SIGMOD '88 Proceedings of the 1988 ACM SIGMOD international conference on Management of data
Statistical Service Assurances for Applications in Utility Grid Environments
MASCOTS '02 Proceedings of the 10th IEEE International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems
Dynamic Provisioning of Multi-tier Internet Applications
ICAC '05 Proceedings of the Second International Conference on Automatic Computing
Adaptive overload control for busy internet servers
USITS'03 Proceedings of the 4th conference on USENIX Symposium on Internet Technologies and Systems - Volume 4
Load shedding in a data stream manager
VLDB '03 Proceedings of the 29th international conference on Very large data bases - Volume 29
Autonomic Provisioning of Backend Databases in Dynamic Content Web Servers
ICAC '06 Proceedings of the 2006 IEEE International Conference on Autonomic Computing
Communications of the ACM
Robust and flexible power-proportional storage
Proceedings of the 1st ACM symposium on Cloud computing
Characterizing, modeling, and generating workload spikes for stateful services
Proceedings of the 1st ACM symposium on Cloud computing
Automated control for elastic storage
Proceedings of the 7th international conference on Autonomic computing
The SCADS director: scaling a distributed storage system under stringent performance requirements
FAST'11 Proceedings of the 9th USENIX conference on File and stroage technologies
Hi-index | 0.00 |
In cloud, service performances are expected to meet various QoS requirements stably, and a great challenge for achieving this comes from the great workload fluctuations in stateful systems. So far, few previous works have endeavored for handling overload caused by such fluctuations. In this paper, we propose an efficient overload control strategy to solve this problem. Crucial server status information is indexed by R-tree to provide global view for data movement. Based on index, a two-step filtering approach is introduced to eliminate irrational server candidates. A server selection algorithm considering workload patterns is presented afterwards to acquire load-balancing effects. Extensive experiments are conducted to evaluate the performance of our strategy.