Abstract
Ring aggregation networks are often employed by network carriers because of their efficiency and high fault tolerance. A fairness scheme is required in ring aggregation to achieve per-flow throughput fairness and bufferbloat avoidance, because frames are forwarded along multiple ring nodes. N Rate N + 1 Color Marking (NRN + 1CM) was proposed to achieve fairness in ring aggregation networks consisting of Layer-2 Switches (SWs). With NRN + 1CM, frames are selectively discarded based on color and the frame-dropping threshold. To avoid the accumulation of a queuing delay, frames are discarded at upstream nodes in advance through the notification process for the frame-dropping threshold. However, in the previous works, NRN + 1CM was assumed to be employed in a logical daisy chain topology linked to one Edge Router (ER). The currently available threshold notification process of NRN + 1CM cannot be employed for ring networks with multiple ERs. Therefore, this paper proposes a method for applying NRN + 1CM to a ring aggregation network with multiple ERs. With the proposed algorithm, an SW dynamically selects the dropping threshold to send in order to avoid excess frame discarding. The performance of the proposed scheme was confirmed through computer simulations.
Highlights
In current networks, bufferbloat is a prevalent problem [1]
The existing NRN + 1CM cannot be employed in such networks because it does not assume multiple Edge Router (ER) and switching paths
We verified the effect of the selected threshold notification on throughput fairness under a simple condition
Summary
Bufferbloat is a prevalent problem [1]. The presence of bufferbloat in 3G/4G cellular networks was confirmed in [2]. A fairness scheme is required in ring aggregation to achieve per-flow throughput fairness and bufferbloat avoidance, because frames are forwarded along multiple SWs. To overcome bufferbloat, the Active Queue Management (AQM) approach has been reconsidered to control the queuing delay, such as Controlled Delay (CoDel) [10]. A Proportional Integral controller-Enhanced (PIE) scheduler [11] was proposed to control the average queuing latency The goal of these AQM schemes is overcoming bufferbloat; not per-flow fairness. To achieve per-flow fairness and bufferbloat avoidance in ring aggregation networks, N Rate N + 1 Color Marking (NRN + 1CM) was proposed [17,18,19]. The performance of NRN + 1CM was confirmed through mathematical formulation and computer simulations; per-flow fairness can be achieved regardless of traffic types, the number of SWs and buffer sizes.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
