To cope with rapidly increasing Internet usage nowadays, providing Internet services using multiple servers has become a necessity. To ensure sufficient service quality and server utilization at the same time, effective methods are needed to spread load among servers properly. Existing load balancing methods often assume servers are homogeneous and consider only one type of resource, such as CPU. Such methods suffer from the fact that different requests often demand multiple types of resources with different requirements; trying to balance the usage of only one resource type may induce an inadvertent performance bottleneck, leading to low resource utilization and service quality. To address this problem, we propose a load balancing method based on the concept of distributed market mechanism, where requests are priced with respect to the load of multiple resources on each server. By migrating jobs among servers to balance inter-server load and minimize intra-server job cost at the same time, our method shows significant improvement in terms of load imbalance degrees, server utilization, and response time when compared to other published methods, especially when server heterogeneity increases.

Modern Web-server systems use multiple servers to handle an increased user demand. Such systems need effective methods to spread the load among web servers evenly in order to keep web server utilization high while providing sufficient quality of service for end users. In conventional DNS-based load balancing architecture, a Doman Name Server (DNS) dispatches requests to web servers based on their load status. Because web servers need to inform the DNS server about their load status from time to time, a so-called load buffer range is often employed to reduce the update frequency. Without care, however, using a load buffer range may result in load oscillation among web servers. To address this problem, we propose a Random Early Detection (RED) method with the intuition that the probability for a web server to become overloaded in near future is directly proportional to its current load. Simulation confirms that our method helps reducing the oscillation of the web server load significantly.