The DLPS: A New Framework for Benchmarking Blockchains

Abstract

Distributed Ledger Technologies (DLT) promise to revolutionize business ecosystems by permitting secure transactions without intermediaries. A widely recognized challenge that inhibits the uptake of DLT is scalability and performance. Hence, quantifying key metrics such as throughput and latency is crucial for designing DLT-based infrastructures, applications, and ecosystems. However, current benchmarking frameworks for blockchains do not cover the whole benchmarking process; impeding transparent comparisons of different DLT networks. In this paper, we present the Distributed Ledger Performance Scan (DLPS), an open-source framework for end-to-end performance characterizations of blockchains, addressing the need to transparently and automatically evaluate the performance of highly customizable configurations. We describe our new framework and argue that it significantly improves existing DLT benchmarking solutions. To demonstrate the capabilities of the DLPS, we also summarize the main results obtained from a series of experiments that we have conducted with it, giving a first comprehensive comparison of essential scalability properties of several commonly used enterprise blockchains.