SlideShare ist ein Scribd-Unternehmen logo
1 von 18
Downloaden Sie, um offline zu lesen
Institute  of  Computer  Science
Chair  of  Communication  Networks
Prof.  Dr.-­Ing.  P.  Tran-­Gia
Investigating  the  Impact  of  Network  Topology  
on  the  Processing  Times  of  SDN  Controllers
Christopher  Metter,  Steffen  Gebert,  Stanislav  Lange,  
Thomas  Zinner,  Phuoc Tran-­Gia,  Michael  Jarschel*
www3.informatik.uni-­wuerzburg.de
* Nokia GmbH, Munich
Steffen  Gebert
2Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
Metadata
u This  paper  was  presented  at  the  “The  Seventh  IFIP/IEEE  
International  Workshop  on  Management  of  the  Future  Internet”  
(ManFI),  colocated with  IM`2015  in  Ottawa,  Canada.
u http://manfi.org
u Paper  author  version:
http://www3.informatik.uni-­wuerzburg.de/staff/steffen.gebert/author_versions/conf_510_author_versions.pdf
Steffen  Gebert
3Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
Motivation  – SDN  Architecture
Data PlaneData Plane
C
D
C
D
Control Plane
Steffen  Gebert
4Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
Impact  of  Controller  Implementations  on  Message  Processing  Times
u Analysis  of  processing  times  for  a  L2  learning  switch
§ Switching  modules  of  OpenDayLight (ODL)  and  Ryu
§ Packet-­in/packet-­out  pairs
– e.g.,  in  case  of  unconnected  switches  or  requests  for  non-­reachable  
destinations  
§ Packet-­in-­packet-­out/flow-­mod  pairs
– e.g.,  to  set  a  forwarding  path  in  the  flow  table  to  a  certain  destination  
u Topology  emulation  and  packet-­in  traffic  generation
§ Controlled,  low-­rate  traffic  emulation  to  reduce  waiting  times
§ Impact  of  network  topology  on  the  service  times  
– Unconnected  topology  (baseline)
– Fat-­tree  topology  (data  center)
– WAN  topologies  (261  topologies  from  the  Internet  Topology  Zoo)
Steffen  Gebert
5Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
OFCProbe
github.com/lsinfo3/ofcprobe
u Controller  performance  analysis  tool
§ Emulation  of  virtual  OpenFlow-­enabled  switches
§ Generate  OF  messages  per  switch
§ PCAP  file  playback
§ Configurable  payload  IAT  distributions
§ Per-­switch  statistics  collection
C
Steffen  Gebert
6Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
Topology Discovery
u Switches  are configured to connect to the controller
u Controller  has no information about links  between switches
??
?
1
2
3
C
Steffen  Gebert
7Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
?
Topology Discovery
u Controller  injects LLDP  packets (Link  Layer  Discovery  Protocol)
u Switches  flood them to all  ports
?
LLDP
LLDP
PacketOut
(flood)
LLDP
?
1
2
3
C
Steffen  Gebert
8Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
C
Topology Discovery
u Other  switches receive packets and  report packet-­in  to controller
u Controller  learns topology from information about incoming ports
?
PacketOut
(flood)
LLDP
PacketIn
(port 7)
LLDP
PacketIn
(port 2)
LLDP
Switch Port To Switch
2 2 1
3 7 1
1
2
3
2
7
Steffen  Gebert
9Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
Processing  Steps of  a  Controller  for Packet-­Ins
1. Data  reception
2. Decoding  of  OpenFlow message  and  classification
3. Attribute  extraction  and  delegation  to  application  module
4. Module-­specific  processing  
5. Response  generation  (e.g.,  packet-­out  /  flow-­mod  message  …)
6. Data  transmission
Impact  of  controller  implementation  on  module-­specific  processing?
Steffen  Gebert
10Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
Controller
C
Baseline  Topology
u 20  switches,  1  host  per  switch,  no  connections  between  switches
u Resulting  action  in  packet-­out  message:  flood
Steffen  Gebert
11Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
Baseline  Performance
u Emulated switches generate 16  packet-­ins  one after  another
u Controller  responds with packet-­out  message
u Periodic behavior in  response times (per  switch)
u Higher  variation of  processing times for ODL  (JVM?)
u ODL  outperforms Ryu
Steffen  Gebert
12Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
C
u 20  switches,  represents data center topology
Fat-­Tree
Steffen  Gebert
13Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
Packet-­Out:  Fat-­Tree
u Bursts  of 16  packets are sent from one host  to another
u Increased processing times compared to unconnected switches
u Similar average processing times,  but  high  variation and outliers
when using ODL
Steffen  Gebert
14Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
Flow-­Mod:  Fat-­Tree
u Flow-­mods by utilizing Layer  2  learning switch module
u Processing  times for flow-­mods 2-­8  times higher than packet-­outs
u ODL  processing times stabilize after  ~20  seconds
u Ryu has nearly constant processing times and outperforms ODL
Steffen  Gebert
15Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
u Experiment  performed on  261  WAN  topologies
u CDF  shows all  measured processing times
u Ryu:  Topology has no effect on  processing times,  flow-­mod
processing slightly slower than packet-­out  processing
u ODL:  High  variation for packet-­out,  flow-­mods significantly longer
Internet  Topology Zoo  – Aggregated Results
Steffen  Gebert
16Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
Topology  Investigation
u Analysis  of  the  topology  impact  on  the  processing  times
u Ryu:  rather  constant  behavior,  similar  processing  times  for  flow-­
mods  and  packet-­outs
u ODL:  high  impact  of  message  type,  varying  processing  times  
Steffen  Gebert
17Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
u Investigation  of  controller  processing  times
§ Emulated  different  network  topologies  using  OFCProbe
§ Time  measurements  done  directly  within  the  controllers
§ Using Layer  2  switch modules (as starting point)
u Benchmarking  results:
§ Ryu shows  very  constant  response  times
§ ODL  faster  for  packet-­out,  Ryu for  flow-­mod
u Main  take-­aways:
§ Test  your  own  forwarding  algorithms  using  different  topologies!
§ OFCProbe now  allows  importing  GraphML topologies
§ Measurement  probes  available  for  ODL  and  Ryu controllers1
1 source code on  github.com/lsinfo3  (manfi2015-­perf-­*  repos)
Summary
Steffen  Gebert
18Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers
OFCProbe Architecture
OF  Controller
Best  Effort  
Traffic  Generator
OpenFlow  Connection  Handler  Module
Configuration
Communication  
Layer
Traffic  Generation  
Layer
Traffic  Generation  Module
Statistics
Statistics:  
PPS  Module
Statistics:  RTT  
Module
Statistics:  
Custom  
Module
Interface
Configuration
Packets
Statistics
Implementation
Payload
OF Message
OF Reply
OF ReplyOF Reply

