Abstract
This paper will focus on the issue of load balancing on a hypercube network of N processors. We will investigate a typical nearest neighbor balancing strategy - in which workloads among neighboring processors are averaged at discrete time steps. The computation model allows tasks, described by independent random variables, to be generated and terminated at all times.
We assume that the random variables at all nodes have equal expected value and their variances are bounded by a constant d2, and we let the difference DIFF between the actual load on each node and the average load on the system describe the deviation of the load on a node from the global average value. The following analytical results are obtained:
The expected value of DIFF, denoted by E(DIFF), is 0.
The variance of DIFF, denoted by Var(DIFF), is independent of time t, and Var(DIFF)≤ 1.386d2 + 0.231 logN.
- 1 M.A.Igbal, J.H.Salts, S.H.Bokhari, A Comparative analysis of Static and Dynamic Load Balancing Strategies, Proceedings of the 1986 International Conference on Parallel Processing, Aug. 1986. pp.1040-1045.Google Scholar
- 2 Winifred I. Williams, Load Balancing and Hypercubes: A preliminary look, Proceedings of The Second Conference on Hypercube Multiprocessors, Sept. 1986. pp. 108- 113.Google Scholar
- 3 M.Livny, The study of load balancing algorithms for decentralized processing systems, ph.D. Dissertation. Veizman Institute of Science, Aug. 1983.Google Scholar
- 4 D. Eager, E.D.Lazowska and J.Zahorjan, A comparison of receiver initiated and sender initiated dynamic load sharing, Proc. of the ACM-SIGMETRICS Conf. on Measurement and Modeling of Computer Systems, Aug. 1985. Google Scholar
Digital Library
- 5 Kai Hwang and Raymond Chowkwanyun, Dynamic Load Balancing Methods for Distributed Supercomputing, CRI-TR87-02, Computer Research Institute, University of Southern California, Feb. 1987.Google Scholar
- 6 F.C.Lin and R.M.Keller, Gradient Model: A Demand- Driven Load Balancing Scheme, IEEE Conference on Distributed Systems, pp.329-336. 1986.Google Scholar
- 7 L.M.ni, C.Xu and T.B.Gendreau, A Distributed Drafting Algorithm for Load Balancing, IEEE Transections on Computers, vol. SE-11, No.10, pp. 1153-1161. Oct. 1985. Google Scholar
Digital Library
Index Terms
From local to global: an analysis of nearest neighbor balancing on hypercube
Recommendations
From local to global: an analysis of nearest neighbor balancing on hypercube
SIGMETRICS '88: Proceedings of the 1988 ACM SIGMETRICS conference on Measurement and modeling of computer systemsThis paper will focus on the issue of load balancing on a hypercube network of N processors. We will investigate a typical nearest neighbor balancing strategy - in which workloads among neighboring processors are averaged at discrete time steps. The ...
Load balancing on a hypercube
IPPS '91: Proceedings of the Fifth International Parallel Processing SymposiumAssuming there is a hypercube with p=2/sup d/ processors and that each hypercube processor, PE(i), has L/sub i/ units of load. The load is always in integral units and may be distributed to other processors in integral units. In the load balancing ...
Randomized Approximate Nearest Neighbor Search with Limited Adaptivity
Special Issue on SPAA 2016We study the complexity of parallel data structures for approximate nearest neighbor search in d-dimensional Hamming space {0,1}d. A classic model for static data structures is the cell-probe model [27]. We consider a cell-probe model with limited ...






Comments