Table of Contents
Fetching ...

Efficient Repair of (k+2, k) Degraded Read Friendly MDS Array Codes With Sub-packetization 2

Jie Li, Xiaohu Tang

TL;DR

The paper advances practical high-rate MDS array codes by introducing two degraded-read-friendly constructions with sub-packetization $N=2$ and two parity nodes, valid for arbitrary code length. Construction 1 achieves asymptotically optimal average repair bandwidth for $(n=k+2,k)$ codes over small finite fields, while Construction 2 provides two repair mechanisms to balance repair bandwidth and rebuilding access depending on helper-node computation. Both constructions operate over relatively small field sizes (e.g., $q=2^{2t}$) and expand the parameter range beyond previously known $(n=k+2,k)$ DRF codes with $N=2$. Collectively, these results reduce repair traffic and disk I/O during degraded reads, improving efficiency and resilience in distributed storage systems.

Abstract

In this paper, we present two constructions of degraded read friendly (DRF) MDS array codes with two parity nodes and a sub-packetization level of 2 over small finite fields, applicable for any arbitrary code length. The first construction achieves the smallest repair bandwidth among all existing constructions with the same parameters, and is asymptotically optimal with respect to the lower bound on the average repair bandwidth characterized by Zhang et al. The second construction supports two repair mechanisms, depending on whether computation within the helper nodes is permitted or not during the node repair process, thereby optimizing either the repair bandwidth or the rebuilding access.

Efficient Repair of (k+2, k) Degraded Read Friendly MDS Array Codes With Sub-packetization 2

TL;DR

The paper advances practical high-rate MDS array codes by introducing two degraded-read-friendly constructions with sub-packetization and two parity nodes, valid for arbitrary code length. Construction 1 achieves asymptotically optimal average repair bandwidth for codes over small finite fields, while Construction 2 provides two repair mechanisms to balance repair bandwidth and rebuilding access depending on helper-node computation. Both constructions operate over relatively small field sizes (e.g., ) and expand the parameter range beyond previously known DRF codes with . Collectively, these results reduce repair traffic and disk I/O during degraded reads, improving efficiency and resilience in distributed storage systems.

Abstract

In this paper, we present two constructions of degraded read friendly (DRF) MDS array codes with two parity nodes and a sub-packetization level of 2 over small finite fields, applicable for any arbitrary code length. The first construction achieves the smallest repair bandwidth among all existing constructions with the same parameters, and is asymptotically optimal with respect to the lower bound on the average repair bandwidth characterized by Zhang et al. The second construction supports two repair mechanisms, depending on whether computation within the helper nodes is permitted or not during the node repair process, thereby optimizing either the repair bandwidth or the rebuilding access.
Paper Structure (9 sections, 7 theorems, 38 equations, 1 table)

This paper contains 9 sections, 7 theorems, 38 equations, 1 table.

Key Result

Lemma 1

Let $q=2^{2t}$ for some positive integer $t$, and let $w$ be a primitive element of the finite field $\mathbf{F}_q$. Then we have

Theorems & Definitions (17)

  • Remark 1
  • Lemma 1
  • proof
  • Theorem 1
  • proof
  • Theorem 2
  • proof
  • Theorem 3
  • proof
  • Theorem 4
  • ...and 7 more