Weitere ähnliche Inhalte

Was ist angesagt?

Frenetic: A Programming Language for OpenFlow Networks
Frenetic: A Programming Language for OpenFlow NetworksFrenetic: A Programming Language for OpenFlow Networks
Frenetic: A Programming Language for OpenFlow NetworksOpen Networking Summits
 
VeriFlow Presentation
VeriFlow PresentationVeriFlow Presentation
VeriFlow PresentationKrystle Bates
 
Introduction to OpenFlow
Introduction to OpenFlowIntroduction to OpenFlow
Introduction to OpenFlowrjain51
 
The dark side of SDN and OpenFlow
The dark side of SDN and OpenFlowThe dark side of SDN and OpenFlow
The dark side of SDN and OpenFlowDiego Kreutz
 
SDN Networks Programming Languages
SDN Networks Programming LanguagesSDN Networks Programming Languages
SDN Networks Programming LanguagesFlavio Vit
 
software defined network, openflow protocol and its controllers
software defined network, openflow protocol and its controllerssoftware defined network, openflow protocol and its controllers
software defined network, openflow protocol and its controllersIsaku Yamahata
 
Loadbalancing In-depth study for scale @ 80K TPS
Loadbalancing In-depth study for scale @ 80K TPSLoadbalancing In-depth study for scale @ 80K TPS
Loadbalancing In-depth study for scale @ 80K TPSShrey Agarwal
 
LF_DPDK17_GRO/GSO Libraries: Bring Significant Performance Gains to DPDK-base...
LF_DPDK17_GRO/GSO Libraries: Bring Significant Performance Gains to DPDK-base...LF_DPDK17_GRO/GSO Libraries: Bring Significant Performance Gains to DPDK-base...
LF_DPDK17_GRO/GSO Libraries: Bring Significant Performance Gains to DPDK-base...LF_DPDK
 
