2019•Unpublished venueRequires access

Improved Constructions for Optimal Multi-erasure Locally Recoverable Codes for Big Data Storage

Jianfa Qian, Lina Zhang

Open publisher page 0 citations

Abstract

Multi-erasure locally recoverable codes play a very significant role in distributed data storage. The advantage of multi-erasure locally recoverable codes is that it has local and global erasure-correcting characteristics. Recently, based on classical algebraic geometry codes, Huang et al. constructed a family of explicit optimal multi-erasure locally recoverable codes over small finite fields F4. In this work, based on the work of Huang et al., we use cyclic codes to construct a family of new optimal multi-erasure locally recoverable codes over small finite fields F3. It turns out that our multi-erasure locally recoverable codes have smaller finite fields than the previously known results.

About this research paper

What this paper is about

Multi-erasure locally recoverable codes play a very significant role in distributed data storage. The advantage of multi-erasure locally recoverable codes is that it has local and global erasure-correcting characteristics. Recently, based on classical algebraic geometry codes, Huang et al. constructed a family of explicit optimal multi-erasure locally recoverable codes over small finite fields F4. In this work, based on the work of Huang et al., we use cyclic codes to construct a family of new optimal multi-erasure locally recoverable codes over small finite fields F3. It turns out that our multi-erasure locally recoverable codes have smaller finite fields than the previously known results.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Multi-erasure locally recoverable codes play a very significant role in distributed data storage. The advantage of multi-erasure locally recoverable codes is that it has local and global erasure-correcting characteristics. Recently, based on classical algebraic geometry codes, Huang et al. constructed a family of explicit optimal multi-erasure locally recoverable codes over small finite fields F4. In this work, based on the work of Huang et al., we use cyclic codes to construct a family of new optimal multi-erasure locally recoverable codes over small finite fields F3. It turns out that our multi-erasure locally recoverable codes have smaller finite fields than the previously known results.

Key concepts: Erasure, Erasure code, Online codes, Construct (python library), Computer science, Finite field, Tornado code, Storage efficiency

Related papers

Back to paper searchBrowse research topicsOriginal source
Improved Constructions for Optimal Multi-erasure Locally Recoverable Codes for Big Data Storage — Research Paper | ScholarLens