``A Practical Analysis of Low-Density Parity-Check Erasure Codes for Wide-Area Storage Applications''

James S. Plank and Michael G. Thomason.

Appearing in DSN-2004: The International Conference on Dependable Systems and Networks, Florence, Italy, June, 2004.

PDF of the paper.

If the material in this paper interests you, please also read the expanded version of the paper: Technical Report UT-CS-03-510, University of Tennessee, September, 2003.

Abstract

As peer-to-peer and widely distributed storage systems proliferate, tolerating multiple server failures effectively becomes crucial to performance and efficiency. At the heart of this sort of fault-tolerance lie erasure codes, most typically Reed-Solomon codes. However, more recently Low-Density Parity-Check (LDPC) codes have arisen as alternative codes that trade off vastly improved decoding performance for inefficiencies in the amount of data that must be acquired to perform decoding. The scores of papers written on LDPC codes typically analyze their collective and asymptotic behavior. Unfortunately, their practical application requires the generation and analysis of individual codes for finite systems.

This paper attempts to illuminate the practical considerations of LDPC codes for peer-to-peer and distributed storage systems. We employ simulation to generate a huge variety of individual codes, and then analyze their behavior in order to address several important heretofore unanswered questions about employing LDPC codes in real-world systems.


PDF of the paper.

Citation Information