Ch 03 --- the OpenFlow protocols
Ch 03 --- the OpenFlow protocolsCh 03 --- the OpenFlow protocols
Ch 03 --- the OpenFlow protocolsYoram Orzach
 
IPv6 Segment Routing : an end-to-end solution ?
IPv6 Segment Routing : an end-to-end solution ?IPv6 Segment Routing : an end-to-end solution ?
IPv6 Segment Routing : an end-to-end solution ?Olivier Bonaventure
 
Open stack with_openflowsdn-torii
Open stack with_openflowsdn-toriiOpen stack with_openflowsdn-torii
Open stack with_openflowsdn-toriiHui Cheng
 
Introduction to Software Defined Networking (SDN)
Introduction to Software Defined Networking (SDN)Introduction to Software Defined Networking (SDN)
Introduction to Software Defined Networking (SDN)rjain51
 
LF_DPDK17_Integrating and using DPDK with Open vSwitch
LF_DPDK17_Integrating and using DPDK with Open vSwitchLF_DPDK17_Integrating and using DPDK with Open vSwitch
LF_DPDK17_Integrating and using DPDK with Open vSwitchLF_DPDK
 
Sun RPC (Remote Procedure Call)
Sun RPC (Remote Procedure Call)Sun RPC (Remote Procedure Call)
Sun RPC (Remote Procedure Call)Peter R. Egli
 
SDN: is it a solution for network security?
SDN: is it a solution for network security?SDN: is it a solution for network security?
SDN: is it a solution for network security?ARCCN
 
Modern Software Architecture
Modern Software Architecture Modern Software Architecture
Modern Software Architecture Ahmed Marzouk
 
Tutorial on SDN data plane evolution
Tutorial on SDN data plane evolutionTutorial on SDN data plane evolution
Tutorial on SDN data plane evolutionAntonio Capone
 
Dynamic Classification in a Silicon-Based Forwarding Engine
Dynamic Classification in a Silicon-Based Forwarding EngineDynamic Classification in a Silicon-Based Forwarding Engine
Dynamic Classification in a Silicon-Based Forwarding EngineTal Lavian Ph.D.
 

Was ist angesagt? (20)

Frenetic: A Programming Language for OpenFlow Networks
Frenetic: A Programming Language for OpenFlow NetworksFrenetic: A Programming Language for OpenFlow Networks
Frenetic: A Programming Language for OpenFlow Networks
 
VeriFlow Presentation
VeriFlow PresentationVeriFlow Presentation
VeriFlow Presentation
 
Introduction to OpenFlow
Introduction to OpenFlowIntroduction to OpenFlow
Introduction to OpenFlow
 
The dark side of SDN and OpenFlow
The dark side of SDN and OpenFlowThe dark side of SDN and OpenFlow
The dark side of SDN and OpenFlow
 
SDN Networks Programming Languages
SDN Networks Programming LanguagesSDN Networks Programming Languages
SDN Networks Programming Languages
 
software defined network, openflow protocol and its controllers
software defined network, openflow protocol and its controllerssoftware defined network, openflow protocol and its controllers
software defined network, openflow protocol and its controllers
 
Loadbalancing In-depth study for scale @ 80K TPS
Loadbalancing In-depth study for scale @ 80K TPSLoadbalancing In-depth study for scale @ 80K TPS
Loadbalancing In-depth study for scale @ 80K TPS
 
LF_DPDK17_GRO/GSO Libraries: Bring Significant Performance Gains to DPDK-base...
LF_DPDK17_GRO/GSO Libraries: Bring Significant Performance Gains to DPDK-base...LF_DPDK17_GRO/GSO Libraries: Bring Significant Performance Gains to DPDK-base...
LF_DPDK17_GRO/GSO Libraries: Bring Significant Performance Gains to DPDK-base...
 
Ch 03 --- the OpenFlow protocols
Ch 03 --- the OpenFlow protocolsCh 03 --- the OpenFlow protocols
Ch 03 --- the OpenFlow protocols
 
IPv6 Segment Routing : an end-to-end solution ?
IPv6 Segment Routing : an end-to-end solution ?IPv6 Segment Routing : an end-to-end solution ?
IPv6 Segment Routing : an end-to-end solution ?
 
