SlideShare a Scribd company logo
Control Plane Progress
Tiered Evaluation Strategy
Project : T.2 Testbed for Optical Aggregation Networks (TOAN)
Investigators: Wissinger, Kilper, Bergman, Cvijetic, Willner, Glick, Norwood
Mission Other Progress
Industry Collaborators
• Research: enable CIAN components, devices, and networking concepts to be evaluated
within a realistic network system infrastructure; link research efforts across the Center;
instantiate the systems vision of CIAN.
• Industry: provide a venue for R&D collaboration with industry partners; accelerate transfer
and productization of CIAN and member company technologies.
• Education: support the training of students with multi-disciplinary systems engineering,
hands-on lab work, and project leadership; provide a platform for collaborative projects
between CIAN students and industry.
A significant strength of CIAN is the comprehensive testbed infrastructure that supports its
vertical development chain from the materials level, through to chip scale components,
packaged devices, and integrated subsystems.
 Investigates impairment-aware and service-aware adaptation and reconfiguration of optical
paths and coding/modulation schemes, especially in context of multi-node transmission
management.
 Explores application of software programmable network behaviors and hardware
virtualization to enable intelligent and cost-effective aggregation schemes and to establish
on-demand high BW connections over the WAN.
 Evaluates CIAN technologies (e.g., chip-scale photonic devices, subsystems, control
planes) within network system context, compare to benchmarks, provide feedback to device
designers on deployment and interoperability issues.
 Studies network stability under increasingly dynamic and transparent reconfigurations.
1. Hierarchical SDN Control Plane Architecture
University of Arizona | California Institute of Technology | Columbia University | Cornell University | Norfolk State University | Tuskegee | UC Berkeley | UC Los Angeles | UC San Diego | University of Southern California
Future Plans and Milestones
Center for Integrated Access Networks: Creating transformative optical technologies to enable an affordable faster internet for the future.
CIAN ERC Center for Integrated Access Networks under grant #EEC-0812072
Fig. 3 TOAN Testbed, UA Optical Sciences
Network Transport Nodes
CIAN Insertion Stations
Fig. 4 Network Control and Management Software
System Level Testbed
Local Lab Device
Characterization
HigherLevel
Integration/Packaging
SubsystemEvaluation
SystemEvaluationin
Testbed
• optical/electrical characterization
• response times
• yield, repeatability/reproducibility
• packaged IL
• thermal
• robustness
• cost
• integrated onboard functionalities
• cross coupling, timing issues
• packet-based performance
• application performance
• COTS equipment compatibility
Rapidchipscaleinsertion
Fig. 1 Role of System-level testbeds within CIAN’s tiered
evaluation process
FPGA FPGA FPGAFPGA
Extended SDN Controller Distribution
Control Data Plane Interface (CDPI)
Client/Servers2
Client/Servers3
Client/Servers1
Client/Servers4
IIC Protocol
LLC Ethernet
OpenFlow Extension
(e.g. OFPT_CFlOW_MOD)
CDC ROADM
Virtual Network 1 Virtual Network 2
VirtualMapping
Logical Packet
SDN Switch
OpenFlow 1.3
 Virtualization Layer: OpenFlow distribution layer provides northbound virtualization mapping service to
tenants which can then share the same physical network resource with different service priorites (e.g.
latency, cost, bandwidth).
 SDN Control Layer (Global Controll Layer): extended OpenFlow 1.3+ on SDN controller to support: (1)
Configuration on optical network (e.g. EDFA’s maximum output power). (2) Data collection (e.g. Topology
