Modelling and Analyzing Distributed Ledger Technology Protocols for Decentralized Software Applications
| Typ | Bachelorarbeit oder Masterarbeit | |
|---|---|---|
| Aushang | ||
| Betreuer | Wenden Sie sich bei Interesse oder Fragen bitte an: Robert Heinrich (E-Mail: heinrich@kit.edu, Telefon: +49-721-608-45963), Niclas Kannengießer |
The selection of insufficient distributed ledger technology (DLT) protocols for decentralized software applications (DApps) can have detrimental consequences for the operation of DApps; especially, because DLT protocols can hardly be changed after being set up. For example, all nodes in a DLT system would need to set up another DLT protocol on their nodes to operate a new DLT system. Data stored in the previous DLT system may not be migrated into the new one. A modelling and simulation application is needed to support practitioners in their selections and configurations of DLT protocols so that corresponding DLT systems meet individual DApp requirements. By using such a modelling and simulation tool, potential drawbacks of misconfigured DLT systems can be identified prior to setting up DLT systems, which can ultimately reduce resource consumption (e.g., in terms of cost) of setting up and operating DLT systems.
Palladio is a model-based analysis approach currently focused on the modelling and analysis of quality properties on the level of software architectures. Palladio serves as a platform that can be adapted and extended for various model-based analysis. In this thesis, Palladio should be extended to analyze DLT protocols for DApps for quality properties, especially performance.
Thesis Goals • Specification of a modeling language to represent DApps by extending and refining Palladio’s modeling language • Specification of a model-based analysis to investigate DApps by extending and refining Palladio’s analyses • A working DLT extension should be implemented in the Palladio modeling and simulation application. • The DLT extension for Palladio should be evaluated in terms of accuracy in predicting the performance of modeled DLT systems.