Open stack with_openflowsdn-torii
Open stack with_openflowsdn-toriiOpen stack with_openflowsdn-torii
Open stack with_openflowsdn-torii
 
Introduction to Software Defined Networking (SDN)
Introduction to Software Defined Networking (SDN)Introduction to Software Defined Networking (SDN)
Introduction to Software Defined Networking (SDN)
 
LF_DPDK17_Integrating and using DPDK with Open vSwitch
LF_DPDK17_Integrating and using DPDK with Open vSwitchLF_DPDK17_Integrating and using DPDK with Open vSwitch
LF_DPDK17_Integrating and using DPDK with Open vSwitch
 
message passing
 message passing message passing
message passing
 
Sun RPC (Remote Procedure Call)
Sun RPC (Remote Procedure Call)Sun RPC (Remote Procedure Call)
Sun RPC (Remote Procedure Call)
 
SDN: is it a solution for network security?
SDN: is it a solution for network security?SDN: is it a solution for network security?
SDN: is it a solution for network security?
 
Parallel computing(1)
Parallel computing(1)Parallel computing(1)
Parallel computing(1)
 
Modern Software Architecture
Modern Software Architecture Modern Software Architecture
Modern Software Architecture
 
Tutorial on SDN data plane evolution
Tutorial on SDN data plane evolutionTutorial on SDN data plane evolution
Tutorial on SDN data plane evolution
 
Dynamic Classification in a Silicon-Based Forwarding Engine
Dynamic Classification in a Silicon-Based Forwarding EngineDynamic Classification in a Silicon-Based Forwarding Engine
Dynamic Classification in a Silicon-Based Forwarding Engine
 

Andere mochten auch

The Development Infrastructure of the TYPO3 Project
The Development Infrastructure of the TYPO3 ProjectThe Development Infrastructure of the TYPO3 Project
The Development Infrastructure of the TYPO3 ProjectSteffen Gebert
 
Cv Ebo Bakker 2016
Cv Ebo Bakker 2016Cv Ebo Bakker 2016
Cv Ebo Bakker 2016Ebo Bakker
 
Mobile Cloud Computing : The Upcoming Trend !
Mobile Cloud Computing : The Upcoming Trend !Mobile Cloud Computing : The Upcoming Trend !
Mobile Cloud Computing : The Upcoming Trend !Sai Natkar
 
Vehicular Delay Tolerant Network (VDTN): Routing Perspectives
Vehicular Delay Tolerant Network (VDTN):Routing PerspectivesVehicular Delay Tolerant Network (VDTN):Routing Perspectives
Vehicular Delay Tolerant Network (VDTN): Routing PerspectivesSyed Hassan Ahmed
 
Distributed routing
Distributed routingDistributed routing
Distributed routingMurali Reddy
 
Innovation in SDN Tools and Platforms
Innovation in SDN Tools and PlatformsInnovation in SDN Tools and Platforms
Innovation in SDN Tools and PlatformsUmesh Krishnaswamy
 
Cloud Computing #Manoj_Rockstar
Cloud Computing #Manoj_RockstarCloud Computing #Manoj_Rockstar
Cloud Computing #Manoj_RockstarManoj Magatapalli
 
Is SDN ready for primetime?
Is SDN ready for primetime?Is SDN ready for primetime?
Is SDN ready for primetime?APNIC
 
Webinar: Hybrid Cloud Integration - Why It's Different and Why It Matters
Webinar: Hybrid Cloud Integration - Why It's Different and Why It MattersWebinar: Hybrid Cloud Integration - Why It's Different and Why It Matters
Webinar: Hybrid Cloud Integration - Why It's Different and Why It MattersSnapLogic
 
Routing, Network Performance, and Role of Analytics
Routing, Network Performance, and Role of AnalyticsRouting, Network Performance, and Role of Analytics
Routing, Network Performance, and Role of AnalyticsAPNIC
 
Real world hybrid cloud session - OpenStack DACH 2015
Real world hybrid cloud session - OpenStack DACH 2015Real world hybrid cloud session - OpenStack DACH 2015
Real world hybrid cloud session - OpenStack DACH 2015assafleb
 
