出版社:Academy & Industry Research Collaboration Center (AIRCC)
摘要:Persona-sized HPC clusters are widely used in many small labs, because they are cost-effectiveand easy to build. Instead of adding costly new nodes to old clusters, we may try to make use ofsome servers’ idle times by including them working independently on the same LAN, especiallyduring the night. However such extension across a firewall raises not only some securityproblem with NFS but also a load balancing problem caused by heterogeneity. In this paper, wepropose a method to solve such problems using only old techniques applicable to old systems asis, without requiring any upgrade for hardware or software. Some experimental results dealingwith heterogeneity and load balancing are presented using a two-queue overflow queuingnetwork problem.