Decoupling bandwidth and delay properties in class based queuing

2003

Eur. Trans. Telecomm.

SUMMARYCircuit switching, suited to providing real-time services due to the low and fixed switching delay, is not cost effective for building integrated services networks because it is based on static allocation of resources which is not efficient with bursty data traffic. Moreover it cannot handle flows that are not integer multiple of 64 Kb/s, preventing the usage of low bit rate codecs.This work explores the most suitable alternatives to the circuit switching technology (i.e. packet/cell switching) from the efficiency point of view, assuming that a PGPS scheduler is deployed in the network nodes. The paper defines an index to measure the efficiency of packet telephony, i.e. the volume of real-time traffic with deterministically guaranteed quality plus the amount of data carried related to the amount of network resources used. Furthermore it determines the maximum efficiency obtainable by packet networks, compares different network technologies and explores the problems of the deploying of low bit-rate codecs.

Section: Delay Bound and Call Admission Controlmentioning

confidence: 99%

On the efficiency of PGPS-based packet and cell switching technologies for traffic with guaranteed delay

2003

Eur. Trans. Telecomm.

“…An in-depth analysis of CBQ, on the other hand, shows several problems; most of them have been pointed out in [3] and [7]. This paper aims at the completion of [7] by summarizing the D-CBQ characteristics, presenting the implementation issues, and characterizing this new scheduler.…”

Section: Introductionmentioning

confidence: 99%

Implementation and Characterization of an Advanced Scheduler

2001

Networking — ICN 2001

Decoupled-CBQ, a CBQ derived scheduler, has been proved being a substantial improvement over CBQ. D-CBQ main advantages are a new set of rules for distributing excess bandwidth and the ability to guarantee bandwidth and delay in a separate way, whence the name "decoupled". This paper aims at the characterization of D-CBQ by means of an extended set of simulations and a real implementation into the ALTQ framework.

“…Except for some minor differences we were able to find all the required basic blocks. Basically, we needed hierarchical schedulers (we used the standard one in Cisco and D-CBQ [12] in ns-2 and FreeBSD), queue managers (WRED in Cisco and RIO 4 [16] in FreeBSD), marker/droppers and meters.…”

Section: On-field Experiencementioning

confidence: 99%

Engineering a quality of service-enabled network: a case study

Proceedings 10th IEEE International Conference on Networks (ICON 2002). Towards Network Superiority (Cat. No.02EX588)

Self Cite

This paper presents a case study conceming a network operator that wants to provide services with a certain degree of quality to its customers. Usen are presented with a set of services with different Quality of Service capabilities and the network is engineered in order to satisfy the requests and to maximize provider's revenues. This paper presents the service model adopted and the technical choices that allowed this network to work, taking in mind scalability issues. Finally, it presents the problems that arose when this network was put in place by using both commercial and experimental routers.