COOLedger: A COnfiguration toOL for Distributed Ledgers

Among many applications of Distributed Ledger technology (DLT) in fields such as finance or internet of things, each has its own specific requirements for configuring DLT features, such as high throughput or anonymity. Due to trade-offs between DLT characteristics (e.g., availability vs. consistency), there will not be a one-size-fits-all DLT design, which leads to a steadily growing number of differing DLT designs. A DLT design describes a particular implementation of a distributed ledger, for instance, Bitcoin or Ethereum. To meet the specific requirements of the different applications, advancements in the area of DLT are currently highly application-driven, leading to an ever-increasing number of different configured DLT designs. Since most DLT designs are open source, one can easily change DLT characteristics such as block size or block creation interval. Nevertheless, how the re-configuration of individual DLT characteristics effect DLT clients is currently unknown because there is only scattered analysis of potential side effects caused by DLT design configurations. So far, extant research hardly goes beyond the analysis of most popular blockchains (e.g., Bitcoin, Ethereum, HyperLedger), which would be essential for gaining a deep understanding on implications and side effects resulting from applied configurations of DLT characteristics. Especially security assessment becomes difficult when it comes to emerging technologies such as DLT. Due to the combination of several disciplines of computer science in DLT (e.g., distributed systems and encryption), DLT designs come with not obvious interdependencies between DLT characteristics concluding into trade-offs and unexpected side-effects. For example, high performance of a DLT design mostly comes at the cost of its level of security. Current research justifies the trade-off between performance and security by the fact that most attacks result from an increased likelihood of forks, which is a result of a misconfiguration of block size and block creation interval. Besides trade-offs between DLT characteristics already established theories such as CAP Theorem can be applied to DLT. Trade-offs between DLT characteristics impair suitability of DLT designs, because one must always make a compromise. Due to the current lack of migratibility of data stored on a distributed ledger from one DLT design to another DLT design, the decision of which DLT design is best fitting to the requirements of a given use case is crucial.

To support the identification of an appropriate configuration of distributed ledgers, this research project aims at the development of a model that explains effects of particular DLT characteristics on others. The model will be instantiated in a browser application and will be publicly usable.