Security for minimum storage regenerating codes and locally repairable codes
- 1 June 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 21, 1028-1032
Abstract
We consider the problem of designing repair efficient distributed storage systems, which are information-theoretically secure against a passive eavesdropper that can gain access to a limited number of storage nodes. We present a framework that enables design of a broad range of secure storage codes through a joint construction of inner and outer codes. As case studies, we focus on two specific families of storage codes: (i) minimum storage regenerating (MSR) codes, and (ii) maximally recoverable (MR) codes, which are a class of locally repairable codes (LRCs). The main idea of this framework is to utilize the existing constructions of storage codes to jointly design an outer coset code and inner storage code. Finally, we present a construction of an outer coset code over small field size to secure locally repairable codes presented by Tamo and Barg for the special case of an eavesdropper that can observe any subset of nodes of maximum possible size.Keywords
This publication has 13 references indexed in Scilit:
- Explicit Maximally Recoverable Codes With LocalityIEEE Transactions on Information Theory, 2014
- A Family of Optimal Locally Recoverable CodesIEEE Transactions on Information Theory, 2014
- Optimal Locally Repairable and Secure Codes for Distributed Storage SystemsIEEE Transactions on Information Theory, 2013
- Data secrecy in distributed storage systems under exact repair2013
- On the Locality of Codeword SymbolsIEEE Transactions on Information Theory, 2012
- Secure Network Coding for Wiretap Networks of Type IIIEEE Transactions on Information Theory, 2012
- Information-Theoretically Secure Regenerating Codes for Distributed Storage2011
- Securing Dynamic Distributed Storage Systems Against Eavesdropping and Adversarial AttacksIEEE Transactions on Information Theory, 2011
- Universal Secure Network Coding via Rank-Metric CodesIEEE Transactions on Information Theory, 2011
- Wire-Tap Channel IIAT&T Bell Laboratories Technical Journal, 1984