STOC '97 Proceedings of the twenty-ninth annual ACM symposium on Theory of computing
Chord: a scalable peer-to-peer lookup protocol for internet applications
IEEE/ACM Transactions on Networking (TON)
Kademlia: A Peer-to-Peer Information System Based on the XOR Metric
IPTPS '01 Revised Papers from the First International Workshop on Peer-to-Peer Systems
Design Considerations for Distributed Caching on the Internet
Design Considerations for Distributed Caching on the Internet
SOSP '03 Proceedings of the nineteenth ACM symposium on Operating systems principles
A hierarchical internet object cache
ATEC '96 Proceedings of the 1996 annual conference on USENIX Annual Technical Conference
Dynamo: amazon's highly available key-value store
Proceedings of twenty-first ACM SIGOPS symposium on Operating systems principles
Smaller Universal Spiking Neural P Systems
Fundamenta Informaticae
| Hi-index | 0.00 |
Distributed caching system is usually used to alleviate database load in constructing an enterprise web application system. It helps to speed up dynamic web applications. In order to improve the utilization of caching cluster, an appropriate data partitioning and placement scheme is usually applied. This paper proposes a Balanced Partition Scheme (BPS) to solve load imbalance problems and highly skewed data requests in web application. In the BPS, which is based on consistent hash algorithm, the partition and placement schemes are designed respectively to guarantee a system's load balance even when the requests of this system are highly skewed. The range of hash function is divided into several groups equally and those groups will be relocated when caching nodes are overloaded. The implementation and evaluation of the BPS is also presented in this paper. The effectiveness of the BPS has been verified in the simulation experiment and the BPS can successfully solve load imbalance problems when faced with a large number of get/set requests.