University of Twente Student Theses

Login

Gossip Layer Optimization for the IOTA Network

Mendler, Gereon (2023) Gossip Layer Optimization for the IOTA Network.

[img] PDF
5MB
Abstract:With the goal of optimizing the gossip layer of the IOTA 2.0 network, we show that Push-Pull gossip offers a worthwhile tradeoff to the currently used flooding by substantially decreasing the network load. This comes at the cost of a predictable speed penalty, but the strong resiliency and reliability are retained. Further, we identify a strong relationship with the network topology, which defines the limits of gossip protocol potential but also embodies the mechanism by which its advantages can be fully exhausted and its weaknesses remedied. This optimization is motivated by the arbitrarily defined throughput limitation of the IOTA system, hence lower network overhead could facilitate correspondingly more transactions per second. For this purpose, we explored various network topologies within the bounds of the specifications and investigated the behavior and resiliency of the current system through event-based simulation of the P2P network with Peersim. Push-Pull gossip is identified as the most suitable protocol for this project, and good configurations emerge through a parameter sensitivity analysis enabled by the simulation. This configuration is then evaluated for resiliency and in conjunction with alternative topologies to identify recommendations with regard to the peering mechanism. By these means, we can propose a gossip protocol to replace flooding.
Item Type:Essay (Master)
Clients:
IOTA Foundation, Berlin, Germany
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science MSc (60300)
Link to this item:https://purl.utwente.nl/essays/97820
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page