Scaling Solutions for the Blockchain

November 2015 - November 2018
About 69,500 €
Funding organization: 

Politecnico di Torino

Person(s) in charge: 
Executive summary: 

The main goal of this project is to study the performance of protocols that aim to scale the blockchain, which currently supports just few transactions per second and prevents Bitcoin and other cryptocurrencies from becoming mainstream payment systems.


Bitcoin is a decentralized crypto-currency. “Decentralized” means that it is operated by a peer-to-peer network, without any central point of control. “Crypto” means that the currency is created and controlled by cryptography: there is no bank issuing the currency.
The blockchain is the ledger which stores all economic transactions occurred in Bitcoin. A transaction is a transfer of cryptocurrency from a party to another. Bitcoin is decentralized in the sense that each peer stores a copy of the blockchain, each peer can validate new transactions and each peer can create new currency. This is possible thanks to the blockchain protocol, which allows all peers to reach distributed consensus, i.e. all peers agree on which transactions occurred: this lets the Bitcoin payment system work.

In particular, the key novelty of the blockchain is to have a distributed consensus in a public environment, without identifying and trusting peer, and so without the need of a trusted central entity. The main blockchain limit is the scalability. To keep Bitcoin decentralized, blockchain’s growth is limited by design, and only 7 transactions per second can be registered in the Bitcoin blockchain. A category of scaling proposals is constituted by off-chain solutions.
The basic idea of off-chain solutions is to allow off-chain transactions, i.e. transactions that don't need to be written to the blockchain and so are not subject to the blockchain limit of 7 transactions per second. This is obtained by creating a layer-two payment network on top of the blockchain, through which it is possible to route off-chain payments without making transactions on the blockchain.


Our research goal is to measure the performance of the Lightning Network, the protocol which implements an off-chain payment network on top of the Bitcoin blockchain.
In particular, we want to study how the performance is affected by factors such as: the network topology; the distribution of bitcoins among the peers of the network; the presence of misbehaving peers that do not respect the protocol; the number and amount of payments performed in the network. Due to some limitations in the Lightning Network protocol, our intuition is that the Lightning Network may be unfeasible in a totally distributed network topology. We expect that many payments fail due to the absence of viable routes. Moreover, we expect that misbehaving peers can cause serious economic damages to the network.


1) A short paper titled “Blockchain for the Internet of Things: a Systematic Literature Review”, with authors Marco Conoscenti, Antonio Vetrò and Juan Carlos De Martin, accepted for publication in the proceedings of The Third International Symposium on Internet of Things: Systems, Management and Security (IOTSMS 2016). It is a Systematic Literature Review, i.e. a review of the literature conducted with a rigorous and auditable method and with precise goals and research questions. In particular, the goal was to gather knowledge on the current uses of the blockchain and to document its current degree of anonymity, integrity and scalability.
2) A short paper titled “Peer to Peer for Privacy and Decentralization in the Internet of Things”, with authors Marco Conoscenti, Antonio Vetrò and Juan Carlos De Martin, accepted for publication in the proceedings of the 39th International Conference of Software Engineering (ICSE 2017). The paper is a two-page extended abstract describing an idea of leveraging P2P applications and the blockchain to foster a decentralized private-by-design Internet of Things.
3) A simulator of the Lightning Network protocol. This simulator takes various input parameters such as the network topology, the distribution of bitcoins in the network, the percentage of misbehaving peers. It outputs some performance measures, such as the number of failed payments, the average payment time and the average route length of payments.
Currently, we are writing a paper on the results provided by the first simulations we ran.