discovery and available wavelengths). (3) Action on optical network (e.g. Routing and Wavelength
Assignment).
 Control Plane Data Interface (CDPI layer) : enables the communication between Controller Layer and
FPGA Local Control Layer.
 Local Control Layer (FPGA layer) : FPGA controller supports the capability of distributed local control logic.
 Physical Layer: contains several Optical ROADMs and are directly controlled by FPGAs.
Fig. 5 High Level View of SDN Control Architecture
Milestone/Deliverable Date
Characterize OADM chips from UA from second Sandia run Jan 2015
Prototype SDN control plane with topology abstraction and
virtualizer
Mar 2015
Characterize and insert OPM chips from Cornell April 2015
Characterize and insert MEMS optical switch from Berkeley May 2015
Complete setup of SDN Datacenter switching emulation Aug 2015
Characterize and insert OSA on-chip from Columbia Sept 2015
Extend CIAN-boxes for optical flow switching and aggregation Oct 2016
Create agile aggregation CIAN-Box network May 2017
Fig. 2 Chip-Scale Testing, UA Optical Sciences
2. Dynamic Cross-layer Protection and Resource Allocation in OFDM Network
CDPI
FPGA 1
FPGA 4
FPGA 2
FPGA 3
ofp_flow_mod
ofp_cflow_mod
LLC packet
Command execution
ofp_flow_m
od
OpenFlow 1.3+
CDPI Control Channel:
Ethernet to Ethernet
Electrical Connection:
Ethernet to Ethernet
Optical Connection
Traffic Path
FPGA Control Channel:
Ethernet to RS232
RYU SDN Controller
Optical ROADM
with OSNR Monitor
Wavelength Tunable
Transceiver
SDN Packet Switch
TT
Server/Client
ROADM 1
ROADM 2
ROADM 3
ROADM 4
Switch 1
Switch 3
Source Destination
TT
TT
TT
TT
ASE
ASE
Amplified Spontaneous
Emission
Switch 2
Switch 4
Fig. 6 TOAN’s SDN-based control plane was enhanced to provide hierarchical
mechanisms for local-global control. The testbed is better able to support
investigation into candidate bandwidth-on-demand architectures that rely on agile
mechanisms for flexible wavelength assignment and dynamic adjustment of
modulation schemes based on real-time network monitoring.
100G PIC chip-scale
characterization
S-parameter, eye diagram, and BER
characterization of CWDM 1310 nm
transmitter, comprising four channels
at 25Gbs, targeted for DC
deployment. Statistical
characterization of large numbers of
passive OADM devices from the
CIAN chip.
Unified solar power-comms
grids disaster recovery
experiments (w/ QESST ERC,
ASU)
Studies to evaluate potential benefits
of jointly optimizing power grid and
communications grids for disaster
recovery. Physical layer emulations
consider stability and cascading
failure modes. Evaluation of the
performance of multi-domain SDN
control using different architectures.
Real-time software defined
OFDM resource allocation
Impairment-aware autonomous
format assignment and bandwidth
allocation was demonstrated using
real-time insitu OSNR monitors. SDN
controller reconfigures the OFDM by
changing number of subcarriers and
modulation formats based on channel
conditions.
TiSER commercial
interoperability test
Demonstration of real-time, in-service
BER estimation of streaming packet
data at 10Gb/s using Fujitsu FW9500
system. Establishes feasibility of
using TiSER as embedded network
monitoring system.
Holographic switch for rapidly
reconfigurable DC
interconnects
Feasibility study for DLP-based us
optical switch based on diffractive
switching method for providing a free
space rack-rack interconnect in the
DC. Analysis of throughput and
packet drops as a function of
datarate.
28 Gb/s 32 Gb/s
-60
-58
-56
-54
-52
-50
-48
-46
-44
-42
-40
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
S21(dB)
Frequency (GHz)
50mA, -2V EAM bias, 10dBm RF power, MORE MISALIGNED
S21(dB)
High Speed Characterization

