Predictable Packet Lossand Proportional Buffer Scaling Mechanism

Abstract—Packet losses at IP network are common behavior at the time of congestion. The TCP traffic is explained as in terms of load and capacity. The load should be measured as number of sender actively competes for a bottleneck link and the capacity as the total network buffering available to those senders. Though there are many congestion mechanism already in practice like congestion window, slow start, congestion avoidance, fast transmit but still we see erratic behavior when there is a large traffic. The TCP protocol that controls sources send rates degrades rapidly if the network cannot store at least a few packets per active connection. Thus the amount of router buffer space required for good performance scales with the number of active connections and the bandwidth utilization by each active connections. As in the current practice, the buffer space does not scale in this way and router drops the packet without looking at bandwidth utilization of each connections. The result is global synchronization and phase effect as well as packet from the unlucky sender will be frequently dropped. The simultaneous requirements of low queuing delay and of large buffer memories for large numbers of connections pose a problem. Routers should enforce a dropping policy by proportional to the bandwidth utilization by each active connection. Router will provision the buffering mechanism when processing slows down. This study explains the existing problem with drop-tail and RED routers and proposes the new mechanism to predict the effective bandwidth utilization of the clients depending on their history of utilization and drop the packet in different pattern after analyzing the network bandwidth utilization at each specific interval of time Keywords—Random Early Detection; Drop-Tail; packet dropping probability