A comparison of two buffer insertion ring architectures with fairness algorithms

Francisco, M.J.; Yuan, Fengjie; Huang, Changcheng; Peng, Hong

doi:10.1109/icc.2003.1204247

Cited by 11 publications

(12 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3 shows the RPR network topology with Hub scenario in our simulation. Four nodes (6,7,8,9) send packets on the outer ring to the terminal node 0, the Hub. Observations are made in the Hub.…”

Section: Simulation Results and Analysismentioning

confidence: 99%

See 1 more Smart Citation

A new scheduling scheme for resilient packet ring networks with single transit buffer

Zhu

Matrawy

Lambadaris

IEEE Global Telecommunications Conference Workshops, 2004. GlobeCom Workshops 2004.

View full text Add to dashboard Cite

-Resilient Packet Ring (RPR) technology known as IEEE 802.17 standard is an emerging network architecture and technology designed to meet the requirements of a packet-based metropolitan area network. The medium access control architecture of RPR network can have single transit buffer (STB), or dual transit buffer (DTB). STB has lower implementation complexity. But the weakness of single transit buffer architecture is that high priority traffic trying to access the ring will be delayed by any low priority traffic already transiting on the node. To improve the quality of service for high priority traffic transmission, we propose a new RPR scheduling scheme which uses Deficit Round-Robin (DRR) algorithm to alternately select packets from the single transit buffer and the high priority transmit buffer. If there is no packet in these two buffers, low priority transmit buffer is then served. Simulation results show certain improvement on overall delay and delay jitter performance of RPR networks with single transit buffer by using our scheme.

show abstract

Section: Simulation Results and Analysismentioning

confidence: 99%

“…RPR has two ring architectures: Single Transit Buffer (STB), and Dual Transit Buffer (DTB) [1] [2][3] [7]. Fig.…”

Section: Introductionmentioning

confidence: 99%

A new scheduling scheme for resilient packet ring networks with single transit buffer

Zhu

Matrawy

Lambadaris

IEEE Global Telecommunications Conference Workshops, 2004. GlobeCom Workshops 2004.

View full text Add to dashboard Cite

show abstract

“…This node, the tail node, resets the advertised rate to the full link rate (if it is not itself more congested), to allow its upstream nodes to send as much traffic as they need. This tail node behavior can produce permanent oscillations in certain scenarios which degrades the average throughput ( [3], [4], [7], [6]). We study these two scenarios in this paper, shown in Figure 3 To prevent these oscillations under unbalanced traffic scenarios, several proposals were made: In [7], a dynamic bandwidth allocation algorithm called Distributed Virtual-time Scheduling in Rings (DVSR) is proposed.…”

Section: Fairness Algorithmmentioning

confidence: 99%

Improved fairness algorithm to prevent tail node induced oscillations in RPR

Robichaud

Huang

IEEE International Conference on Communications, 2005. ICC 2005. 2005

View full text Add to dashboard Cite

-The IEEE P802.17 workgroup has standardized a ring network architecture and associated protocol called Resilient Packet Ring (RPR). RPR tries to overcome previous MAN technologies shortcomings in high-speed networks. The RPR fairness mechanism suffers from severe permanent oscillations under certain conditions. Several propositions were made to solve this problem but all needed significant modifications of the current RPR protocol. In this paper, we propose an improved algorithm to prevent tail node induced oscillations. It requires only simple modifications to the current fair rate advertisement mechanism of RPR. Our simulation results show that the improvements work under various conditions and increase the average throughput of RPR.

show abstract

“…The first is bandwidth oscillation due to the inaccuracy in fair rate estimation. The results in [4], [5] have shown that with "unbalanced node traffic" scenario, RPR fairness algorithm is prone to severe and permanent oscillations. Such oscillations may result in significant throughput loss [6].…”

Section: A Limitations Of Rpr Fairness Algorithmmentioning

confidence: 99%

“…Such oscillations may result in significant throughput loss [6]. The second is long access delay for the node traffic since there is a delay from the time the fair rate is calculated to the time the fair rate is taken into effect at all nodes [4]. This is a natural result of feedback control scheme.…”

Section: A Limitations Of Rpr Fairness Algorithmmentioning

confidence: 99%