Dynamic Core VNT Adaptability Based on Predictive Metro-Flow Traffic Models

F. Morales; Ll. Gifre; F. Paolucci; M. Ruiz; F. Cugini; P. Castoldi; L. Velasco

doi:10.1364/JOCN.9.001202

Journal of Optical Communications and Networking
Vol. 9,
Issue 12,
pp. 1202-1211
(2017)
•https://doi.org/10.1364/JOCN.9.001202

Dynamic Core VNT Adaptability Based on Predictive Metro-Flow Traffic Models

F. Morales, Ll. Gifre, F. Paolucci, M. Ruiz, F. Cugini, P. Castoldi, and L. Velasco

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
F. Morales, Ll. Gifre, F. Paolucci, M. Ruiz, F. Cugini, P. Castoldi, and L. Velasco, "Dynamic Core VNT Adaptability Based on Predictive Metro-Flow Traffic Models," J. Opt. Commun. Netw. 9, 1202-1211 (2017)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Abstract

MPLS-over-optical virtual network topologies (VNTs) can be adapted to near-future traffic matrices based on predictive models that are estimated by applying data analytics on monitored origin-destination (OD) traffic. However, the deployment of independent SDN controllers for core and metro segments can bring large inefficiencies to this core network reconfiguration based on traffic prediction when traffic flows from metro areas are rerouted to different ingress nodes in the core. In such cases, OD traffic patterns in the core might severely change, thus affecting the quality of the predictive OD models. New traffic model re-estimation usually takes a long time, during which no predictive capabilities are available for the network operator. To alleviate this problem, we propose to extend data analytics to metro networks to obtain predictive models for the metro flows; by knowing how these flows are aggregated into OD pairs in the core, we can also aggregate their predictive models, thus accurately predicting OD traffic and therefore enabling core VNT reconfiguration. To obtain quality metro-flow models, we propose an estimation algorithm that processes monitored data and returns a predictive model. In addition, a flow controller is proposed for the control architecture to allow metro and core controllers to exchange metro-flow model information. The proposed model aggregation is evaluated through exhaustive simulation, and eventually experimentally assessed together with the flow controller in a testbed connecting premises in CNIT (Pisa, Italy) and UPC (Barcelona, Spain).

Full Article | PDF Article

More Like This

Virtual Network Topology Adaptability Based on Data Analytics for Traffic Prediction

Fernando Morales, Marc Ruiz, Lluís Gifre, Luis M. Contreras, Víctor López, and Luis Velasco
J. Opt. Commun. Netw. 9(1) A35-A45 (2017)

Matheuristic With Machine-Learning-Based Prediction for Software-Defined Mobile Metro-Core Networks

Rodolfo Alvizu, Sebastian Troia, Guido Maier, and Achille Pattavina
J. Opt. Commun. Netw. 9(9) D19-D30 (2017)

Knowledge management in optical networks: architecture, methods, and use cases [Invited]

Marc Ruiz, Fatemehsadat Tabatabaeimehr, and Luis Velasco
J. Opt. Commun. Netw. 12(1) A70-A81 (2020)

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (11)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (3)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (8)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

INPUT $Y$ , $T$
OUTPUT model
1: $⟨ perStart, perDur ⟩ \leftarrow identifyPeriod$ ( $Y$ )
2: $X \leftarrow []$ ; $nSegm \leftarrow ⌈ perDur / T ⌉$
3: for $y = ⟨ t i m e, bitrate ⟩$ in $Y$ do
4: $t \leftarrow ⌊ ((y . t i m e - perStart) % perDur) / T ⌋$
5: push( $y . bitrate, X {t}$ )
6: $U \leftarrow []$ ; $V \leftarrow []$ ; $μ \leftarrow \emptyset$ ; $σ^{2} \leftarrow \emptyset$
7: $segmLength \leftarrow perDur / nSegm$
8: for $t$ in $0 \dots nSegm$ do
9: if $t < nSegm$ then
10: $U [t] \leftarrow compute$ mean $u (X {t}$ ) (Eq. (1))
11: $V [t] \leftarrow compute$ variance $v (X {t}$ ) (Eq. (2))
12: if $t = 0$ continue
13: if $t < nSegm$ then $t^{'} \leftarrow t - 1$ else $t \leftarrow 0$ ; $t^{'} \leftarrow nSegm - 1$
14: $a_{μ} \leftarrow U [t]$ ; $a_{σ 2} \leftarrow V [t]$
15: $b_{μ} \leftarrow compute$ slope $b (U [t^{'}], U [t], segmLength)$ (Eq. (3))
16: $b_{σ 2} \leftarrow compute$ slope $b (V [t^{'}], V [t], segmLength)$ (Eq. (3))
17: $μ [t] \leftarrow ⟨ a_{μ}, b_{μ} ⟩$ ; $σ^{2} [t] \leftarrow ⟨ a_{σ 2}, b_{σ 2} ⟩$
18: return $model = ⟨ perStart, perDur, nSegm, μ, σ^{2} ⟩$

