Load balance algorithms for anycast

摘要: Increasingly, replicated anycast servers are being used to deliver network applications and service ever increasing user requests. Therefore, the strategies guarantee bandwidth prerequisites perform load balancing across nodes of an group critical performance online applications. In this paper, we model requests, congestion latency, server using a combination hydro-dynamics queuing theory develop efficient job distribution strategy. Current, research does not explicitly consider system within groups when distributing heavily loaded can quickly become congested uneven as jobs routed closely linked which already saturated with comparison, further away systems remain relatively unused because other issues such latency during these times. Our redirects requests from busy idle, remotely nodes, process faster in spite slower access. Using empirical study, show technique improve request performance, throughput minimal probing overhead.