Erasure Coding Vs. Replication: A Quantitative Comparison
IPTPS '01 Revised Papers from the First International Workshop on Peer-to-Peer Systems
SOSP '03 Proceedings of the nineteenth ACM symposium on Operating systems principles
Disk failures in the real world: what does an MTTF of 1,000,000 hours mean to you?
FAST '07 Proceedings of the 5th USENIX conference on File and Storage Technologies
Optimal recovery of single disk failure in RDP code storage systems
Proceedings of the ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Network coding for distributed storage systems
IEEE Transactions on Information Theory
Availability in globally distributed storage systems
OSDI'10 Proceedings of the 9th USENIX conference on Operating systems design and implementation
Rethinking erasure codes for cloud file systems: minimizing I/O for recovery and degraded reads
FAST'12 Proceedings of the 10th USENIX conference on File and Storage Technologies
IEEE Transactions on Information Theory
Erasure coding in windows azure storage
USENIX ATC'12 Proceedings of the 2012 USENIX conference on Annual Technical Conference
XORing elephants: novel erasure codes for big data
Proceedings of the VLDB Endowment
Hi-index | 0.00 |
Erasure codes, such as Reed-Solomon (RS) codes, are being increasingly employed in data centers to combat the cost of reliably storing large amounts of data. Although these codes provide optimal storage efficiency, they require significantly high network and disk usage during recovery of missing data. In this paper, we first present a study on the impact of recovery operations of erasure-coded data on the data-center network, based on measurements from Facebook's warehouse cluster in production. To the best of our knowledge, this is the first study of its kind available in the literature. Our study reveals that recovery of RS-coded data results in a significant increase in network traffic, more than a hundred terabytes per day, in a cluster storing multiple petabytes of RS-coded data. To address this issue, we present a new storage code using our recently proposed Piggybacking framework, that reduces the network and disk usage during recovery by 30% in theory, while also being storage optimal and supporting arbitrary design parameters. The implementation of the proposed code in the Hadoop Distributed File System (HDFS) is underway. We use the measurements from the warehouse cluster to show that the proposed code would lead to a reduction of close to fifty terabytes of cross-rack traffic per day.