Skip to main content
Log in

Optimizing I/O server placement for parallel I/O on switch-based irregular networks

The Journal of Supercomputing Aims and scope Submit manuscript

Abstract

In this paper, we study I/O server placement for optimizing parallel I/O performance on switch-based clusters, which typically adopt irregular network topologies to allow construction of scalable systems with incremental expansion capability. Finding optimal solution to this problem is computationally intractable. We quantified the number of messages travelling through each network link by a workload function, and developed three heuristic algorithms to find good solutions based on the values of the workload function. The maximum-workload-based heuristic chooses the locations for I/O nodes in order to minimize the maximum value of the workload function. The distance-based heuristic aims to minimize the average distance between the compute nodes and I/O nodes, which is equivalent to minimizing average workload on the network links. The load-balance-based heuristic balances the workload on the links based on a recursive traversal of the routing tree for the network.

Our simulation results demonstrate performance advantage of our algorithms over a number of algorithms commonly used in existing parallel systems. In particular, the load-balance-based algorithm is superior to the other algorithms in most cases, with improvement ratio of 10 to 95% in terms of parallel I/O throughput.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

References

  1. Bae M, Bose B (1997) Resource placement in torus-based networks. IEEE Trans. Computers 46(10):1083–1092

    Article  MathSciNet  Google Scholar 

  2. Boden NJ, Cohen D, Felderman RF, Kulawik AE, Seitz CL, Seizovic J, Su W (1995) Myrinet—A gigabit per second local area network. IEEE Micro, pp 29–36

  3. Brezany P, Mueck TA, Schikuta E (1996) A software architecture for massively parallel input-output. In: Proc. 3rd International Workshop PARA’96, LNCS Springer Verlag

  4. Carns PH, Ligon III WB, Ross RB, Thakur R (2000) Pvfs: A parallel file system for linux clusters. In: Proc. 4th Annual Linux Showcase and Conference, pp 317–327

  5. Cho Y, Winslett M, Subramaniam M, Chen Y, Kuo SW, Seamons KE (1997) Exploiting local data in parallel array i/o on a practical network of workstations. In: Proc. Fifth Workshop on I/O in Parallel and Distributed Systems (IOPADS)

  6. Dan A, Sitaram D (1995) An on-line video placement policy based on bandwidth to space ratio. In: ACM SIGMOD International Conf. Management of Data, pp 376–385

  7. Dukes J, Jones J (2003) Dynamic replication of content in the hammerhead multimedia server. Technical report, Department of Computer Science, Trinity College Dublin, Ireland

  8. Schroeder MD et al. (1990) Autonet: A high-speed, self-configuring local area network using point-to-point links. Technical Report SRC research report 59, DEC

  9. Harry M, Rosario J, Choudhary A (1995) Vipfs: A virtual parallel file system for high performance parallel anddistributed computing. In: Proc. 9th International Parallel Processing Symposium

  10. Huber J, Elford CL, Reed DA, Chien AA, Blumenthal DS (1995) Ppfs: A high performance portable parallel file system. In: Proc. 9th ACM International Conference on Supercomputing, pp 485–394

  11. Hwang Y, Mckinley PK (1996) Efficient collective operations with atm network interface support. In: Proceedings of ICPP

  12. Moyer S, Sunderam V (1994) Pious: A scalable parallel i/o system for distributed computing environments. Technical Report Computer Science Report CSTR-940302, Department of Math and Computer Science, Emory University

  13. Nieuwejaar N (1996) Galley: A New Parallel File System for Scientific Workload. PhD thesis, Dept. of Computer Science, Dartmouth College

  14. Ramananthan P, Chalasani S (1995) Resource placement with multiple adjacency constraints in k-ary n-cubes. IEEE Trans. Parallel and Distributed Systems 6(5):511–519

    Article  Google Scholar 

  15. Seamons KE, Chen Y, Jones P, Jozwiak J, Winslett M (2001) Reading in Disk Array and Parallel I/O, chapter server-directed collective I/O in Panda. IEEE Computer Society Press

  16. Serpanos DN, Georgiadis L, Bouloutas T (1998) MMPacking: A Load and Storage Balancing Algorithm for Distributed Multimedia Servers. IEEE Trans. Circuits and Systems for Video Technology 8(1):13–17

    Article  Google Scholar 

  17. Subramany SR, Narahari B, Simha R (1998) Placement of storage nodes in a network. In: International Conference on Parallel and Distributed Processing Techniques and Applications

  18. Thakur R, Choudhary A, Bordawekar R, More S, Kuditipudi S (1996) Passion: Optimized I/O for parallel applications. IEEE Computer 29(6):70–78

    Google Scholar 

  19. Tseng NF, Feng GL (1996) Resource allocation in cube network systems based on the covering radius. IEEE Trans. Parallel and Distributed Systems 7(4):323–342

    Google Scholar 

  20. Venkatasubramanian N, Ramanathan S (1997) Load management in distributed video servers. In: Inter. Conf. Distributed Computing Systems

  21. Wang Y, Lin J, Du D, Hsieh J (1996) Efficient video allocation for video-on-demand services. In: IEEE Multimedia Conference

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Jan-Jan Wu.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Lin, YF., Wang, CM. & Wu, JJ. Optimizing I/O server placement for parallel I/O on switch-based irregular networks. J Supercomput 36, 201–217 (2006). https://doi.org/10.1007/s11227-006-8293-2

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11227-006-8293-2

Keywords

Navigation