Cloud computing Brief
Cloud computing BriefCloud computing Brief
Cloud computing BriefJustin Lee
 

Andere mochten auch (13)

The Development Infrastructure of the TYPO3 Project
The Development Infrastructure of the TYPO3 ProjectThe Development Infrastructure of the TYPO3 Project
The Development Infrastructure of the TYPO3 Project
 
ccna3
ccna3ccna3
ccna3
 
Cv Ebo Bakker 2016
Cv Ebo Bakker 2016Cv Ebo Bakker 2016
Cv Ebo Bakker 2016
 
Mobile Cloud Computing : The Upcoming Trend !
Mobile Cloud Computing : The Upcoming Trend !Mobile Cloud Computing : The Upcoming Trend !
Mobile Cloud Computing : The Upcoming Trend !
 
Vehicular Delay Tolerant Network (VDTN): Routing Perspectives
Vehicular Delay Tolerant Network (VDTN):Routing PerspectivesVehicular Delay Tolerant Network (VDTN):Routing Perspectives
Vehicular Delay Tolerant Network (VDTN): Routing Perspectives
 
Distributed routing
Distributed routingDistributed routing
Distributed routing
 
Innovation in SDN Tools and Platforms
Innovation in SDN Tools and PlatformsInnovation in SDN Tools and Platforms
Innovation in SDN Tools and Platforms
 
Cloud Computing #Manoj_Rockstar
Cloud Computing #Manoj_RockstarCloud Computing #Manoj_Rockstar
Cloud Computing #Manoj_Rockstar
 
Is SDN ready for primetime?
Is SDN ready for primetime?Is SDN ready for primetime?
Is SDN ready for primetime?
 
Webinar: Hybrid Cloud Integration - Why It's Different and Why It Matters
Webinar: Hybrid Cloud Integration - Why It's Different and Why It MattersWebinar: Hybrid Cloud Integration - Why It's Different and Why It Matters
Webinar: Hybrid Cloud Integration - Why It's Different and Why It Matters
 
Routing, Network Performance, and Role of Analytics
Routing, Network Performance, and Role of AnalyticsRouting, Network Performance, and Role of Analytics
Routing, Network Performance, and Role of Analytics
 
Real world hybrid cloud session - OpenStack DACH 2015
Real world hybrid cloud session - OpenStack DACH 2015Real world hybrid cloud session - OpenStack DACH 2015
Real world hybrid cloud session - OpenStack DACH 2015
 
Cloud computing Brief
Cloud computing BriefCloud computing Brief
Cloud computing Brief
 

Ähnlich wie Investigating the Impact of Network Topology on the Processing Times of SDN Controllers

Analytical Modeling of End-to-End Delay in OpenFlow Based Networks
Analytical Modeling of End-to-End Delay in OpenFlow Based NetworksAnalytical Modeling of End-to-End Delay in OpenFlow Based Networks
Analytical Modeling of End-to-End Delay in OpenFlow Based NetworksAzeem Iqbal
 
Network programmability: an Overview
Network programmability: an Overview Network programmability: an Overview
Network programmability: an Overview Aymen AlAwadi
 
IJSRED-V1I1P2
IJSRED-V1I1P2IJSRED-V1I1P2
IJSRED-V1I1P2IJSRED
 
SDN Security Talk - (ISC)2_3
SDN Security Talk - (ISC)2_3SDN Security Talk - (ISC)2_3
SDN Security Talk - (ISC)2_3Wen-Pai Lu
 
IRJET- SDN Simulation in Mininet to Provide Security Via Firewall
IRJET- SDN Simulation in Mininet to Provide Security Via FirewallIRJET- SDN Simulation in Mininet to Provide Security Via Firewall
IRJET- SDN Simulation in Mininet to Provide Security Via FirewallIRJET Journal
 
Software-Defined Networking Changes for the Paradigm for Mission-Critical Ope...
Software-Defined Networking Changes for the Paradigm for Mission-Critical Ope...Software-Defined Networking Changes for the Paradigm for Mission-Critical Ope...
Software-Defined Networking Changes for the Paradigm for Mission-Critical Ope...Wheeler Flemming
 
