Comparative Analysis of SEC DED, EVEN-ODD, and Pyramid Codes for Distributed Storage Systems: A MATLAB-Based Study
This study presents a comparative analysis of three erasure coding schemes—Single Error Correction Double Error Detection (SEC DED), EVEN-ODD, and Pyramid codes—within distributed storage systems, utilizing MATLAB for simulation. The research focuses on key metrics such as storage overhead, recovery...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
EDP Sciences
2025-01-01
|
| Series: | ITM Web of Conferences |
| Online Access: | https://www.itm-conferences.org/articles/itmconf/pdf/2025/04/itmconf_iwadi2024_03021.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This study presents a comparative analysis of three erasure coding schemes—Single Error Correction Double Error Detection (SEC DED), EVEN-ODD, and Pyramid codes—within distributed storage systems, utilizing MATLAB for simulation. The research focuses on key metrics such as storage overhead, recovery time, and fault tolerance to elucidate the performance characteristics and practical applicability of each scheme. Single Error Correction Double Error Detection coding balances recovery speed and fault tolerance but incurs higher storage overhead due to its double-parity structure. EVEN-ODD coding, renowned for its storage efficiency and rapid recovery capabilities, exhibits limitations in fault tolerance. Conversely, Pyramid codes offer robust multidimensional error correction, making them suitable for large-scale storage environments, albeit at the cost of increased storage and extended recovery times. By examining these schemes under various scenarios, the analysis highlights their respective trade-offs and provides insights for selecting the optimal coding scheme based on specific system requirements. The study concludes with recommendations for enhancing erasure coding effectiveness in distributed environments and suggests areas for future research, including hybrid coding schemes and machine learning optimizations. |
|---|---|
| ISSN: | 2271-2097 |