Programmable residues defined networks for edge data centres

Magnos Martinello; Alextian Liberato; Arash Farhadi Beldachi; Koteswararao Kondepu; Roberta Gomes; Rodolfo Villaca; Moises Ribeiro; Emilio Hugues Salas; Dimitra Simeonidou

doi:10.23919/CNSM.2017.8255987

Programmable residues defined networks for edge data centres

Magnos Martinello, Alextian Liberato, Arash Farhadi Beldachi, Koteswararao Kondepu, Roberta Gomes, Rodolfo Villaca, Moises Ribeiro, Emilio Hugues Salas, Dimitra Simeonidou

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

4 Citations (Scopus)

448 Downloads (Pure)

Abstract

Edge Data Centres (EDC) are often managed by a single administrative entity with logically centralized control. The architectural split of control and data planes and the new control plane abstractions have been touted as Software-Defined Networking (SDN), where the OpenFlow protocol is one common choice for the standardized programmatic interface to data plane devices. However, in the design of an SDN architecture, there is no clear distinction between functional network parts such as core and edge elements. It means that all switches require to support lookups over hundreds of bits with complex actions that have to be specified by multiple tables. In this paper, we propose a new programmable architecture for EDC networks, named Residues Defined Networks (RDN). In RDN, a controller defines a network policy (e.g. connectivity protection) setting flow entries at the edges. Based on these entries, the edge switches assign route-IDs to flows. A route is defined as the remainder of the division (Residue) between a route-ID and a set of switch-IDs within RDN core. In case of failures, emergency routes are compactly encoded as programmable residues forwarding paths written into the packets. RDN scalability is evaluated considering 2-tier Clos topologies which cover mostly EDC deployments supporting up to 2304 servers. A RDN proof-of-concept prototype is implemented in Mininet for network emulation. Also, to increase the accuracy on latency measures, we implement RDN in NetFPGA that is validated in a testbed with 10Gbps Ethernet boards. RDN offers ultra-fast failure recovery (sub-milliseconds carrier grade), achieves low latency with RDN switching time per hop ( 0.6 microseconds) and no jitter within the RDN core.

Original language	English
Title of host publication	2017 13th International Conference on Network and Service Management (CNSM 2017)
Subtitle of host publication	Proceedings of a meeting held 26-30 November 2017, Tokyo, Japan
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	117-125
Number of pages	9
ISBN (Electronic)	9783901882982
ISBN (Print)	9781538621530
DOIs	https://doi.org/10.23919/CNSM.2017.8255987
Publication status	Published - Feb 2018
Event	IEEE Conference on Network and Service Management - Tokyo Duration: 27 Nov 2017 → 30 Nov 2017 http://www.cnsm-conf.org/2017

Publication series

Name
ISSN (Print)	2165-963X

Conference

Conference	IEEE Conference on Network and Service Management
Abbreviated title	CNSM
Period	27/11/17 → 30/11/17
Internet address	http://www.cnsm-conf.org/2017

Access to Document

10.23919/CNSM.2017.8255987

Full-text PDF (accepted author manuscript)
This is the author accepted manuscript (AAM). The final published version (version of record) is available online via IEEE at http://ieeexplore.ieee.org/document/8255987/ . Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 3.18 MB

Handle.net

Persistent link

Cite this

Martinello, M., Liberato, A., Farhadi Beldachi, A., Kondepu, K., Gomes, R., Villaca, R., Ribeiro, M., Hugues Salas, E., & Simeonidou, D. (2018). Programmable residues defined networks for edge data centres. In 2017 13th International Conference on Network and Service Management (CNSM 2017) : Proceedings of a meeting held 26-30 November 2017, Tokyo, Japan (pp. 117-125). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.23919/CNSM.2017.8255987

@inproceedings{d2f312fc031c4a84864c5cb676c001e6,

title = "Programmable residues defined networks for edge data centres",

abstract = "Edge Data Centres (EDC) are often managed by a single administrative entity with logically centralized control. The architectural split of control and data planes and the new control plane abstractions have been touted as Software-Defined Networking (SDN), where the OpenFlow protocol is one common choice for the standardized programmatic interface to data plane devices. However, in the design of an SDN architecture, there is no clear distinction between functional network parts such as core and edge elements. It means that all switches require to support lookups over hundreds of bits with complex actions that have to be specified by multiple tables. In this paper, we propose a new programmable architecture for EDC networks, named Residues Defined Networks (RDN). In RDN, a controller defines a network policy (e.g. connectivity protection) setting flow entries at the edges. Based on these entries, the edge switches assign route-IDs to flows. A route is defined as the remainder of the division (Residue) between a route-ID and a set of switch-IDs within RDN core. In case of failures, emergency routes are compactly encoded as programmable residues forwarding paths written into the packets. RDN scalability is evaluated considering 2-tier Clos topologies which cover mostly EDC deployments supporting up to 2304 servers. A RDN proof-of-concept prototype is implemented in Mininet for network emulation. Also, to increase the accuracy on latency measures, we implement RDN in NetFPGA that is validated in a testbed with 10Gbps Ethernet boards. RDN offers ultra-fast failure recovery (sub-milliseconds carrier grade), achieves low latency with RDN switching time per hop ( 0.6 microseconds) and no jitter within the RDN core. ",

author = "Magnos Martinello and Alextian Liberato and {Farhadi Beldachi}, Arash and Koteswararao Kondepu and Roberta Gomes and Rodolfo Villaca and Moises Ribeiro and {Hugues Salas}, Emilio and Dimitra Simeonidou",

