Cross-Rack Update Bandwidth for Distributed Storage Systems
Authors:
- 4 December 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3699-3704
Abstract
In distributed storage systems, storage nodes are organized in racks in which the cross-rack communication cost is much more expensive than the intra-rack communication cost. It is critical to design erasure codes that minimize the cross-rack update bandwidth which is defined as the total amount of symbols transferred across different racks during an update process. In this paper, we analyze the cross-rack update bandwidth of erasure codes for distributed storage systems. We derive a lower bound on the cross-rack update bandwidth and show that the proposed lower bound is achievable under certain parameters.Keywords
Funding Information
- National Natural Science Foundation of China (62071121,61901115,12025104)
- National Key R&D Program of China (2020YFA0712300)
This publication has 16 references indexed in Scilit:
- STAR+ Codes: Triple-Fault-Tolerant Codes with Asymptotically Optimal Updates and Efficient Encoding/Decoding2021
- An Efficient Piggybacking Design with Lower Repair Bandwidth and Lower Sub-packetization2021
- Optimal Rack-Coordinated Updates in Erasure-Coded Data Centers2021
- Toward Optimality in Both Repair and Update via Generic MDS Code Transformation2020
- Minimum Storage Rack-Aware Regenerating Codes with Exact Repair and Small Sub-Packetization2020
- TIP-Code: A Three Independent Parity Code to Tolerate Triple Disk Failures with Optimal Update Complextiy2015
- Partial-MDS Codes and Their Application to RAID Type of ArchitecturesIEEE Transactions on Information Theory, 2013
- Zigzag Codes: MDS Array Codes With Optimal RebuildingIEEE Transactions on Information Theory, 2012
- Network Coding for Distributed Storage SystemsIEEE Transactions on Information Theory, 2010
- On lowest density MDS codesIEEE Transactions on Information Theory, 1999