INPUT model, time
OUTPUT $⟨ μ, σ^{2} ⟩$
1: $segmLength \leftarrow model.perDur / model.nSegm$
2: $t \leftarrow (t i m e - model.perStart) % model.perDur$
3: $s \leftarrow ⌊ t / segmLength ⌋$
4: $offset \leftarrow t % segmLength$
5: $[a_{μ}, b_{μ}] \leftarrow model . μ [s]$
6: $[a_{σ 2}, b_{σ 2}] \leftarrow model . σ^{2} [s]$
7: $μ \leftarrow a_{μ}, + b_{μ} * offset$
8: $σ^{2} \leftarrow a_{σ 2} + b_{σ 2} * offset$
9: return $⟨ μ, σ^{2} ⟩$

INPUT: $Q = {⟨ o d, m, F (o d) ⟩}$
OUTPUT: $S = {⟨ o d, m^{'} ⟩}$
1: $Q_{obs} \leftarrow getObsoleteModels (Q)$
2: $S \leftarrow \emptyset$
3: for each $q = ⟨ o d, m, F (o d) ⟩$ in $Q_{obs}$ do
4: if $type (m) = NEW_CORE$ then
5: $m^{'} \leftarrow m$
6: else if $type (m) = NEW_METRO$ then
7: $m^{'} \leftarrow newAggregate (F (o d))$ (Eqs. (4),(5))
8: else if $t y p e (m) = UPDATE$ then
9: $m^{'} \leftarrow updateAggregate (m, F (o d))$ (Eqs. (6),(7))
10: $S \leftarrow S \cup {⟨ o d, m^{'} ⟩}$
11: return $S$

INPUT $Y$ , $T$
OUTPUT model
1: $⟨ perStart, perDur ⟩ \leftarrow identifyPeriod$ ( $Y$ )
2: $X \leftarrow []$ ; $nSegm \leftarrow ⌈ perDur / T ⌉$
3: for $y = ⟨ t i m e, bitrate ⟩$ in $Y$ do
4: $t \leftarrow ⌊ ((y . t i m e - perStart) % perDur) / T ⌋$
5: push( $y . bitrate, X {t}$ )
6: $U \leftarrow []$ ; $V \leftarrow []$ ; $μ \leftarrow \emptyset$ ; $σ^{2} \leftarrow \emptyset$
7: $segmLength \leftarrow perDur / nSegm$
8: for $t$ in $0 \dots nSegm$ do
9: if $t < nSegm$ then
10: $U [t] \leftarrow compute$ mean $u (X {t}$ ) (Eq. (1))
11: $V [t] \leftarrow compute$ variance $v (X {t}$ ) (Eq. (2))
12: if $t = 0$ continue
13: if $t < nSegm$ then $t^{'} \leftarrow t - 1$ else $t \leftarrow 0$ ; $t^{'} \leftarrow nSegm - 1$
14: $a_{μ} \leftarrow U [t]$ ; $a_{σ 2} \leftarrow V [t]$
15: $b_{μ} \leftarrow compute$ slope $b (U [t^{'}], U [t], segmLength)$ (Eq. (3))
16: $b_{σ 2} \leftarrow compute$ slope $b (V [t^{'}], V [t], segmLength)$ (Eq. (3))
17: $μ [t] \leftarrow ⟨ a_{μ}, b_{μ} ⟩$ ; $σ^{2} [t] \leftarrow ⟨ a_{σ 2}, b_{σ 2} ⟩$
18: return $model = ⟨ perStart, perDur, nSegm, μ, σ^{2} ⟩$

Abstract

Cited By

Figures (11)

Tables (3)

Equations (8)

Journal of Optical Communications and Networking