Decoupling packet loss from blocking in proactive reservation-based switching

Abstract

We consider the maximization of network throughput in buffer-constrained optical networks using aggregate bandwidth allocation and reservation-based transmission control. Assuming that all flows are subject to loss-based TCP congestion control, we quantify the effects of buffer capacity constraints on bandwidth utilization efficiency through contention-induced packet loss. The analysis shows that the ability of TCP flows to efficiently utilize successful reservations is highly sensitive to the available buffer capacity. Maximizing the bandwidth utilization efficiency under buffer capacity constraints thus requires decoupling packet loss from contention-induced blocking of transmission requests. We describe a confirmed (two-way) reservation scheme that eliminates contention-induced loss, so that no packets are dropped at the network's core, and loss can be incurred only at the adequately buffer-provisioned ingress routers, where it is exclusively congestion-induced. For the confirmed signaling scheme, analytical and simulation results indicate that TCP aggregates are able to efficiently utilize the successful reservations independently of buffer constraints.

Analysis of a transmission scheduling algorithm for supporting bandwidth guarantees in bufferless networks

“…In the reservation framework, the ingress border router of a connection regularly generates requests for time slots at a rate equal to the connection's bandwidth allocation, and releases constant-size data frames into the network according to the schedule of acknowledged transmission requests. 4 The transmission requests are transmitted from the ingress router in the form of reservation packets that are processed by core link schedulers as they traverse the connection's path. A link scheduler may block requests due to contention.…”

“…4 Arriving packets are grouped into constant-size frames at the ingress router. 5 The volume of reservation traffic generated by a connection should be made much smaller than the volume of its data traffic by the appropriate choice of the frame size [19].…”

“…Given an expression for the tail of the blocking rate distribution of a connection, the network operator can overprovision the bandwidth allocated to the connection so that the expected rate of successful reservations is arbitrarily close to the nominal connection bandwidth allocation. Additional details about the reservation-based framework can be found in [4].…”