High Availability in DHTs: Erasure Coding vs. Replication

By Rodrigo Seromenho Miragaia Rodrigues, Barbara H. Liskov. International Workshop on Peer-to-Peer Systems, 2005

424 citations recorded by the source corpus.

Subfields: Computer Networks and Communications

View Paper: https://www.semanticscholar.org/paper/4b49d374c9306b929743e7d213c28cd47fc2d4fc

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

Citations in the corpus

Using redundancy to cope with failures in a delay tolerant network (2005, background): “Encouragingly, some basic studies show the advantages of using erasure coding with replication [23, 16].”
Rethinking erasure codes for cloud file systems: minimizing I/O for recovery and degraded reads (2012, background): “Two recent research projects have demonstrated how the RAID6 codes RDP and EVENODD may recover from single disk failures by reading significantly smaller subsets of codeword symbols than the previous standard practice of recovering from the parity drive [51, 49].”
Availability in Globally Distributed Storage Systems (2010, background): “Distributed storage systems increase resilience to failures by using replication [2] or erasure encoding across nodes [28].”
Coding for fast content download (2012, background): “The cost of this repair regeneration was considered much higher in coded than in replication system [2], until [1] established existence and advantages of new regenerating codes.”
On the Delay-Storage Trade-Off in Content Download from Coded Distributed Storage Systems (2013, background): “But in general, the cost of repair regeneration was considered much higher in coded than in replication systems [8], until [6] established existence and advantages of new regenerating codes.”
TOFEC: Achieving optimal throughput-delay trade-off of cloud storage using erasure codes (2013, background): “FEC has also been extensively studied in the context of distributed storage from the points of high durability and availability while attaining high storage efficiency [15], [16], [17].”
Efficient Replica Maintenance for Distributed Storage Systems (2006, background): “For systems that use erasure codes, there is an additional read cost since a complete copy of the object is needed in order to generate a new fragment [32].”
Analyzing the download time of availability codes (2015, background): “Excessive network traffic generated during recovery aggravates the contention at system resources and degrades the overall system performance [45].”
XORing Elephants: Novel Erasure Codes for Big Data (2013, methodology): “RS codes are usually repaired with the simple method that requires transferring 10 blocks and recreating the original 10 data blocks even if a single block is lost [29], hence creating a 10× overhead in repair bandwidth and disk I/O.”