Dynamic Load Balancing for I/O-Intensive Tasks on Heterogeneous Clusters

  • Xiao Qin
  • Hong Jiang
  • Yifeng Zhu
  • David R. Swanson
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2913)


Since I/O-intensive tasks running on a heterogeneous cluster need a highly effective usage of global I/O resources, previous CPU- or memory-centric load balancing schemes suffer significant performance drop under I/O-intensive workload due to the imbalance of I/O load. To solve this problem, we develop two I/O-aware load-balancing schemes, which consider system heterogeneity and migrate more I/O-intensive tasks from a node with high I/O utilization to those with low I/O utilization. If the workload is memory-intensive in nature, the new method applies a memory-based load balancing policy to assign the tasks. Likewise, when the workload becomes CPU-intensive, our scheme leverages a CPU-based policy as an efficient means to balance the system load. In doing so, the proposed approach maintains the same level of performance as the existing schemes when I/O load is low or well balanced. Results from a trace-driven simulation study show that, when a workload is I/O-intensive, the proposed schemes improve the performance with respect to mean slowdown over the existing schemes by up to a factor of 8. In addition, the slowdowns of almost all the policies increase consistently with the system heterogeneity.


Load Balance Remote Node Heterogeneous Cluster Dynamic Load Balance Page Fault 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Harchol-Balter, M., Downey, A.: Exploiting process lifetime distributions for load balancing. ACM Transactions on Computer Systems 15, 253–285 (1997)CrossRefGoogle Scholar
  2. 2.
    Acharva, A., Setia, S.: Availability and utility of idle memory in workstation clusters. In: Proceedings of the ACM SIGMETRICS Conf. on Measuring and Modeling of Computer Systems (1999)Google Scholar
  3. 3.
    Xiao, L., Zhang, X., Qu, Y.: Effective load sharing on heterogeneous networks of workstations. In: Proc. of International Symposium on Parallel and Distributed Processing (2000)Google Scholar
  4. 4.
    Qin, X., Jiang, H., Zhu, Y., Swanson, D.: A dynamic load balancing scheme for I/O-intensive applications in distributed systems. In: Proceedings of the 32nd International Conference on Parallel Processing Workshops (2003)Google Scholar
  5. 5.
    Qin, X., Jiang, H., Zhu, Y., Swanson, D.: Boosting performance for I/O-intensive workload by preemptive job migrations in a cluster system. In: Proc. of the 15th Symp. on Computer Architecture and High Performance Computing, Brazil (2003)Google Scholar
  6. 6.
    Scheuermann, P., Weikum, G., Zabback, P.: Data partitioning and load balancing in parallel disk systems. The VLDB Journal, 48–66 (1998)Google Scholar
  7. 7.
    Cho, Y., Winslett, M.S., Kuo, J.L., Chen, Y.: Parallel I/O for scientific applications on heterogeneous clusters: A resource-utilization approach. In: Proceedings of Supercomputing (1999)Google Scholar
  8. 8.
    Zhu, Y., Jiang, H., Qin, X., Feng, D., Swanson, D.: Scheduling for improved write performance in a cost-effective, fault-tolerant parallel virtual file system (CEFTPVFS). In: The Fourth LCI International Conference on Linux Clusters (2003)Google Scholar
  9. 9.
    Zhu, Y., Jiang, H., Qin, X., Feng, D., Swanson, D.: Improved read performance in a cost-effective, fault-tolerant parallel virtual file system (ceft-pvfs). In: Proc. of the 3rd IEEE/ACM Intl. Symp. on Cluster Computing and the Grid (2003)Google Scholar
  10. 10.
    Ma, X., Winslett, M., Lee, J., Yu, S.: Faster collective output through active buffering. In: Proceedings of the International Symposium on Parallel and Distributed Processing (2002)Google Scholar
  11. 11.
    Qin, X., Jiang, H., Zhu, Y., Swanson, D.: Dynamic load balancing for I/O- and memory-intensive workload in clusters using a feedback control mechanism. In: Kosch, H., Böszörményi, L., Hellwagner, H. (eds.) Euro-Par 2003. LNCS, vol. 2790, pp. 224–229. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  12. 12.
    Forney, B., Arpaci-Dusseau, A.C., Arpaci-Dusseau, R.H.: Storage-aware caching: Revisiting caching for heterogeneous storage systems. In: Proceedings of the 1st Symposium on File and Storage Technology, Monterey, California, USA (2002)Google Scholar
  13. 13.
    Geoffray, P.: Opiom: Off-processor I/O with myrinet. Future Generation Computer Systems 18, 491–499 (2002)zbMATHCrossRefGoogle Scholar
  14. 14.
    Franklin, M., Govindan, V.: A general matrix iterative model for dynamic load balancing. Parallel Computing 33 (1996)Google Scholar
  15. 15.
    Eager, D., Lazowska, E., Zahorjan, J.: Adaptive load sharing in homogeneous distributed systems. IEEE Trans. on Software Eng. 12, 662–675 (1986)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Xiao Qin
    • 1
  • Hong Jiang
    • 1
  • Yifeng Zhu
    • 1
  • David R. Swanson
    • 1
  1. 1.Department of Computer Science and EngineeringUniversity of Nebraska – LincolnLincolnUSA

Personalised recommendations