As a post-doc at ESiLV engineering school, Gabriel Lhéricy who is also a member of De Vinci Research Center, has been working on the “Moneytrack” partnership research project around cryptocurrencies. The work was done in collaboration with Cyril Grunspan (ESiLV, De Vinci Research Center) and Ricardo Pérez-Marco (CNRS, IMJ-PRG).
In his talk, Gabriel Lehéricy presents how “ant” routing works, and tries to estimate the efficiency of the concerned algorithm and the effectiveness of its implementation, whereby the ultimate goal would be to see how many transactions this “ant” algorithm can theoretically handle.
It’s about estimating the capacity of the ant routing algorithm for the Bitcoin Lightning Network.
What is Moneytrack?
Moneytrack is the big umbrella project that aims to respond to a recurring need in the payment sector: controlling the use of a payment platform made available to a consumer, either personally or in a professional capacity. Under the initiative of Truffle Capital and conducted in partnership with ESiLV, Inria, Keyrus, Ocaml Pro, and Wizypay; the project benefits from the labels of Finance Innovation and Systématic Région, and is supported by the Fonds Unique Interministériel.
Bitcoin and The Lightning Network
The lightning network (LN) is a payment network laid on top of the bitcoin network and consists of an aggregation of payment channels that allow its users to do off-chain micropayments in bitcoins. The concept was introduced in the hopes of solving bitcoin’s scalibility problem. In their proposal, Gabriel, Cyril, and Ricardo suggest a decentralized routing algorithm that can be implemented in Bitcoin Lightning Network: all nodes in the network contribute equally to path searching. The algorithm is inspired from ant path searching algorithms.
In fact, paying via the LN requires finding a route from the paying node to the payee, hence the need for a suitable routing algorithm. However, since the current routing algorithm in use in the LN poses threats to decentralization, Cyril Grunspan and Ricardo Pérez-Marco proposed in one of their papers, a more decentralized routing algorithm, called the “ant” routing algorithm. In this talk, Gabriel explains the protocol that needs to be followed in order to perform the “ant” routing task and also elaborates on the implementation of this protocol.
I will estimate the efficiency of this implementation. In particular, I will give an estimate of the number of transactions which the ant routing algorithm can handle.
After estimating that one node can manage to process more than 10,000 transactions per second, Gabriel deduced that the whole network can manage far more than that because not all the nodes are needed to be able to process all the incoming routing tasks.
Find out more about the Finance Group members and their publications here.