SLA-Aware Dynamic Bandwidth Allocation for QoS in EPONs

Choi, Sangdun; Park, Jae-Hyung

doi:10.1364/jocn.2.000773

Cited by 30 publications

(12 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [12], the proposal introduces an approach called cyclicpolling-based dynamic bandwidth allocation with service level agreements (CPBA-SLA) based on a two-layer allocation scheme: SLA-layer and ONU-layer. In the SLA-layer allocation, ONUs are distributed on two or more groups based on their SLA parameters.…”

Section: Upstream Bandwidth Alloaction Mechanismmentioning

confidence: 99%

Dynamic reallocation of SLA parameters in passive optical network based on clustering analysis

Frigui

Lemlouma

Gosselin

et al. 2018

2018 21st Conference on Innovation in Clouds, Internet and Networks and Workshops (ICIN)

View full text Add to dashboard Cite

Section: Upstream Bandwidth Alloaction Mechanismmentioning

confidence: 99%

Dynamic reallocation of SLA parameters in passive optical network based on clustering analysis

Frigui

Lemlouma

Gosselin

et al. 2018

2018 21st Conference on Innovation in Clouds, Internet and Networks and Workshops (ICIN)

View full text Add to dashboard Cite

“…Thus, the delay of various traffic classes should be load independent. The algorithms must impart a centralized delay distribution to the traffic classes, where the delay for a complete range of load is identically distributed [6]. Also, as the delay and the jitter needed is very much dependent on the application of the user, the users must be served according to the QoS parameter needed which is referred as the Parameterized QoS control (PQoS) [7].…”

Section: Keeping Required Delay Variation (Jitter) Bound -mentioning

confidence: 99%

“…Online approaches such as IPACT achieve a high throughput but have a variable cycle time. The variable cycle time leads to variable packet delays or high jitter and is not suitable for the high-priority jitter-sensitive applications [6,12].…”

Section: A Online Vs Offline Qos Algorithmsmentioning

confidence: 99%

“…This changes the service order of the ONUs in every cycle, thereby impairing the estimation of the high-priority traffic. Furthermore, to provide a constant jitter performance, several offline scheduling algorithms with either separate cycle time for each traffic class (such as Hybrid Grant Protocol (HGP) protocol [12] Cyclic-Polling-based Bandwidth Allocation with SLAs (CPBA-SLA) [6]) have been proposed. In the HGP protocol, there is a separate cycle for each traffic class and there are idle periods between each cycle, which limits the throughput.…”

Section: A Online Vs Offline Qos Algorithmsmentioning

confidence: 99%

“…Figure 11 shows the probability of the delay of EF traffic for half and full load scenarios for DAWS, MDAWS and CPBA. CPBA is a fixed cycle algorithm [6], and is used to benchmark the performance of the MDAWS algorithm. We clearly see that the MDAWS algorithm minimizes the interwindow jitter and has a centralized delay distribution comparable to CPBA, and thus is immune to fluctuations in load.…”

Section: ) Delaymentioning

confidence: 99%

See 2 more Smart Citations

Dynamic Bandwidth Allocation With SLA Awareness for QoS in Ethernet Passive Optical Networks

Dixit¹,

Lannoo²,

Das³

et al. 2013

J. Opt. Commun. Netw.

View full text Add to dashboard Cite

Abstract-Quality-of-service (QoS) support in Ethernet passive optical networks (EPON) is a crucial concern. We propose a new dynamic bandwidth allocation (DBA) algorithm, for service differentiation that meets service level agreements (SLAs) of the users. The proposed delay aware (DA) online DBA algorithm provides constant and predictable average packet delay and reduced delay variation for the high and medium priority traffic while keeping the packet loss rate under check. We prove the effectiveness of the proposed algorithm by exhaustive simulations. EPON is a tree-structured PON technology where in the downstream direction, an optical line terminal (OLT) at the central office (CO) broadcasts data to every optical network unit (ONU) at the user's end, as shown in Fig. 1. In the upstream direction, TDMA techniques are used for scheduling data transmissions from the ONUs to the OLT to avoid any collisions between the users' data. Multi-Point Control Protocol (MPCP) specified in IEEE 802. 3ah is used as the signaling protocol. GATE and REPORT messages are 64 byte Ethernet control messages specified in MPCP. The ONU sends the REPORT messages carrying bandwidth request information based on its queue size, and the OLT sends back the GATE message to the ONU informing allocated bandwidth. Several dynamic bandwidth allocation (DBA) algorithms have been proposed for efficient resource allocation in EPONs [3]. The resource allocation algorithms need to address further many performance dimensions with proliferation of voice and video applications like voice-over-IP (VOIP), IPTV, network gaming, and many others. These emerging applications require a guaranteed bound on many network parameters like: bandwidth, packet delay (latency), packet delay variation (jitter), and packet-loss ratio. This parameter bound delivery of content is referred as Quality of Service (QoS). In this paper, we investigate QoS issues in optical access networks and in particular in EPONs. Index Terms-

show abstract