Throughput concavity and response time convexity
Information Processing Letters
Resource allocation problems: algorithmic approaches
Resource allocation problems: algorithmic approaches
Optimal allocation of multiple class resources in computer systems
SIGMETRICS '88 Proceedings of the 1988 ACM SIGMETRICS conference on Measurement and modeling of computer systems
Simulated annealing: theory and applications
Simulated annealing: theory and applications
The placement optimization program: a practical solution to the disk file assignment problem
SIGMETRICS '89 Proceedings of the 1989 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Optimal Partitioning of Cache Memory
IEEE Transactions on Computers
Approximation algorithms for NP-hard problems
Approximation algorithms for NP-hard problems
Disk load balancing for video-on-demand systems
Multimedia Systems
The art of computer programming, volume 3: (2nd ed.) sorting and searching
The art of computer programming, volume 3: (2nd ed.) sorting and searching
A Fast Selection Algorithm and the Problem of Optimum Distribution of Effort
Journal of the ACM (JACM)
Probability and statistics with reliability, queuing and computer science applications
Probability and statistics with reliability, queuing and computer science applications
Computer Performance Modeling Handbook
Computer Performance Modeling Handbook
Tabu Search
Dynamic Load Balancing on Web-Server Systems
IEEE Internet Computing
The state of the art in locally distributed Web-server systems
ACM Computing Surveys (CSUR)
A cost-oriented approach for infrastructural design
Proceedings of the 2004 ACM symposium on Applied computing
IEEE Transactions on Parallel and Distributed Systems
SLA based profit optimization in autonomic computing systems
Proceedings of the 2nd international conference on Service oriented computing
On Honey Bees and Dynamic Server Allocation in Internet Hosting Centers
Adaptive Behavior - Animals, Animats, Software Agents, Robots, Adaptive Systems
Dynamic placement for clustered web applications
Proceedings of the 15th international conference on World Wide Web
Provisioning servers in the application tier for e-commerce systems
ACM Transactions on Internet Technology (TOIT)
SLA based resource allocation policies in autonomic environments
Journal of Parallel and Distributed Computing
Fault-tolerant framework for load balancing system
SEPADS'06 Proceedings of the 5th WSEAS International Conference on Software Engineering, Parallel and Distributed Systems
A kind of spatial data storage system based on global subdivision theory for mobile GIS
WiCOM'09 Proceedings of the 5th International Conference on Wireless communications, networking and mobile computing
A weighted metric based adaptive algorithm for web server load balancing
IITA'09 Proceedings of the 3rd international conference on Intelligent information technology application
Rating the raters: a reputation system for wiki-like domains
Proceedings of the 3rd international conference on Security of information and networks
Distribution strategies for parallel VPN servers
ACOS'06 Proceedings of the 5th WSEAS international conference on Applied computer science
Characterizing problems for realizing policies in self-adaptive and self-managing systems
Proceedings of the 6th International Symposium on Software Engineering for Adaptive and Self-Managing Systems
Host selection through collective decision
ACM Transactions on Autonomous and Adaptive Systems (TAAS) - Special section on formal methods in pervasive computing, pervasive adaptation, and self-adaptive systems: Models and algorithms
Hi-index | 0.00 |
In this article we propose a novel, yet practical, scheme which attempts to optimally balance the load on the servers of a clustered Web farm. The goal in solving this performance problem is to achieve minimal average response time for customer requests, and thus ultimately achieve maximal customer throughput. The article decouples the overall problem into two related but distinct mathematical subproblems, one static and one dynamic. We believe this natural decoupling is one of the major contributions of our article. The static component algorithm determines good assignments of sites to potentially overlapping servers. These cluster assignments, which, due to overhead, cannot be changed too frequently, have a major effect on achievable response time. Additionally, these assignments must be palatable to the sites themselves. The dynamic component algorithm is designed to handle real-time load balancing by routing customer requests from the network dispatcher to the servers. This algorithm must react to fluctuating customer request load while respecting the assignments of sites to servers determined by the static component. The static and dynamic components both employ in various contexts the same so-called goal setting algorithm. This algorithm determines the theoretically optimal load on each server, given hypothetical cluster assignments and site activity. We demonstrate the effectiveness of the overall load-balancing scheme via a number of simulation experiments.