On the network impact of dynamic server selection

Widespread replication of information can ameliorate the problem of server overloading but raises the allied question of server selection. Clients may be assigned to a replica in a static manner or they may choose among replicas based on client-initiated measurements. The latter technique, called dynamic server selection (DSS), can provide signi®cantly improved response time to users when compared with static server assignment policies (for example, based on network distance in hops). In the ®rst part of this paper we demonstrate the idea of DSS using experiments performed in the Internet. We compare a range of policies for DSS and show that obtaining additional information about servers and paths in the Internet before choosing a server improves response time signi®cantly. The best policy we examine adopts a strategy of never adding more than 1% additional trac to the network, and is still able to provide nearly all the bene®ts of the most expensive policies. While these results suggest that DSS is bene®cial from the network user's standpoint, the system-wide e€ects of DSS schemes should also be closely examined. In the second part of this paper we use large-scale simulation to study the system-wide network impact of dynamic server selection. We use a simulated network of over 100 hosts that allows local-area e€ects to be distinguished from wide-area e€ects within trac patterns. In this environment we compare DSS with static server selection schemes and con®rm that client bene®ts remain even when many use DSS simultaneously. Importantly, we also show that DSS confers system-wide bene®ts from the network standpoint, as compared to static server selection. First, overall data trac volume in the network is reduced, since DSS tends to diminish network congestion. Second, trac distribution improves ± trac is shifted from the backbone to regional and local networks. Ó 1999 Elsevier Science B.V. All rights reserved.