observations on the Dynamics of a Congestion Control Algorithm: The Effects of Two-Way Traffic. In Proc. SIGCOMM, 1991. Google ScholarDigital Library
Index Terms
- Bullet trains: a study of NIC burst behavior at microsecond timescales
Recommendations
Network traffic characteristics of data centers in the wild
IMC '10: Proceedings of the 10th ACM SIGCOMM conference on Internet measurementAlthough there is tremendous interest in designing improved networks for data centers, very little is known about the network-level traffic characteristics of data centers today. In this paper, we conduct an empirical study of the network traffic in 10 ...
Adaptive FEC for 802.11 burst losses reduction
MobiMedia '06: Proceedings of the 2nd international conference on Mobile multimedia communicationsOne of the challenges that video transmission over wireless networks (such as IEEE 802.11) must face is the packet loss that can heavily decrease the received video quality. What makes the challenge more difficult is the bursty nature of wireless ...
A New Quorum-Based Power Saving Protocol to Maintain QoS for MANETs with Burst Traffics
WAINA '10: Proceedings of the 2010 IEEE 24th International Conference on Advanced Information Networking and Applications WorkshopsIn this paper we address the Quality of Service (QoS) issues for applications such as voice and video that generate bursty traffic in mobile ad hoc networks while asynchronous power saving protocols are applied to wireless hosts. To provide QoS, we ...
Comments