Fire ant: an algorithm providing high-performance distributed fault-tolerant communication for internet-scale network survivability

摘要: In today's world, data information, exchange, and processing are sustenance to corporations, organizations, government agencies. Communications networks relied upon meet the computing needs of such organizations. These groups have demands that require large amounts power, use supercomputing systems consisting massively parallel processors (MPP) for a wide range intensive operations. With advances in high-bandwidth network technology, performance found MPP can be replicated with workstations (NOW). When large-scale disaster situations occur, intelligent routing algorithms needed maintain necessary resources required by NOW. must robust enough accommodate changing conditions while still providing communications acceptable load balancing flow. Mobile multiple autonomous units or multi-agents used make fault-tolerant decisions way similar those swarm insects. this research, we develop mobile multi-agent algorithm avoidance communications. This algorithm, called Fault-Tolerant Intelligent Routing Environment AntNet FIRE Ant, is designed using activity structures. novel approach determining design parameters requisite behavior algorithm. The activities NOWs environment analyzed order find its relevant components. Activity structures conceptual methodological tools capturing adaptive natural artificial systems. research they allow interaction diverse high-performance, communication network. Ant compared simplified version most well known based AntNet, BGP UPDATE procedure Internet routing. first agent address failures situations. cost optimize average flow an entire as method during Network added study determine effectiveness three worst-case scenarios study, results show provides up 35.7% throughput improvement over when experiencing situation.