Dynamic optical-network architectures and technologies for existing and emerging grid services

Simeonidou, Dimitra; Nejabati, Reza; Zervas, Georgios; Klonidis, Dimitrios; Tzanakaki, Anna; O'Mahony, M.J.

doi:10.1109/jlt.2005.856254

Cited by 95 publications

(53 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are two types of models proposed for scheduling network and grid resources in wavelength switched grids [9]. The first is the overlay model, where applications request resource scheduling from the grid middleware.…”

Section: The Need For a Low Latency Grid Architecturementioning

confidence: 99%

See 1 more Smart Citation

Joint resource and network scheduling with adaptive offset determination for optical burst switched grids

Koseoglu

Karasan

2010

Future Generation Computer Systems

View full text Add to dashboard Cite

a b s t r a c tOptical burst switching (OBS) is a promising technology for optical grids with short-lived and interactive data communication requirements. On the other hand, burst losses are in the nature of the OBS protocol and these losses severely affect the grid job completion times. This paper first proposes a joint grid resource and network provisioning method to avoid congestion in the network in order to minimize grid job completion times. Simulations show that joint provisioning significantly reduces completion times in comparison to other methods that perform network provisioning after grid scheduling. An adaptive extra offset based quality of service (QoS) mechanism is also proposed in order to reduce grid burst losses in case of network congestion. Results show that this adaptive mechanism significantly reduces grid completion times by exploiting the trade-off between decreasing loss probability and increasing delay introduced by the extra offset time.

show abstract

Section: The Need For a Low Latency Grid Architecturementioning

confidence: 99%

“…The grid burst loss probability π l (∆ j ) can be computed by performing fixed-point iterations using (6)- (9). However, to simplify this analysis we assume that the low priority link loss probability, π B l , is independent of ∆ j , and we call it q B l .…”

Section: Analysis Of Optimum Offset Timementioning

confidence: 99%

Joint resource and network scheduling with adaptive offset determination for optical burst switched grids

Koseoglu

Karasan

2010

Future Generation Computer Systems

View full text Add to dashboard Cite

show abstract

“…3, and the input parameters are summarized in Table 1. The first used a fairly densely meshed European backbone network (the "Large Topology" taken from [22]), with artificially generated job arrival rates at each site (each rate λ i was with 30% chance uniformly chosen in [1,15] and 70% from [30,60]). …”

Section: Scenariomentioning

confidence: 99%

“…This bound was obtained by assuming that jobs only may be executed at the closest server site (see the simplifying assumption used in the server site selection approach of Section 3.1). Given this assumption, the maximal local processing rate at server site k equals 1-L k , with L k = L calculated using ErlangB formula (1) for N = n k the number of servers at site k, and λ = λ k the aggregate arrival rate of its closest Grid sites.…”

Section: Local Processing Ratesmentioning

confidence: 99%

On dimensioning optical grids and the impact of scheduling

et al. 2008

View full text Add to dashboard Cite

Abstract-When deploying Grid infrastructure, the problem of dimensioning arises: how many servers to provide, where to place them, and which network to install for interconnecting server sites and users generating Grid jobs? In contrast to classical optical network design problems, it is typical of optical Grids that the destination of traffic (jobs) is not known beforehand. This leads to so-called anycast routing of jobs. For network dimensioning, this implies the absence of a clearly defined (source,destination)-based traffic matrix, since only the origin of Grid jobs (and their data) is known, but not their destination. The latter depends not only on the state of Grid resources, including network, storage, and computational resources, but also the Grid scheduling algorithm used. We present a phased solution approach to dimension all these resources, and use it to evaluate various scheduling algorithms in two European network case studies. Results show that the Grid scheduling algorithm has a substantial impact on the required network capacity. This capacity can be minimized by appropriately choosing a (reasonably small) number of server site locations: an optimal balance can be found, in between the single server site case requiring a lot of network traffic to this single location, and an overly fragmented distribution of server capacity over too many sites without much statistical multiplexing opportunities, and hence a relatively large probability of not finding free servers at nearby sites.

show abstract

“…Hence, service differentiation must be provided for such applications in order to reduce the loss rate while maintaining the lowest possible end-to-end delay. Accordingly, the high burst loss probability evident in OBS networks has become a critical issue that must be addressed in order to enable real deployment of OBS networks [5][6][7]. Most existing research in this area can be categorized into one of the mechanism shown in Fig.…”

Section: Introductionmentioning

confidence: 99%

A Novel QoS Provisioning Scheme for OBS Networks

Askar

Zervas

Hunter

et al. 2012

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

Abstract-This paper presents Classified Cloning, a novel QoS provisioning mechanism for OBS networks carrying real-time applications (such as video on demand, Voice over IP, online gaming and Grid computing). It provides such applications with a minimum loss rate while minimizing end-to-end delay and jitter. ns-2 has been used as the simulation tool, with new OBS modules having been developed for performance evaluation purposes. Ingress node performance has been investigated, as well as the overall performance of the suggested scheme. The results obtained showed that new scheme has superior performance to classical cloning. In particular, QoS provisioning offers a guaranteed burst loss rate, delay and expected value of jitter, unlike existing proposals for QoS implementation in OBS which use the burst offset time to provide such differentiation. Indeed, classical schemes increase both end-to-end delay and jitter. It is shown that the burst loss rate is reduced by 50% reduced over classical cloning.

show abstract

Dynamic optical-network architectures and technologies for existing and emerging grid services

Cited by 95 publications

References 20 publications

Joint resource and network scheduling with adaptive offset determination for optical burst switched grids

Joint resource and network scheduling with adaptive offset determination for optical burst switched grids

On dimensioning optical grids and the impact of scheduling

A Novel QoS Provisioning Scheme for OBS Networks

Contact Info

Product

Resources

About