An Implementation of Depth-First and Breadth-First Search Algorithms for Tip Selection in IOTA Distributed Ledger.

IOTA is one of the most promising distributed ledgers due to its performance, scalability, and fee-less transaction capability. The interaction with the IOTA ledger requires client software to perform complex logic. As this software is often running on low-power Internet of Things (IoT) devices, it is important for it to be highly optimized. Our work highlights the benefits of DAG-based distributed ledgers. We start with an overview of IOTA tip selection algorithms, including the Markov Chain Monte Carlo (MCMC) algorithm and time-based tip selection algorithm. The original IOTA Reference Implementation (IRI) of CumulativeWeight Calculation (CWC) is based on Breadth-First Search (BFS) algorithm. The main contribution of our paper is to show how the Depth-First Search (DFS) algorithm is a better alternative solution to the CWC problem. We describe a sample software implementation, a method for collecting a snapshot from the Tangle, and finally we discuss the experimental outcomes.