Cryptology ePrint Archive: Report 2018/654

Proofs of Replicated Storage Without Timing Assumptions

Ivan Damgård and Chaya Ganesh and Claudio Orlandi

Abstract: In this paper we provide a formal treatment of proof of replicated storage, a novel cryptographic primitive recently proposed in the context of a novel cryptocurrency, namely Filecoin.

In a nutshell, proofs of replicated storage is a solution to the following problem: A user stores a file $m$ on $n$ different servers to ensure that the file will be available even if some of the servers fail. Using proof of retrievability, the user could check that every server is indeed storing the file. However, what if the servers collude and, in order to save on resources, decide to only store one copy of the file? A proof of replicated storage guarantees that, unless the server is indeed reserving the space necessary to store $n$ copies of the file, the user will not accept the proof. While some candidate proofs of replicated storage have already been proposed, their soundness relies on timing assumptions i.e., the user must reject the proof if the prover does not reply within a certain time-bound.

In this paper we provide the first construction of a proof of replication which does not rely on any timing assumptions.

Category / Keywords: replicated storage, proof of replication, encodings

Original Publication (with minor differences): IACR-CRYPTO-2019

Date: received 3 Jul 2018, last revised 29 Jun 2019

Contact author: ivan at cs au dk, chaya ganesh at gmail com, orlandi at cs au dk

Available format(s): PDF | BibTeX Citation

Note: comparison with related work, clarified instantiation of oracle

Version: 20190629:172249 (All versions of this report)

Short URL:

[ Cryptology ePrint archive ]