Lecture12 ie321 dr_atifshahzad - networks
Lecture12 ie321 dr_atifshahzad - networksLecture12 ie321 dr_atifshahzad - networks
Lecture12 ie321 dr_atifshahzad - networksAtif Shahzad
 
FOREST FIRE DETECTION SYSTEM USING XBEE
FOREST FIRE DETECTION SYSTEM USING XBEEFOREST FIRE DETECTION SYSTEM USING XBEE
FOREST FIRE DETECTION SYSTEM USING XBEETalvinder Singh
 
What is SDN and how to approach it with Python
What is SDN and how to approach it with PythonWhat is SDN and how to approach it with Python
What is SDN and how to approach it with PythonJustin Park
 
Analysis of IT Monitoring Using Open Source Software Techniques: A Review
Analysis of IT Monitoring Using Open Source Software Techniques: A ReviewAnalysis of IT Monitoring Using Open Source Software Techniques: A Review
Analysis of IT Monitoring Using Open Source Software Techniques: A ReviewIJERD Editor
 
btNOG 5: Network Automation
btNOG 5: Network AutomationbtNOG 5: Network Automation
btNOG 5: Network AutomationAPNIC
 
Software defined network
Software defined networkSoftware defined network
Software defined networkBogamoga1
 
Naveen nimmu sdn future of networking
Naveen nimmu sdn   future of networkingNaveen nimmu sdn   future of networking
Naveen nimmu sdn future of networkingOpenSourceIndia
 
Naveen nimmu sdn future of networking
Naveen nimmu sdn   future of networkingNaveen nimmu sdn   future of networking
Naveen nimmu sdn future of networkingsuniltomar04
 

Ähnlich wie Investigating the Impact of Network Topology on the Processing Times of SDN Controllers (20)

Analytical Modeling of End-to-End Delay in OpenFlow Based Networks
Analytical Modeling of End-to-End Delay in OpenFlow Based NetworksAnalytical Modeling of End-to-End Delay in OpenFlow Based Networks
Analytical Modeling of End-to-End Delay in OpenFlow Based Networks
 
Network programmability: an Overview
Network programmability: an Overview Network programmability: an Overview
Network programmability: an Overview
 
SDN approach.pptx
SDN approach.pptxSDN approach.pptx
SDN approach.pptx
 
Lecture 1.pdf
Lecture 1.pdfLecture 1.pdf
Lecture 1.pdf
 
MininetasSDNPlatform.pdf
MininetasSDNPlatform.pdfMininetasSDNPlatform.pdf
MininetasSDNPlatform.pdf
 
IJSRED-V1I1P2
IJSRED-V1I1P2IJSRED-V1I1P2
IJSRED-V1I1P2
 
SDN Security Talk - (ISC)2_3
SDN Security Talk - (ISC)2_3SDN Security Talk - (ISC)2_3
SDN Security Talk - (ISC)2_3
 
CCNA project-report
CCNA project-reportCCNA project-report
CCNA project-report
 
IRJET- SDN Simulation in Mininet to Provide Security Via Firewall
IRJET- SDN Simulation in Mininet to Provide Security Via FirewallIRJET- SDN Simulation in Mininet to Provide Security Via Firewall
IRJET- SDN Simulation in Mininet to Provide Security Via Firewall
 
sdnppt.pdf
sdnppt.pdfsdnppt.pdf
sdnppt.pdf
 
Open Networking
Open NetworkingOpen Networking
Open Networking
 
Software-Defined Networking Changes for the Paradigm for Mission-Critical Ope...
Software-Defined Networking Changes for the Paradigm for Mission-Critical Ope...Software-Defined Networking Changes for the Paradigm for Mission-Critical Ope...
Software-Defined Networking Changes for the Paradigm for Mission-Critical Ope...
 