More Related Content

PDF
Machine learning for 5G and beyond
 
PDF
Solutions and Protocolos for Vehicles Accident Notifications Using IoT
PDF
IRJET- A Review Paper of Performance Analysis Slotted CSMA/CA
PDF
Dubey Kaushalchandra resume
PPT
Wang lab introduction_1204_2018
PDF
ENHANCED PARTICLE SWARM OPTIMIZATION FOR EFFECTIVE RELAY NODES DEPLOYMENT IN ...
PPT
Wang lab introduction_1204_2018
PDF
2021 itu challenge_reinforcement_learning
Machine learning for 5G and beyond
 
Solutions and Protocolos for Vehicles Accident Notifications Using IoT
IRJET- A Review Paper of Performance Analysis Slotted CSMA/CA
Dubey Kaushalchandra resume
Wang lab introduction_1204_2018
ENHANCED PARTICLE SWARM OPTIMIZATION FOR EFFECTIVE RELAY NODES DEPLOYMENT IN ...
Wang lab introduction_1204_2018
2021 itu challenge_reinforcement_learning

What's hot (20)

PPT
A smart Multi-hop hierarchical routing protocol for Efficient VIdeo communica...
PDF
Orchestration of Ethernet Services in Software-Defined and Flexible Heterogen...
PDF
Routing protocol for hetrogeneous wireless mesh network
PDF
A Novel Routing Strategy Towards Achieving Ultra-Low End-to-End Latency in 6G...
PPTX
Adaptive weight factor estimation from user review 1
PDF
Ijarcet vol-2-issue-4-1420-1427
PDF
Power Networks Demonstration Centre, 5 Feb 14
DOCX
PDF
Statistical Dissemination Control in Large Machine-to-Machine Communication N...
PDF
AN ADVANCED QOS ANALYSIS AND EVALUATION METHOD FOR MOBILE INTERNET ACCESS
PDF
DESIGN OF ENERGY EFFICIENT ROUTING ALGORITHM FOR WIRELESS SENSOR NETWORK (WSN...
DOCX
Motivation and results coverage enhancment for 3GPP NR Rel.17
PDF
Scheduling wireless virtual networks functions
PDF
Novel Approach for Evaluating Video Transmission using Combined Scalable Vide...
PDF
Energy packet networks with energy harvesting
PDF
Hpc, grid and cloud computing - the past, present, and future challenge
PDF
Network experiences with Public Cloud Services @ TNC2017
PPT
Network Planning & Design: An Art or a Science?
PDF
IRJET- An Evaluation of the Performance Parameters of CMOS and CNTFET based D...
A smart Multi-hop hierarchical routing protocol for Efficient VIdeo communica...
Orchestration of Ethernet Services in Software-Defined and Flexible Heterogen...
Routing protocol for hetrogeneous wireless mesh network
A Novel Routing Strategy Towards Achieving Ultra-Low End-to-End Latency in 6G...
Adaptive weight factor estimation from user review 1
Ijarcet vol-2-issue-4-1420-1427
Power Networks Demonstration Centre, 5 Feb 14
Statistical Dissemination Control in Large Machine-to-Machine Communication N...
AN ADVANCED QOS ANALYSIS AND EVALUATION METHOD FOR MOBILE INTERNET ACCESS
DESIGN OF ENERGY EFFICIENT ROUTING ALGORITHM FOR WIRELESS SENSOR NETWORK (WSN...
Motivation and results coverage enhancment for 3GPP NR Rel.17
Scheduling wireless virtual networks functions
Novel Approach for Evaluating Video Transmission using Combined Scalable Vide...
Energy packet networks with energy harvesting
Hpc, grid and cloud computing - the past, present, and future challenge
Network experiences with Public Cloud Services @ TNC2017
Network Planning & Design: An Art or a Science?
IRJET- An Evaluation of the Performance Parameters of CMOS and CNTFET based D...
Ad

Viewers also liked (20)

