Reputation-Based Consensus Protocol for Blockchain-Based Scientific Workflow Provenance
Faculty Mentor Information
Matthew Miller, Marshall University; Gaby Dagher, Boise State University; and Min Long, Boise State University
Presentation Date
7-2023
Abstract
A scientific workflow consists of ordered, non-linear tasks connected by their inputs and outputs. Provenance records play a crucial role in documenting the history of these scientific workflows, including data origin, analysis stages, parameters used, intermediate computations, and transformations. Such records ensure clarity, reproducibility, and credibility for subsequent scientific investigations. However, a 2021 survey on scientific misconduct revealed approximately 8% of researchers admit to fabricating or falsifying scientific data between 2017 and 2020. This emphasizes the need for a solution that can reliably store provenance records. Blockchain technology offers answers by providing immutability, cryptographic verification, and transparency to the data it stores. A potential solution suggests storing provenance records on a blockchain- this allows scientists to collaborate and verify the integrity of the records themselves. Most existing implementations operate under the assumption of an effective consensus mechanism without actually implementing one. We propose a blockchain-enabled scientific workflow provenance audit with reputation-based consensus capabilities. Our solution aims to address gaps in other solutions by focusing on consensus mechanisms in blockchain-based scientific workflow provenance.
Reputation-Based Consensus Protocol for Blockchain-Based Scientific Workflow Provenance
A scientific workflow consists of ordered, non-linear tasks connected by their inputs and outputs. Provenance records play a crucial role in documenting the history of these scientific workflows, including data origin, analysis stages, parameters used, intermediate computations, and transformations. Such records ensure clarity, reproducibility, and credibility for subsequent scientific investigations. However, a 2021 survey on scientific misconduct revealed approximately 8% of researchers admit to fabricating or falsifying scientific data between 2017 and 2020. This emphasizes the need for a solution that can reliably store provenance records. Blockchain technology offers answers by providing immutability, cryptographic verification, and transparency to the data it stores. A potential solution suggests storing provenance records on a blockchain- this allows scientists to collaborate and verify the integrity of the records themselves. Most existing implementations operate under the assumption of an effective consensus mechanism without actually implementing one. We propose a blockchain-enabled scientific workflow provenance audit with reputation-based consensus capabilities. Our solution aims to address gaps in other solutions by focusing on consensus mechanisms in blockchain-based scientific workflow provenance.