Adaptive Protection of Scientific Backbone Networks Using Machine Learning

Mogyorosi, Ferenc; Pasic, Alija; Cziva, Richard; Revisnyei, Peter; Kenesi, Zsolt; Tapolcai, Janos

doi:10.1109/TNSM.2021.3050964

Title: Adaptive Protection of Scientific Backbone Networks Using Machine Learning

Journal Article · Mon Jan 11 00:00:00 EST 2021 · IEEE Transactions on Network and Service Management

DOI:https://doi.org/10.1109/TNSM.2021.3050964· OSTI ID:1813357

^[1];

^[2];

^[1]; Kenesi, Zsolt ^[3];

^[1]

Budapest Univ. of Technology and Economics (BME) (Hungary)
Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Ericsson Hungary Ltd., Budapest (Hungary)

In this article, we propose a new protection scheme for backbone networks to guarantee high service availability. The presented scheme does not require any reconfiguration immediately after the failure (i.e., it is proactive). At the same time, it does not require any reserved backup network resources either. To achieve these seemingly contradictory goals, we utilize the recent advancements in Machine Learning (ML) to implement a network intelligence that periodically re-allocates the unused capacity as protection bandwidth to meet the service availability requirements of each connection. Our goal is achieved by two components (1) predicting the traffic for the next period on each link, and (2) intelligently selecting the best fit dedicated protection scheme for the next period depending on the estimated unused (spare) bandwidth and the previous service availability violations. Note that re-allocating protection bandwidth affects neither the operational connections nor the current best practice of operators to over-provision network bandwidth to support elephant flows. Finally, we provide a case study on the real traffic from Energy Sciences Network (ESnet), a high-speed, international scientific backbone network. The key benefit of our framework is that adaptively utilizing the over-provisioned bandwidth for spare capacity is sufficient to improve the availability from three-nines to five-nines (in ESnet for the 30 examined connections). The drawback is negligible bandwidth limitations; the user perceives a minor and very temporal bandwidth limitation in less than 0.1% of the time.

View Accepted Manuscript (DOE)

Cite

Export

Save

Research Organization:: Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)

Sponsoring Organization:: USDOE Office of Science (SC); National Science Foundation (NSF)

Grant/Contract Number:: AC02-05CH11231; 2018754

OSTI ID:: 1813357

Journal Information:: IEEE Transactions on Network and Service Management, Vol. 18, Issue 1; ISSN 1932-4537

Publisher:: IEEECopyright Statement

Country of Publication:: United States

Language:: English

References (17)

