STOC '97 Proceedings of the twenty-ninth annual ACM symposium on Theory of computing
Compact, adaptive placement schemes for non-uniform requirements
Proceedings of the fourteenth annual ACM symposium on Parallel algorithms and architectures
The power of two choices in randomized load balancing
The power of two choices in randomized load balancing
Dynamic and Redundant Data Placement
ICDCS '07 Proceedings of the 27th International Conference on Distributed Computing Systems
SPREAD: an adaptive scheme for redundant and fair storage in dynamic heterogeneous storage systems
Proceedings of the nineteenth annual ACM-SIAM symposium on Discrete algorithms
On the improvement of active XML (AXML) representation and query evaluation
Information Systems Frontiers
Hi-index | 0.01 |
Current p2p environments are often based on Consistent Hashing as underlying distributed hash table (DHT). The drawback of Consistent Hashing for small or mid-sized environments is that it is not able to efficiently use the available storage capacity, if the environment consists of heterogeneous peers. Inside this paper, we investigate the Redundant Share strategy and introduce the peer-Replication strategy, which are able to ensure an optimal capacity efficiency, even in case of data replication. While Redundant Share is not always able to retrieve all copies inside the view of a client, peer-Replication achieves this property by introducing a small number of additional communication rounds. The trade-off between the proposed peer-Replication strategy and Consistent Hashing is the number of required communication rounds vs. the quality of the data distribution. Inside this paper we show that a very small number of additional communication rounds enables us to significantly increase the capacity efficiency.