PDF
Ciena Overview
PDF
RECONFIGURABLE OPTICAL ADD AND DROP MULTIPLEXERS A REVIEW
PDF
OTN tour 2015 opening speech
PDF
Software Defined Optical Networks - Mayur Channegowda
PPTX
Webinar: Learn How To Deploy High-Scale, Low-Latency Cost-Efficient Solutions...
PPTX
Otn barcelona presentation
PDF
2014 11 13-field-deployment-of-advanced-photonic-technologies-for-ultra-high-...
PPTX
Transport Solutions
PPT
A System's View of Metro and Regional Optical Networks
PDF
Optical network evolution
PDF
Evaluating efficiency of multi layer switching in future optical transport ne...
PDF
Benefits of multi layer bandwidth management in next generation core optical ...
PPTX
ECI - ElastiGRID™ - For utilities and strategic industries
PPTX
PPTX
OFC/NFOEC: Software Defined Optical Networks
PDF
OTN EMEA TOUR 2016 - OBIEE12c New Features for End-Users, Developers and Sys...
PPTX
ECI Risk Free Transition to Packet-UTC LATAM-April 2016
PPTX
What is l co s based wavelength selective switch
PPTX
Amplification, ROADM and Optical Networking activities at CPqD
PPTX
Emerging Trends and Applications for Cost Effective ROADMs
Ciena Overview
RECONFIGURABLE OPTICAL ADD AND DROP MULTIPLEXERS A REVIEW
OTN tour 2015 opening speech
Software Defined Optical Networks - Mayur Channegowda
Webinar: Learn How To Deploy High-Scale, Low-Latency Cost-Efficient Solutions...
Otn barcelona presentation
2014 11 13-field-deployment-of-advanced-photonic-technologies-for-ultra-high-...
Transport Solutions
A System's View of Metro and Regional Optical Networks
Optical network evolution
Evaluating efficiency of multi layer switching in future optical transport ne...
Benefits of multi layer bandwidth management in next generation core optical ...
ECI - ElastiGRID™ - For utilities and strategic industries
OFC/NFOEC: Software Defined Optical Networks
OTN EMEA TOUR 2016 - OBIEE12c New Features for End-Users, Developers and Sys...
ECI Risk Free Transition to Packet-UTC LATAM-April 2016
What is l co s based wavelength selective switch
Amplification, ROADM and Optical Networking activities at CPqD
Emerging Trends and Applications for Cost Effective ROADMs
Ad

Similar to TestbedLikun_final (20)

PDF
Software Innovations and Control Plane Evolution in the new SDN Transport Arc...
PDF
Towards an Open Data Center with an Interoperable Network (ODIN) Volume 3: So...
PDF
Software Innovations and Control Plane Evolution in the new SDN Transport Arc...
PPTX
On SDN Research Topics - Christian Esteve Rothenberg
PDF
B530429_FinalDissertation
PDF
Light Reading BTE_SDNtoolbox_June_2015
PPTX
Transport SDN Overview and Standards Update: Industry Perspectives
PDF
SDN Software Defined Networks 1st Edition Thomas Nadeau D.
PDF
SDN Software Defined Networks 1st Edition Thomas Nadeau D.
PPTX
Exhibitor session: Ciena
PPT
Lambda Data Grid
PDF
On the latency and jitter evaluation of software defined networks
PPTX
SDN 101: Software Defined Networking Course - Sameh Zaghloul/IBM - 2014
PDF
Software defined optical communication
PDF
OIF-plenary-Jan-20-2015_SDN-WAN-Loukas_oif2015.083
PPTX
sdn_based_controller_for_mobile_network_traffic_management1.pptx
PDF
Cognitive Technique for Software Defined Optical Network (SDON)
PPTX
Cpqd's SDN activities in optical dwdm terabit networks
PDF
My Ph.D. Defense - Software-Defined Systems for Network-Aware Service Compos...
PDF
Information Technology in Industry(ITII) - November Issue 2018
Software Innovations and Control Plane Evolution in the new SDN Transport Arc...
Towards an Open Data Center with an Interoperable Network (ODIN) Volume 3: So...
Software Innovations and Control Plane Evolution in the new SDN Transport Arc...
On SDN Research Topics - Christian Esteve Rothenberg
B530429_FinalDissertation
Light Reading BTE_SDNtoolbox_June_2015
Transport SDN Overview and Standards Update: Industry Perspectives
SDN Software Defined Networks 1st Edition Thomas Nadeau D.
SDN Software Defined Networks 1st Edition Thomas Nadeau D.
Exhibitor session: Ciena
Lambda Data Grid
On the latency and jitter evaluation of software defined networks
SDN 101: Software Defined Networking Course - Sameh Zaghloul/IBM - 2014
Software defined optical communication
OIF-plenary-Jan-20-2015_SDN-WAN-Loukas_oif2015.083
sdn_based_controller_for_mobile_network_traffic_management1.pptx
Cognitive Technique for Software Defined Optical Network (SDON)
Cpqd's SDN activities in optical dwdm terabit networks
My Ph.D. Defense - Software-Defined Systems for Network-Aware Service Compos...
Information Technology in Industry(ITII) - November Issue 2018

