Weak fairness guarantees that continuously enabled actions are executed infinitely often. Strong fairness, on the other hand, guarantees that actions that are enabled infinitely often (but not necessarily continuously) are executed infinitely often. In this paper, we present a distributed algorithm for scheduling actions for execution. Assuming weak fairness for the execution of this algorithm,...

We present a method for providing response-time guarantees for multimedia traffic in Asynchronous Transfer Mode (ATM) networks. The method extends our ResponseTime Analysis (RTA) for ATM networks to better capture the behaviour of MPEG-type multimedia traffic. RTA is intended for admission control of hard real-time traffic, i.e., it determines if a new connection can be admitted without violati...

—This paper presents a new rate based call gapping method. The main advantage is that it provides maximal through-put, priority handling and fairness for traffic classes without queues, unlike Token Bucket which provides only the first two or Weighted Fair Queuing that uses queues. The Token Bucket is used for call gapping because it has good throughput characteristics. For this reason we prese...

We explain why and how a crossbar switch with weighted fair queues (a deterministic device) converges to weighted max-min fair rates of service. Based on a plausible technical hypothesis we prove that a small mean number of buffers suffices to this end. WMMF-rates comprise the unique Nash equilibrium of a natural switching game—a fact that supports further their use.

We study the packet dispersion phenomenon that a traffic flow experiences when it passes through a router. We show that when there are competing flows and the router schedules packets first-come-first-served (FCFS), the dispersion is not described well by the bottleneck spacing effect. We therefore introduce the term FCFS-spacing effect. We also show that for a router implementing weighted fair...

Today’s cloud networks are shared among many tenants. Bandwidth guarantees and work conservation are two key properties to ensure predictable performance for tenant applications and high network utilization for providers. Despite significant efforts, very little prior work can really achieve both properties simultaneously even some of them claimed so. In this paper, we present QShare, a compreh...

In this paper, we investigate the eeectiveness of packet-drop mechanisms in conjunction with fair queuing link scheduling and hierarchical link sharing. Under fair queuing, the link share of a ow changes dynamically due to the arrivals and departures of ows and their bursts. This phenomenon becomes more pronounced in the case of hierarchical link sharing. Packet-drop mechanisms play an integral...

This work aims to analyse different design proposals for a dynamic Weighted Fair Queuing (WFQ) scheduler to provide Quality of Service (QoS) guarantees in general at IP level and specifically for the applications of the PHOENIX project. Several possible design options and configuration parameters have been investigated and studied for a dynamic scheduling discipline able to support QoS in a rel...

There i s an increasing demand for Internet core nodes to have (I) quality-of-service (QoS) guarantees and (2) higher switching capacity. The combined input and output queued (CiOQ) switching has been shown to be a promising solution to meet such demand. However, many of the proposed scheduling algorithms fo r a CIOQ switch cannot be practically implemented due to problems from scalability and ...