Designing a DHT for Low Latency and High Throughput
By Frank Dabek, Jinyang Li, Emil Sit, James Robertson, M. Frans Kaashoek, Robert Morris. Symposium on Networked Systems Design and Implementation, 2004
432 citations recorded by the source corpus.
View Paper: https://www.semanticscholar.org/paper/15bbc0d5d3125d3086b0571d4562ffab2cc54bdf
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- John Risson
- A. Dimakis
- Frank Dabek
- Andreas Haeberlen
- D. Wallach
- Peter Druschel
- Giuseppe Ateniese
- Brighten Godfrey
Citations in the corpus
- TAO: Facebook's Distributed Data Store for the Social Graph (2013, background): “Li et al. [16] characterize the performance of DHTs under churn while Dabek et al. [9] focus on designing DHTs in a wide-area network.”
- MR-PDP: Multiple-Replica Provable Data Possession (2008, background): “Systems re-replicate data when replicas fail [10, 12], evaluate the correctness of replicas in the system [20], and move replicas among sites to meet availability goals [2,9].”
- OpenDHT: a public DHT service and its uses (2005, unknown): “As such, in evaluating OpenDHT’s performance in Section 5.1, we do not focus on the detailed behavior of the underlying DHT routing or storage algorithms, both of which have been evaluated over short periods elsewhere [8, 10 , 29].”
- Network Coding for Distributed Storage Systems (2007, methodology): “Several designs [8], [4], [5] use erasure codes instead of r eplication.”
- RACS: a case for cloud storage diversity (2010, methodology): “Peer-to-peer storage [19–21, 32] have employed RAID-like techniques to ensure the availability and durability of data.”
- Survey of research towards robust peer-to-peer networks: Search methods (2006, background): “One may increase the number of replicas [150], use parallel lookups [291, 316], use alternate routes on failure [150], or use multiple gateway nodes to enter the DHT [317].”
- Efficient Replica Maintenance for Distributed Storage Systems (2006, methodology): “The line labeled Cates shows a system that keeps track of exactly r L replicas per object; this system approximates the behavior of DHTs like DHash, PAST and OpenDHT.”
- Eclipse Attacks on Overlay Networks: Threats and Defenses (2006, methodology): “The DHash [ 10 ] DHT provides flexibility in selecting fingers so that close by nodes can be used as fingers in Chord, thereby making it vulnerable to Eclipse attacks.”