TestbedLikun_final

  • 1. Control Plane Progress Tiered Evaluation Strategy Project : T.2 Testbed for Optical Aggregation Networks (TOAN) Investigators: Wissinger, Kilper, Bergman, Cvijetic, Willner, Glick, Norwood Mission Other Progress Industry Collaborators • Research: enable CIAN components, devices, and networking concepts to be evaluated within a realistic network system infrastructure; link research efforts across the Center; instantiate the systems vision of CIAN. • Industry: provide a venue for R&D collaboration with industry partners; accelerate transfer and productization of CIAN and member company technologies. • Education: support the training of students with multi-disciplinary systems engineering, hands-on lab work, and project leadership; provide a platform for collaborative projects between CIAN students and industry. A significant strength of CIAN is the comprehensive testbed infrastructure that supports its vertical development chain from the materials level, through to chip scale components, packaged devices, and integrated subsystems.  Investigates impairment-aware and service-aware adaptation and reconfiguration of optical paths and coding/modulation schemes, especially in context of multi-node transmission management.  Explores application of software programmable network behaviors and hardware virtualization to enable intelligent and cost-effective aggregation schemes and to establish on-demand high BW connections over the WAN.  Evaluates CIAN technologies (e.g., chip-scale photonic devices, subsystems, control planes) within network system context, compare to benchmarks, provide feedback to device designers on deployment and interoperability issues.  Studies network stability under increasingly dynamic and transparent reconfigurations. 1. Hierarchical SDN Control Plane Architecture University of Arizona | California Institute of Technology | Columbia University | Cornell University | Norfolk State University | Tuskegee | UC Berkeley | UC Los Angeles | UC San Diego | University of Southern California Future Plans and Milestones Center for Integrated Access Networks: Creating transformative optical technologies to enable an affordable faster internet for the future. CIAN ERC Center for Integrated Access Networks under grant #EEC-0812072 Fig. 3 TOAN Testbed, UA Optical Sciences Network Transport Nodes CIAN Insertion Stations Fig. 4 Network Control and Management Software System Level Testbed Local Lab Device Characterization HigherLevel Integration/Packaging SubsystemEvaluation SystemEvaluationin Testbed • optical/electrical characterization • response times • yield, repeatability/reproducibility • packaged IL • thermal • robustness • cost • integrated onboard functionalities • cross coupling, timing issues • packet-based performance • application performance • COTS equipment compatibility Rapidchipscaleinsertion Fig. 1 Role of System-level testbeds within CIAN’s tiered evaluation process FPGA FPGA FPGAFPGA Extended SDN Controller Distribution Control Data Plane Interface (CDPI) Client/Servers2 Client/Servers3 Client/Servers1 Client/Servers4 IIC Protocol LLC Ethernet OpenFlow Extension (e.g. OFPT_CFlOW_MOD) CDC ROADM Virtual Network 1 Virtual Network 2 VirtualMapping Logical Packet SDN Switch OpenFlow 1.3  Virtualization Layer: OpenFlow distribution layer provides northbound virtualization mapping service to tenants which can then share the same physical network resource with different service priorites (e.g. latency, cost, bandwidth).  SDN Control Layer (Global Controll Layer): extended OpenFlow 1.3+ on SDN controller to support: (1) Configuration on optical network (e.g. EDFA’s maximum output power). (2) Data collection (e.g. Topology discovery and available wavelengths). (3) Action on optical network (e.g. Routing and Wavelength Assignment).  