Abstract
AbstractScaling high performance computer systems needs increasing the fault tolerance at the design stage of a topology. There are several approaches of designing simple fast routing with fault tolerance. One of effective approach is to ensure fault tolerance at the topology level. This article discusses methods for synthesizing fault-tolerant topologies. These topologies are hierarchical, scalable with a simple routing system. This paper discusses the synthesis of a multilevel topological organization consisting of clusters of excess De Bruijn rank connected by Dragonfly-like connections with hierarchical scaling. The main characteristics of this topology are high fault tolerance at the cluster level, as well as ultra-high scaling speed with a linear increasing of topology degree. The issue of routing in this kind of system is considered, the comparison with other topological organizations is made.KeywordsDragonflyDe BruijnMultilevel topological organization
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have