Quality of Service (QoS) in Software Defined Networking (SDN): A survey Karakus, Murat; Durresi, Arjan Journal of Network and Computer Applications, Vol. 80 https://doi.org/10.1016/j.jnca.2016.12.019	journal	February 2017
The Globus Striped GridFTP Framework and Server Allcock, W.; Bresnahan, J.; Kettimuthu, R. ACM/IEEE SC 2005 Conference (SC'05) https://doi.org/10.1109/SC.2005.72	conference	January 2005
Instantaneous recovery of unicast connections in transport networks: Routing versus coding Babarczi, Péter; Pašić, Alija; Tapolcai, János Computer Networks, Vol. 82 https://doi.org/10.1016/j.comnet.2015.02.010	journal	May 2015
How to Measure Network Flexibility? A Proposal for Evaluating Softwarized Networks Kellerer, Wolfgang; Basta, Arsany; Babarczi, Peter IEEE Communications Magazine, Vol. 56, Issue 10 https://doi.org/10.1109/MCOM.2018.1700601	journal	October 2018
Analyzing the Performance of an Anycast CDN Calder, Matt; Flavel, Ashley; Katz-Bassett, Ethan Proceedings of the 2015 Internet Measurement Conference https://doi.org/10.1145/2815675.2815717	conference	October 2015
Applying deep learning approaches for network traffic prediction Vinayakumar, R.; Soman, K. P.; Poornachandran, Prabaharan 2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI) https://doi.org/10.1109/ICACCI.2017.8126198	conference	September 2017
Enhanced Echo-State Restricted Boltzmann Machines for Network Traffic Prediction Sun, Xiaochuan; Ma, Shuhao; Li, Yingqi IEEE Internet of Things Journal, Vol. 7, Issue 2 https://doi.org/10.1109/JIOT.2019.2954283	journal	February 2020
Multilayer networks: an architecture framework Lehman, Tom; Yang, Xi; Ghani, Nasir IEEE Communications Magazine, Vol. 49, Issue 5 https://doi.org/10.1109/MCOM.2011.5762808	journal	May 2011
Long short-term memory neural network for network traffic prediction Zhuo, Qinzheng; Li, Qianmu; Yan, Han 2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE) https://doi.org/10.1109/ISKE.2017.8258815	conference	November 2017
Quality of service terminology in IP networks Gozdecki, J.; Jajszczyk, A.; Stankiewicz, R. IEEE Communications Magazine, Vol. 41, Issue 3 https://doi.org/10.1109/MCOM.2003.1186560	journal	March 2003
Long Short-Term Memory Hochreiter, Sepp; Schmidhuber, Jürgen Neural Computation, Vol. 9, Issue 8 https://doi.org/10.1162/neco.1997.9.8.1735	journal	November 1997
T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction Zhao, Ling; Song, Yujiao; Zhang, Chao IEEE Transactions on Intelligent Transportation Systems, Vol. 21, Issue 9 https://doi.org/10.1109/TITS.2019.2935152	journal	September 2020
Variable heavy tails in Internet traffic Hernández-Campos, Félix; Marron, J. S.; Samorodnitsky, Gennady Performance Evaluation, Vol. 58, Issue 2-3 https://doi.org/10.1016/j.peva.2004.07.008	journal	November 2004
Evaluation of network traffic prediction based on neural networks with multi-task learning and multiresolution decomposition Barabas, Melinda; Boanea, Georgeta; Rus, Andrei B. 2011 IEEE 7th International Conference on Intelligent Computer Communication and Processing https://doi.org/10.1109/ICCP.2011.6047849	conference	August 2011
Understanding Internet traffic streams: dragonflies and tortoises Brownlee, N.; Claffy, K. C. IEEE Communications Magazine, Vol. 40, Issue 10 https://doi.org/10.1109/MCOM.2002.1039865	journal	October 2002
Identifying elephant flows through periodically sampled packets Mori, Tatsuya; Uchida, Masato; Kawahara, Ryoichi Proceedings of the 4th ACM SIGCOMM conference on Internet measurement - IMC '04 https://doi.org/10.1145/1028788.1028803	conference	January 2004
Service level agreement and provisioning in optical networks Fawaz, W.; Daheb, B.; Audouin, O. IEEE Communications Magazine, Vol. 42, Issue 1 https://doi.org/10.1109/MCOM.2004.1262160	journal	January 2004

Similar Records

Flow Ordering and Hierarchical Bottleneck Identification for High Speed Data Networks - Phase I Final Report

Technical Report · Tue Dec 03 00:00:00 EST 2019 · OSTI ID:1813357

Ros-Giralt, Jordi

Bridging the gap between peak and average loads on science networks

Journal Article · Fri May 12 00:00:00 EDT 2017 · Future Generations Computer Systems · OSTI ID:1813357

Nickolay, Sam; Jung, Eun -Sung; Kettimuthu, Rajkumar; +1 more

HEP Science Network Requirements--Final Report

Technical Report · Tue Apr 27 00:00:00 EDT 2010 · OSTI ID:1813357

Bakken, Jon; Barczyk, Artur; Blatecky, Alan; +26 more

Related Subjects

42 ENGINEERING
quality of service (QoS)
availability
traffic prediction
energy sciences network
service level agreement (SLA)
deep learning

Title: Adaptive Protection of Scientific Backbone Networks Using Machine Learning

Citation Formats

References (17)

Similar Records

Related Subjects