Control Plane Data Interface (CDPI layer) : enables the communication between Controller Layer and FPGA Local Control Layer.  Local Control Layer (FPGA layer) : FPGA controller supports the capability of distributed local control logic.  Physical Layer: contains several Optical ROADMs and are directly controlled by FPGAs. Fig. 5 High Level View of SDN Control Architecture Milestone/Deliverable Date Characterize OADM chips from UA from second Sandia run Jan 2015 Prototype SDN control plane with topology abstraction and virtualizer Mar 2015 Characterize and insert OPM chips from Cornell April 2015 Characterize and insert MEMS optical switch from Berkeley May 2015 Complete setup of SDN Datacenter switching emulation Aug 2015 Characterize and insert OSA on-chip from Columbia Sept 2015 Extend CIAN-boxes for optical flow switching and aggregation Oct 2016 Create agile aggregation CIAN-Box network May 2017 Fig. 2 Chip-Scale Testing, UA Optical Sciences 2. Dynamic Cross-layer Protection and Resource Allocation in OFDM Network CDPI FPGA 1 FPGA 4 FPGA 2 FPGA 3 ofp_flow_mod ofp_cflow_mod LLC packet Command execution ofp_flow_m od OpenFlow 1.3+ CDPI Control Channel: Ethernet to Ethernet Electrical Connection: Ethernet to Ethernet Optical Connection Traffic Path FPGA Control Channel: Ethernet to RS232 RYU SDN Controller Optical ROADM with OSNR Monitor Wavelength Tunable Transceiver SDN Packet Switch TT Server/Client ROADM 1 ROADM 2 ROADM 3 ROADM 4 Switch 1 Switch 3 Source Destination TT TT TT TT ASE ASE Amplified Spontaneous Emission Switch 2 Switch 4 Fig. 6 TOAN’s SDN-based control plane was enhanced to provide hierarchical mechanisms for local-global control. The testbed is better able to support investigation into candidate bandwidth-on-demand architectures that rely on agile mechanisms for flexible wavelength assignment and dynamic adjustment of modulation schemes based on real-time network monitoring. 100G PIC chip-scale characterization S-parameter, eye diagram, and BER characterization of CWDM 1310 nm transmitter, comprising four channels at 25Gbs, targeted for DC deployment. Statistical characterization of large numbers of passive OADM devices from the CIAN chip. Unified solar power-comms grids disaster recovery experiments (w/ QESST ERC, ASU) Studies to evaluate potential benefits of jointly optimizing power grid and communications grids for disaster recovery. Physical layer emulations consider stability and cascading failure modes. Evaluation of the performance of multi-domain SDN control using different architectures. Real-time software defined OFDM resource allocation Impairment-aware autonomous format assignment and bandwidth allocation was demonstrated using real-time insitu OSNR monitors. SDN controller reconfigures the OFDM by changing number of subcarriers and modulation formats based on channel conditions. TiSER commercial interoperability test Demonstration of real-time, in-service BER estimation of streaming packet data at 10Gb/s using Fujitsu FW9500 system. Establishes feasibility of using TiSER as embedded network monitoring system. Holographic switch for rapidly reconfigurable DC interconnects Feasibility study for DLP-based us optical switch based on diffractive switching method for providing a free space rack-rack interconnect in the DC. Analysis of throughput and packet drops as a function of datarate. 28 Gb/s 32 Gb/s -60 -58 -56 -54 -52 -50 -48 -46 -44 -42 -40 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 S21(dB) Frequency (GHz) 50mA, -2V EAM bias, 10dBm RF power, MORE MISALIGNED S21(dB) High Speed Characterization