Experiences in implementing an experimental wide-area GMPLS network

Zhu, Xiangfei; Zheng, Xuan; Veeraraghavan, Malathi

doi:10.1109/twc.2007.026906

Cited by 9 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each RSVP message type is characterized by the processing time in Table II (where "Others" includes PathErr, ResvErr, PathTear, etc. ), and their propagation delay depends on the geographical distance between the terminating nodes [30][31][32]. Besides, the time for path computation at the source node and for wavelength selection at the destination node is constant and equal to 0.5 ms [22].…”

Section: Req1(2 Fss) Pcvmentioning

confidence: 99%

Collision-Aware Routing and Spectrum Assignment in GMPLS-Enabled Flexible-Bandwidth Optical Network

Guo

Wang

et al. 2013

J. Opt. Commun. Netw.

View full text Add to dashboard Cite

In a flexible-bandwidth optical network (FBON) with a generalized multiprotocol label switching control plane, the resource reservation protocol with traffic engineering extensions (RSVP-TE) is employed to establish end-to-end bandwidth variable lightpaths. The resource reservation can be blocked due to spectrum contention of different concurrent RSVP instances (called backward blocking), and it is the predominant blocking contribution when traffic load is low and arrival highly dynamic. Also, due to its flexible traffic granularity in FBON, the backward collision would happen more frequently and the collision scenario would be more complicated than that in wavelength-routed networking; thus the need for a collision evaluation and avoiding method. So, a collisionavoiding routing with maximum available spectrum allocation (SA) options and an SA scheme to search the spectrum segment with minimum collision possibility are proposed for backward blocking reduction, as well as an extension to the RSVP-TE. Simulation results show that the proposed algorithm could achieve better blocking performance through significantly reducing the backward reservation collision but with acceptable forward blocking deterioration. At the same time, with the proposed algorithm, different granularity traffic could be served more fairly in FBON.Index Terms-Collision aware; Flexible bandwidth optical network; Routing and spectrum assignment.

show abstract

Section: Req1(2 Fss) Pcvmentioning

confidence: 99%

Collision-Aware Routing and Spectrum Assignment in GMPLS-Enabled Flexible-Bandwidth Optical Network

Guo

Wang

et al. 2013

J. Opt. Commun. Netw.

View full text Add to dashboard Cite

show abstract

“…For example, if circuit durations are 30 minutes, and circuit provisioning, based on our prior studies [24], is on the order of 1 second, it is acceptable to signal a change of circuit to applications and to re-provision the circuit on a new path.…”

Section: Path Switchingmentioning

confidence: 99%

A routing architecture for scheduled dynamic circuit services

Veeraraghavan

Starobinski

2010

Proceedings of the Re-Architecting the Internet Workshop

Self Cite

View full text Add to dashboard Cite

A wide range of applications in the nascent field of cloud computing and in other fields require stable, low-delay widearea connectivity together with capabilities of co-scheduling network and server resources. As a result, both research and commercial providers have recently started to deploy scheduled dynamic circuit services (SDCS) as a complement to their IP-routed and leased-line service offerings. Under such services, corporations, universities, or end-users dynamically place requests for a fixed-rate circuit lasting for a fixed duration, for either earliest possible usage or scheduled usage at some point in the future. The lack of adequate support for inter-domain routing and provisioning represents, however, a major hurdle to the full-scale deployment of SDCS. Indeed, existing protocols, such as BGP, do not cater to future dynamic allocation and scheduling of network resources. In this paper, we introduce a strawman protocol, called scheduled circuit routing protocol (SCRP), geared towards inter-domain routing in SDCS-based architectures. The purpose of this protocol is to report available bandwidth information as a function of time across domains without revealing internal network topologies. We describe tradeoffs associated with the design of such a protocol, more particularly the challenge of ensuring high performance while maintaining low implementation complexity, and present avenues for future research.

show abstract

“…In addition to the above-cited papers on analytical modeling [1], [2], we have published papers reporting experimental work on the CHEETAH testbed [9]. But our question on the optimal per-call circuit rate could not be answered using experimental means because of the lack of a sufficient number of data generating end hosts on this testbed.…”

Section: Related Workmentioning

confidence: 99%

“…We choose C ¼ 10 Gb/s and r ¼ 100 Mb/s. We assume E½T setup ¼ 1 sec, based on measurements reported in [9], and set the file-size Pareto-distribution parameters to 1.0009 and k to 1,000 bytes, based on measurements reported in [15].…”

Section: Determination Of Numerical Values For Design Parametersmentioning

confidence: 99%

“…For the first set, parameters related to the circuit-switched path, capacity on all links in the DC network is known to be 10 Gb/s (OC192). Call-setup delay across different paths can be obtained through measurements as we did on the CHEETAH network [9]. Propagation delays can be obtained using ping between computers at different PoPs.…”

Section: Practical Considerationsmentioning

confidence: 99%

See 1 more Smart Citation

A Hybrid Network Architecture for File Transfers

Fang

Veeraraghavan

2009

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

In prior work, we proposed a hybrid architecture that complements the connectionless Internet with a high-speed, dynamically shared circuit-switched network for file-transfer applications. "Small" files are directed to the connectionless network while for "large" files, a high-speed circuit setup is attempted. If the attempt is unsuccessful (because all circuits are already in use), the application will fall back to using the connectionless path. This paper addresses the question of how to select appropriate values for operational parameters in this hybrid architecture. Specifically, we study the questions of what sized files to direct to the circuitswitched network and how much bandwidth to allocate per file transfer. To answer these questions, we construct a model of this hybrid architecture by combining the Erlang-B call blocking model with TCP delay models. Our model captures a combination of utilization and delay considerations in the selection of minimum file size above which transfers should be directed to the circuit-switched network. The optimal per-call circuit rate and the optimal minimum file size are determined to maximize the benefits of using the circuit network, which is quantified with an average-delay-reduction metric.

show abstract

Experiences in implementing an experimental wide-area GMPLS network

Cited by 9 publications

References 17 publications

Collision-Aware Routing and Spectrum Assignment in GMPLS-Enabled Flexible-Bandwidth Optical Network

Collision-Aware Routing and Spectrum Assignment in GMPLS-Enabled Flexible-Bandwidth Optical Network

A routing architecture for scheduled dynamic circuit services

A Hybrid Network Architecture for File Transfers

Contact Info

Product

Resources

About