The increasing digitalization of food supply chains, driven by the advent of Industry 4.0, has made product traceability an increasingly central issue. In this context, Filiera360 is a web application that leverages blockchain technology to ensure data security and immutability throughout the entire lifecycle of a food product. The platform was developed using an architecture composed of a Node.js middleware for interaction with the Hyperledger Fabric blockchain, a Flask backend for user management and RESTful APIs, and a React frontend for service access by the various supply chain actors. The goal of this work was to assess the robustness and scalability of the platform through a targeted stress test conducted on the product upload endpoint. For the experimentation, Locust—a distributed load testing tool containerized in Docker—was used to measure key metrics such as throughput, latency, and error rate. The results revealed bottlenecks mainly related to the persistence phase of blockchain transactions, which introduces increased latency as the load grows. Initial peaks were also observed due to chaincode instantiation and the establishment of peer-orderer connections. A critical analysis of the data made it possible to outline potential solutions and future developments, such as chaincode optimization, the introduction of caching strategies, and improvements in the Fabric network configuration. In conclusion, the study demonstrates that blockchain represents a valid technology for supply chain traceability, although specific optimizations are required to ensure adequate performance in real-world, high-load scenarios.
La crescente digitalizzazione delle filiere alimentari, favorita dall’avvento dell’Industria 4.0, ha reso sempre più centrale il tema della tracciabilità dei prodotti. In questo contesto si colloca Filiera360, un'applicazione Web che sfrutta la tecnologia blockchain per garantire sicurezza e immutabilità dei dati lungo l’intero ciclo di vita di un prodotto alimentare. La piattaforma è stata sviluppata adottando un’architettura composta da un middleware Node.js per l’interazione con la blockchain Hyperledger Fabric, un backend Flask per la gestione degli utenti e delle API RESTful, e un frontend React per la fruizione dei servizi da parte degli attori della filiera. L’obiettivo di questo lavoro è stato valutare la robustezza e la scalabilità della piattaforma attraverso uno stress test mirato, condotto sull’endpoint di upload dei prodotti. Per la sperimentazione è stato utilizzato Locust, strumento di load testing distribuito e containerizzato in Docker, che ha consentito di misurare metriche fondamentali come throughput, latenza e tasso di errore. I risultati ottenuti hanno evidenziato la presenza di colli di bottiglia principalmente legati alla fase di persistenza delle transazioni in blockchain, la quale introduce un incremento di latenza al crescere del carico. Sono emersi inoltre picchi iniziali dovuti all’instanziamento del chaincode e all’avvio delle connessioni peer-orderer. L’analisi critica dei dati ha permesso di delineare possibili soluzioni e sviluppi futuri, quali l’ottimizzazione del chaincode, l’introduzione di strategie di caching e il miglioramento della configurazione della rete Fabric. In conclusione, il lavoro dimostra come la blockchain rappresenti una tecnologia valida per la tracciabilità nella supply chain, pur richiedendo ottimizzazioni specifiche per garantire prestazioni adeguate in scenari di utilizzo reale e ad alto carico.
BLOCKCHAIN SOTTO PRESSIONE: VALUTAZIONE DELLE PRESTAZIONI DI UN SISTEMA BLOCKCHAIN-BASED PER LA TRACCIABILITÀ DELLA FILIERA PRODUTTIVA
RIZZI, DOMENICO
2024/2025
Abstract
The increasing digitalization of food supply chains, driven by the advent of Industry 4.0, has made product traceability an increasingly central issue. In this context, Filiera360 is a web application that leverages blockchain technology to ensure data security and immutability throughout the entire lifecycle of a food product. The platform was developed using an architecture composed of a Node.js middleware for interaction with the Hyperledger Fabric blockchain, a Flask backend for user management and RESTful APIs, and a React frontend for service access by the various supply chain actors. The goal of this work was to assess the robustness and scalability of the platform through a targeted stress test conducted on the product upload endpoint. For the experimentation, Locust—a distributed load testing tool containerized in Docker—was used to measure key metrics such as throughput, latency, and error rate. The results revealed bottlenecks mainly related to the persistence phase of blockchain transactions, which introduces increased latency as the load grows. Initial peaks were also observed due to chaincode instantiation and the establishment of peer-orderer connections. A critical analysis of the data made it possible to outline potential solutions and future developments, such as chaincode optimization, the introduction of caching strategies, and improvements in the Fabric network configuration. In conclusion, the study demonstrates that blockchain represents a valid technology for supply chain traceability, although specific optimizations are required to ensure adequate performance in real-world, high-load scenarios.I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14253/108487