Cryptomonnaie | EPFL Graph Search

Une cryptomonnaie, dite aussi cryptoactif, cryptodevise, monnaie cryptographique ou encore cybermonnaie, est une monnaie numérique (actif numérique) émise de pair à pair, sans nécessité de banque centrale, utilisable au moyen d'un réseau informatique décentralisé. Elle utilise des technologies de cryptographie et associe l'utilisateur aux processus d'émission et de règlement des transactions. Au , selon CoinMarketCap, il existe , ou cryptomonnaies, pour une valeur de d’euros. Principes de fonctionnement Blockchain Blockchain Une cryptomonnaie repose sur une blockchain, un registre distribué (ou grand livre de comptes), consultable par tous, qui répertorie l'ensemble des actions du réseau depuis l'origine. Les informations à ajouter sont appelées transactions, et sont groupées dans des blocs. Une transaction peut par exemple être un transfert de cryptomonnaie d'une personne à une autre. Les acteurs du réseau, appelés nœuds, possèdent, stockent et vérifient le

A Standalone FPGA-Based Miner for Lyra2REv2 Cryptocurrencies

Alexios Konstantinos Balatsoukas Stimming, Pascal Giard, Michiel Van Beirendonck

Lyra2REv2 is a hashing algorithm that consists of a chain of individual hashing algorithms, and it is used as a proof-of-work function in several cryptocurrencies. The most crucial and exotic hashing algorithm in the Lyra2REv2 chain is a specific instance of the general Lyra2 algorithm. This work presents the first hardware implementation of the specific instance of Lyra2 that is used in Lyra2REv2. Several properties of the aforementioned algorithm are exploited in order to optimize the design. In addition, an FPGA-based hardware implementation of a standalone miner for Lyra2REv2 on a Xilinx Multi-Processor System on Chip is presented. The proposed Lyra2REv2 miner is shown to be significantly more energy efficient than both a GPU and a commercially available FPGA-based miner. Finally, we also explain how the simplified Lyra2 and Lyra2REv2 architectures can be modified with minimal effort to also support the recent Lyra2REv3 chained hashing algorithm.

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC2020

Scaling Byzantine Fault Tolerance

Matej Pavlovic

Online services are becoming more and more ubiquitous and keep growing in scale. At the same time, they are required to be highly available, secure, energy-efficient, and to achieve high performance. To ensure these (and many other) properties, replication and distribution of these services becomes inevitable. Indeed, todayâs online services often involve thousands of processes running on different machines interconnected by a communication network. These processes may experience various kinds of failures, from simply crashing to being compromised by a malicious (Byzantine) adversary. The classic method for dealing with Byzantine faults is state machine replication (SMR). However, SMR fundamentally relies on a solution of the consensus problem, which often proves to be a scalability bottleneck. This dissertation addresses the scalability of Byzantine fault-tolerant systems. We argue that, for a certain class of applications, consensus either does not need to be solved at all, or only needs to be solved among a limited number of processes. By circumventing the consensus problem where solving it is not necessary, we improve the scalability of these applications. We start by focusing on the particular problem of distributed asset transfer, where digital assets are being transferred between user accountsâa problem underlying many cryptocurrency systems, most of which address it using Byzantine fault-tolerant SMR (and thus consensus). We show that consensus is not required for asset transfer by defining it as a sequential object type in the shared memory model and proving that it has consensus number 1 in Herlihyâs hierarchy. We further generalize the asset transfer object type, allowing an account to be shared by up to k owners. We prove that the consensus number of such an object type is k. We also discuss the asset transfer problem in the message passing model. We devise a consensusless asset transfer algorithm that relies on a secure broadcast primitive that, unlike consensus, has fully asynchronous deterministic implementations. Furthermore, since deterministic implementations of secure broadcast have limited scalability, we propose probabilistic secure broadcast, a variant of secure broadcast where some properties are allowed to be violated with a bounded probability. We design a highly scalable randomized algorithm that implements probabilistic secure broadcast with an arbitrarily low bound on the failure probability. Finally, we present Atum, a system for scalable group communication in a Byzantine environment that supports high churn. Atum achieves scalability by partitioning the system into groups of logarithmic size, only executing a consensus protocol inside each group.

EPFL2019

Scaling Byzantine Fault Tolerance

Matej Pavlovic

EPFL2019