This Article 
 Bibliographic References 
 Add to: 
Performance-Guided Load (Un)balancing under Autocorrelated Flows
May 2008 (vol. 19 no. 5)
pp. 652-665
Size-based policies have been shown in the literature to effectively balance load and improve performance in cluster environments. Size-based policies assign jobs to servers based on the job size and their performance improvements are an outcome of separating "short" from "long" jobs, by avoiding having short jobs waiting behind long jobs for service. In this paper, we present evidence that performance improvements due to this separation quickly vanish if the arrival process to the cluster is autocorrelated. Based on our observations, we devise a new size-based policy called {\DEqAL} that still strives to separate jobs to servers according to job size but this separation is now biased by an effort to reduce performance loss due to autocorrelation in the arrival flows of jobs that are directed to each server. As a result of this bias, all servers may not be equally utilized (i.e., load in the system may be "unbalanced"), but performance benefits become significant. {\DEqAL} can be used on-line as it does not assume any a priori knowledge of the incoming workload. Extensive simulations show the effectiveness of {\DEqAL} under autocorrelated and uncorrelated arrival streams and illustrate that the policy successfully self-adjusts the degree of load unbalancing based on monitored performance measures.

[1] D. Abendroth and U. Killat, “Intelligent Shaping: Well Shaped Throughout the Entire Network?” Proc. IEEE INFOCOM '02, vol. 2, pp. 912-919, June 2002.
[2] A.M. Adas and A. Mukherjee, “On Resource Management and QoS Guarantees for Long Range Dependent Traffic,” Proc. IEEE INFOCOM '95, vol. 2, pp. 779-787, Apr. 1995.
[3] M. Andreolini, M. Colajanni, and R. Morselli, “Performance Study of Dispatching Algorithms in Multi-Tier Web Architectures,” ACM SIGMETRICS Performance Evaluation Rev., vol. 30, no. 2, pp.10-20, Sept. 2002.
[4] M. Arlitt and T. Jin, “Workload Characterization of the 1998 World Cup Web Site,” Technical Report HPL-1999-35R1, Hewlett-Packard Laboratories, Sept. 1999.
[5] D. Bushmitch, S.S. Panwar, and A. Pal, “Thinning, Striping and Shuffling: Traffic Shaping and Transport Techniques for Variable Bit Rate Video,” Proc. IEEE GLOBECOM '02, vol. 2, pp. 1485-1491, Nov. 2002.
[6] L. Cherkasova, W. Tang, and S. Singhal, “An SLA-Oriented Capacity Planning Tool for Streaming Media Services,” Proc. Int'l Conf. Dependable Systems and Networks (DSN '04), pp. 743-752, June 2004.
[7] K.J. Christensen and V. Ballingam, “Reduction of Self-Similarity by Application-Level Traffic Shaping,” Proc. 22nd Ann. IEEE Conf. on Local Computer Networks (LCN '97) pp. 511-518, Nov. 1997.
[8] A. Erramilli, O. Narayan, and W. Willinger, “Experimental Queueing Analysis with Long-Range Dependent Packet Traffic,” IEEE/ACM Trans. Networking, vol. 4, no. 2, 209-223, Apr. 1996.
[9] H. Feng, M. Visra, and D. Rubenstein, “Optimal State-Free, Size-Aware Dispatching for Heterogeneous M/G/-Type Systems,” Performance Evaluation J., vol. 62, nos. 1-4, 475-492, Nov. 2005.
[10] M. Harchol-Balter, M. Crovella, and C.D. Murta, “On Choosing a Task Assignment Policy for a Distributed Server System,” J. Parallel and Distributed Computing, vol. 59, no. 2, 204-228, Nov. 1999.
[11] M. Harchol-Balter and A. Downey, “Exploiting Process Lifetime Distributions for Dynamic Load Balancing,” ACM Trans. Computer Systems, vol. 15, no. 3, pp. 253-285, Aug. 1997.
