Showing papers on "Load balancing (computing) published in 1979"
TL;DR: Queueing models for a simple heterogeneous multiple processor system are presented, analyzed, and compared and each model is distinguished by a job routing strategy which is designed to reduce the average job turnaround time by balancing the total load among the processors.
Abstract: Queueing models for a simple heterogeneous multiple processor system are presented, analyzed, and compared. Each model is distinguished by a job routing strategy which is designed to reduce the average job turnaround time by balancing the total load among the processors. In each case an arriving job is routed by a job dispatcher to one of m parallel processors. The job routing strategies are divided into two classes: deterministic and nondeterministic. The nondeterministic policies are described by state independent branching probabilities. For the deterministic policies, the next processor is chosen to minimize or maximize the expected value of a performance related criterion function.
287 citations
TL;DR: In this paper, the cell-formation method developed at the C.I.T. has shown that the grouping of production facilities can be carried out in a very convenient way if the workload of the potential cells is used as a weighting factor.
28 citations
01 Jan 1979
TL;DR: Although this problem isNP-complete, techniques based on the Gomory-Hu tree from network flow theory can be applied to instances of the problem to obtainduction incomplexity, some instances can be solved completely and others reduced sufficiently to be susceptible toumerative techniques.
Abstract: A recently published algorithm shows howtoassign modules toatwo-processor computer system with distributed execu- tion soastominimize thetotal ofexecution costs andinterprocessor communication costs. Inthis paper weconsider thesameproblem except that oneprocessor haslimited memorycapacity. Although this problem isNP-complete, techniques based ontheGomory-Hu treefromnetwork flow theory canbeapplied toinstances ofthe problem toobtain areduction incomplexity. Anewtechnique based onagraph called theinclusive cutgraph isshown tobeanevenmore powerful tooLThese twotechniques cansolve someinstances ofthe problem completely; still others arereduced sufficiently tobe susceptible toenumerative techniques. Intheworstcase, the techniques yield noreduction inproblem size. IndexTerms-Computer networks, cutsets, distributed com- puters, Ford-Fulkerson algorithm, Gomory-Hu cuttrees, load balancing, multiterminal maximal flows, NP-complete problems.