Asynchronous verifiable information dispersal

  • Authors:
  • Christian Cachin;Stefano Tessaro

  • Affiliations:
  • Zurich Research Laboratory, IBM Research, Rüschlikon, Switzerland;Department of Computer Science, ETH Zurich, Zürich, Switzerland

  • Venue:
  • DISC'05 Proceedings of the 19th international conference on Distributed Computing
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

We consider the distribution of data by a client among a set of n storage servers, of which up to t might be faulty exhibiting arbitrary, i.e., Byzantine, behavior. The goal is to ensure that clients can always recover the stored data correctly, independently from the behavior of faulty servers or other, faulty clients. An inefficient solution is based on replication such that every server keeps a copy of the data. The classic alternative is information dispersal (IDA): using an erasure code, the data is split into blocks such that each server holds exactly one block and only a subset of the blocks is needed in order to reconstruct the data.