Reliability and Packet Transfer Efficiency of Sparkle Topologies

Distributed computing through topology control involves making changes to the underlying network (modeled as a graph) to decrease the cost of distributed algorithms when executed across the modified networks. Topology construction builds reduced topology and topology maintenance adopts the reduced topology when the current topology is no longer optimal. This study makes major contributions to topology control by constructing new topologies named sparkle topologies using a ring topology, structured web topology, sun topology, and star topology. The efficiency of the sparkle topologies is checked by calculating the reliability using Wiener Index and data transfer using Cisco packet simulation. Data transmission times, the number of hops required, and the number of potential failure points are all reduced in sparkle topologies compared to ring topologies.