[12] D. Heyman and D. Lucantoni, “Modeling Multiple IP Traffic Streams with Rate Limits,” IEEE/ACM Trans. Networking, vol. 11, no. 6, pp. 948-958, Dec. 2003.
[13] L. Kleinrock, Queueing Systems, Volume I: Theory. Wiley, 1975
[14] N. Mi, Q. Zhang, A. Riska, E. Smirni, and E. Riedel, “Performance Impacts of Autocorrelated Flows in Multi-Tiered Systems,” Performance Evaluation, vol. 64, nos. 9-12, pp. 1082-1101, Oct. 2007.
[15] V.S. Pai, M. Aron, G. Banga, M. Svendsen, P. Druschel, W. Zwaenepoel, and E. Nahum, “Locality-Aware Request Distribution in Cluster-Based Network Servers,” Proc. Eighth Int'l Conf. Architectural Support for Programming Languages and Operating Systems (ASPLOS-VIII), pp. 205-216, Oct. 1998.
[16] A. Riska and E. Riedel, “Disk Drive Level Workload Characterization,” Proc. USENIX Ann. Technical Conf., pp. 97-102, June 2006.
[17] A. Riska and E. Riedel, “Long-Range Dependence at the Disk Drive Level,” Proc. Third Int'l Conf. Quantitative Evaluation of Systems (QEST '06), pp. 41-50, Sept. 2006.
[18] Y.M. Teo and R. Ayani, “Comparison of Load Balancing Strategies on Cluster-Based Web Servers,” Trans. Soc. for Modeling and Simulation, vol. 77, nos. 5-6, pp. 185-195, Nov. 2001.
[19] V. Ungureanu, B. Melamed, P.G. Bradford, and M. Katehakis, “Class-Dependent Assignment in Cluster-Based Servers,” Proc. ACM Symp. Applied Computing (SAC '04), pp. 1420-1425, Mar. 2004.
[20] U. Vallamsetty, K. Kant, and P. Mohapatra, “Characterization of E-Commerce Traffic,” Proc. Fourth IEEE Int'l Workshop Advanced Issues of E-Commerce and Web-Based Information Systems (WECWIS '02), pp. 137-144, 2002.
[21] F. Xue and S.J. B. Yoo, “Self-Similar Traffic Shaping at the Edge Router in Optical Packet-Switched Networks,” Proc. IEEE Int'l Conf. Comm. (ICC '02), vol. 4, pp. 2449-2453, Apr. 2002.
[22] Q. Zhang, L. Cherkasova, and E. Smirni, “FlexSplit: A Workload-Aware, Adaptive Load Balancing Strategy for Media Clusters,” Proc. Multimedia Computing and Networking (MMCN '06), Jan. 2006.
[23] Q. Zhang, A. Heindl, and E. Smirni, “Characterizing the BMAP/MAP/1 Departure Process via the ETAQA Truncation,” Stochastic Models, vol. 21, nos. 2-3, pp. 821-846, 2005.
[24] Q. Zhang, N. Mi, A. Riska, and E. Smirni, “Load Unbalancing to Improve Performance under Autocorrelated Traffic,” Proc. 26th IEEE Int'l Conf. Distributed Computing Systems (ICDCS '06), June 2006.
[25] Q. Zhang, A. Riska, W. Sun, E. Smirni, and G. Ciardo, “Workload-Aware Load Balancing for Clustered Web Servers,” IEEE Trans. Parallel and Distributed Systems, vol. 16, no. 3, pp. 219-233, Mar. 2005.

Index Terms:
Load balancing and task assignment, Algorithm/protocol design and analysis, Adaptable architectures, Distributed architectures, Correlation and regression analysis, Distribution functions
Qi Zhang, Ningfang Mi, Alma Riska, Evgenia Smirni, "Performance-Guided Load (Un)balancing under Autocorrelated Flows," IEEE Transactions on Parallel and Distributed Systems, vol. 19, no. 5, pp. 652-665, May 2008, doi:10.1109/TPDS.2007.70775
Usage of this product signifies your acceptance of the Terms of Use.