S.-M. Baek, H.-K. Rim, and S.-C. Kim (Korea)
Load balancing, TCP request Scheduling, Dispatching
The use of server clusters to provide Internet services can yield both high performance and cost-effectiveness. Contemporary approaches to cluster design must confront the tradeoff between dynamic load-balancing and efficiency when dispatching client requests to servers in the cluster. In this paper we describe a technique called TCP Connection Filtering, intended to be an easily implemented scheme for hosting Internet services on a server cluster in a way transparent to clients. We take a semi-centralized approach, in which client packets sent to the cluster’s single IP address are broadcasted to all of its servers. Each server makes responses to the packets requesting a new TCP session after some duration proportioned to its workload. And, the dispatcher passes only the first response packet, while rejects the others. Finally the first-responding server takes charge of the TCP session. This yields dynamically balanced load-distribution. We describe this approach in detail, and present results of simulations that show it achieves higher performance (in terms of throughput and reliability) than Least Connection First, a representative scheme of the traditional dynamic scheduling approaches.
Important Links:
Go Back
