Proof of Storage-Time: Efficiently Checking Continuous Data Availability

Giuseppe Ateniese, Long Chen, Mohammad Etemad, Qiang Tang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

20 Scopus citations

Abstract

A high-quality outsourced storage service is crucial for many existing applications. For example, hospitals and data centers need to guarantee the availability of their systems to perform routine daily activities. Such a system should protect users against downtime and ensure data availability over time. Continuous data availability is a critical property to measure the quality of an outsourced storage service, which implies that outsourced data is continuously available to the server during the entire storage period. We formally study the Proof of Storage-Time (PoSt), the notion initially proposed in the Filecoin whitepaper, which enables a verifier to audit the continuous data availability of an outsourced storage service. We provide a formal security model of PoSt and generic constructions that are proven secure under our definition. Moreover, our concrete instantiation can yield a PoSt protocol with an extremely efficient verification: a single hash computation to verify a proof of size around 200 bits. This makes our scheme applicable even in the decentralized storage marketplace enabled by blockchain.

Original languageEnglish
Title of host publication27th Annual Network and Distributed System Security Symposium, NDSS 2020
ISBN (Electronic)1891562614, 9781891562617
DOIs
StatePublished - 2020
Event27th Annual Network and Distributed System Security Symposium, NDSS 2020 - San Diego, United States
Duration: 23 Feb 202026 Feb 2020

Publication series

Name27th Annual Network and Distributed System Security Symposium, NDSS 2020

Conference

Conference27th Annual Network and Distributed System Security Symposium, NDSS 2020
Country/TerritoryUnited States
CitySan Diego
Period23/02/2026/02/20

Fingerprint

Dive into the research topics of 'Proof of Storage-Time: Efficiently Checking Continuous Data Availability'. Together they form a unique fingerprint.

Cite this