year = "2018",

month = feb,

doi = "10.23919/CNSM.2017.8255987",

language = "English",

isbn = "9781538621530",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

pages = "117--125",

booktitle = "2017 13th International Conference on Network and Service Management (CNSM 2017)",

address = "United States",

note = "IEEE Conference on Network and Service Management, CNSM ; Conference date: 27-11-2017 Through 30-11-2017",

url = "http://www.cnsm-conf.org/2017",

}

Martinello, M, Liberato, A, Farhadi Beldachi, A, Kondepu, K, Gomes, R, Villaca, R, Ribeiro, M, Hugues Salas, E & Simeonidou, D 2018, Programmable residues defined networks for edge data centres. in 2017 13th International Conference on Network and Service Management (CNSM 2017) : Proceedings of a meeting held 26-30 November 2017, Tokyo, Japan. Institute of Electrical and Electronics Engineers (IEEE), pp. 117-125, IEEE Conference on Network and Service Management, 27/11/17. https://doi.org/10.23919/CNSM.2017.8255987

Programmable residues defined networks for edge data centres. / Martinello, Magnos; Liberato, Alextian; Farhadi Beldachi, Arash et al.
2017 13th International Conference on Network and Service Management (CNSM 2017) : Proceedings of a meeting held 26-30 November 2017, Tokyo, Japan. Institute of Electrical and Electronics Engineers (IEEE), 2018. p. 117-125.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Programmable residues defined networks for edge data centres

AU - Martinello, Magnos

AU - Liberato, Alextian

AU - Farhadi Beldachi, Arash

AU - Kondepu, Koteswararao

AU - Gomes, Roberta

AU - Villaca, Rodolfo

AU - Ribeiro, Moises

AU - Hugues Salas, Emilio

AU - Simeonidou, Dimitra

PY - 2018/2

Y1 - 2018/2

N2 - Edge Data Centres (EDC) are often managed by a single administrative entity with logically centralized control. The architectural split of control and data planes and the new control plane abstractions have been touted as Software-Defined Networking (SDN), where the OpenFlow protocol is one common choice for the standardized programmatic interface to data plane devices. However, in the design of an SDN architecture, there is no clear distinction between functional network parts such as core and edge elements. It means that all switches require to support lookups over hundreds of bits with complex actions that have to be specified by multiple tables. In this paper, we propose a new programmable architecture for EDC networks, named Residues Defined Networks (RDN). In RDN, a controller defines a network policy (e.g. connectivity protection) setting flow entries at the edges. Based on these entries, the edge switches assign route-IDs to flows. A route is defined as the remainder of the division (Residue) between a route-ID and a set of switch-IDs within RDN core. In case of failures, emergency routes are compactly encoded as programmable residues forwarding paths written into the packets. RDN scalability is evaluated considering 2-tier Clos topologies which cover mostly EDC deployments supporting up to 2304 servers. A RDN proof-of-concept prototype is implemented in Mininet for network emulation. Also, to increase the accuracy on latency measures, we implement RDN in NetFPGA that is validated in a testbed with 10Gbps Ethernet boards. RDN offers ultra-fast failure recovery (sub-milliseconds carrier grade), achieves low latency with RDN switching time per hop ( 0.6 microseconds) and no jitter within the RDN core.

AB - Edge Data Centres (EDC) are often managed by a single administrative entity with logically centralized control. The architectural split of control and data planes and the new control plane abstractions have been touted as Software-Defined Networking (SDN), where the OpenFlow protocol is one common choice for the standardized programmatic interface to data plane devices. However, in the design of an SDN architecture, there is no clear distinction between functional network parts such as core and edge elements. It means that all switches require to support lookups over hundreds of bits with complex actions that have to be specified by multiple tables. In this paper, we propose a new programmable architecture for EDC networks, named Residues Defined Networks (RDN). In RDN, a controller defines a network policy (e.g. connectivity protection) setting flow entries at the edges. Based on these entries, the edge switches assign route-IDs to flows. A route is defined as the remainder of the division (Residue) between a route-ID and a set of switch-IDs within RDN core. In case of failures, emergency routes are compactly encoded as programmable residues forwarding paths written into the packets. RDN scalability is evaluated considering 2-tier Clos topologies which cover mostly EDC deployments supporting up to 2304 servers. A RDN proof-of-concept prototype is implemented in Mininet for network emulation. Also, to increase the accuracy on latency measures, we implement RDN in NetFPGA that is validated in a testbed with 10Gbps Ethernet boards. RDN offers ultra-fast failure recovery (sub-milliseconds carrier grade), achieves low latency with RDN switching time per hop ( 0.6 microseconds) and no jitter within the RDN core.

U2 - 10.23919/CNSM.2017.8255987

DO - 10.23919/CNSM.2017.8255987

M3 - Conference Contribution (Conference Proceeding)

SN - 9781538621530

SP - 117

EP - 125

BT - 2017 13th International Conference on Network and Service Management (CNSM 2017)

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - IEEE Conference on Network and Service Management

Y2 - 27 November 2017 through 30 November 2017

ER -

Martinello M, Liberato A, Farhadi Beldachi A, Kondepu K, Gomes R, Villaca R et al. Programmable residues defined networks for edge data centres. In 2017 13th International Conference on Network and Service Management (CNSM 2017) : Proceedings of a meeting held 26-30 November 2017, Tokyo, Japan. Institute of Electrical and Electronics Engineers (IEEE). 2018. p. 117-125 Epub 2018 Jan 15. doi: 10.23919/CNSM.2017.8255987

Programmable residues defined networks for edge data centres

Abstract

Publication series

Conference

Access to Document

Handle.net

Fingerprint

Cite this