Abstract
There is a surging interest in applying erasure codes into distributed storage systems to ensure certain level of reliability because of its spatial efficiency compared with replication-based storage systems. The Low Density Parity Check (LDPC) codes has been a popular alternative compared with Reed-Solomon (R-S) codes. It has been shown that LDPC codes have lower computational cost and repair network traffic compared with R-S based solutions. In this paper, a detailed study on the relation between the reliability, repair traffic and spatial efficiency is provided. The lower bound on the length of the LDPC codes will be derived to achieve predefined reliability and repair traffic. A new algorithm to construct a guided system random LDPC codes is proposed to generate LDPC codes to achieve pre-defined reliability and repair traffic.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