Lecture12 ie321 dr_atifshahzad - networks
Lecture12 ie321 dr_atifshahzad - networksLecture12 ie321 dr_atifshahzad - networks
Lecture12 ie321 dr_atifshahzad - networks
 
FOREST FIRE DETECTION SYSTEM USING XBEE
FOREST FIRE DETECTION SYSTEM USING XBEEFOREST FIRE DETECTION SYSTEM USING XBEE
FOREST FIRE DETECTION SYSTEM USING XBEE
 
What is SDN and how to approach it with Python
What is SDN and how to approach it with PythonWhat is SDN and how to approach it with Python
What is SDN and how to approach it with Python
 
Analysis of IT Monitoring Using Open Source Software Techniques: A Review
Analysis of IT Monitoring Using Open Source Software Techniques: A ReviewAnalysis of IT Monitoring Using Open Source Software Techniques: A Review
Analysis of IT Monitoring Using Open Source Software Techniques: A Review
 
btNOG 5: Network Automation
btNOG 5: Network AutomationbtNOG 5: Network Automation
btNOG 5: Network Automation
 
Software defined network
Software defined networkSoftware defined network
Software defined network
 
Naveen nimmu sdn future of networking
Naveen nimmu sdn   future of networkingNaveen nimmu sdn   future of networking
Naveen nimmu sdn future of networking
 
Naveen nimmu sdn future of networking
Naveen nimmu sdn   future of networkingNaveen nimmu sdn   future of networking
Naveen nimmu sdn future of networking
 

Mehr von Steffen Gebert

Building an IoT SuperNetwork on top of the AWS Global Infrastructure
Building an IoT SuperNetwork on top of the AWS Global InfrastructureBuilding an IoT SuperNetwork on top of the AWS Global Infrastructure
Building an IoT SuperNetwork on top of the AWS Global InfrastructureSteffen Gebert
 
Wenn selbst ‘erlaube allen Verkehr von 0.0.0.0/0’ nicht hilft - Verbindungspr...
Wenn selbst ‘erlaube allen Verkehr von 0.0.0.0/0’ nicht hilft - Verbindungspr...Wenn selbst ‘erlaube allen Verkehr von 0.0.0.0/0’ nicht hilft - Verbindungspr...
Wenn selbst ‘erlaube allen Verkehr von 0.0.0.0/0’ nicht hilft - Verbindungspr...Steffen Gebert
 
Feature Management Platforms
Feature Management PlatformsFeature Management Platforms
Feature Management PlatformsSteffen Gebert
 
Serverless Networking - How We Provide Cloud-Native Connectivity for IoT Devices
Serverless Networking - How We Provide Cloud-Native Connectivity for IoT DevicesServerless Networking - How We Provide Cloud-Native Connectivity for IoT Devices
Serverless Networking - How We Provide Cloud-Native Connectivity for IoT DevicesSteffen Gebert
 
How our Cloudy Mindsets Approached Physical Routers
How our Cloudy Mindsets Approached Physical RoutersHow our Cloudy Mindsets Approached Physical Routers
How our Cloudy Mindsets Approached Physical RoutersSteffen Gebert
 
Jenkins vs. AWS CodePipeline (AWS User Group Berlin)
Jenkins vs. AWS CodePipeline (AWS User Group Berlin)Jenkins vs. AWS CodePipeline (AWS User Group Berlin)
Jenkins vs. AWS CodePipeline (AWS User Group Berlin)Steffen Gebert
 
Jenkins vs. AWS CodePipeline
Jenkins vs. AWS CodePipelineJenkins vs. AWS CodePipeline
Jenkins vs. AWS CodePipelineSteffen Gebert
 
Monitoring Akka with Kamon 1.0
Monitoring Akka with Kamon 1.0Monitoring Akka with Kamon 1.0
Monitoring Akka with Kamon 1.0Steffen Gebert
 
(Declarative) Jenkins Pipelines
(Declarative) Jenkins Pipelines(Declarative) Jenkins Pipelines
(Declarative) Jenkins PipelinesSteffen Gebert
 
An Open-Source Chef Cookbook CI/CD Implementation Using Jenkins Pipelines
An Open-Source Chef Cookbook CI/CD Implementation Using Jenkins PipelinesAn Open-Source Chef Cookbook CI/CD Implementation Using Jenkins Pipelines
An Open-Source Chef Cookbook CI/CD Implementation Using Jenkins PipelinesSteffen Gebert
 
Let's go HTTPS-only! - More Than Buying a Certificate
Let's go HTTPS-only! - More Than Buying a CertificateLet's go HTTPS-only! - More Than Buying a Certificate
Let's go HTTPS-only! - More Than Buying a CertificateSteffen Gebert
 
Cleaning Up the Dirt of the Nineties - How New Protocols are Modernizing the Web
Cleaning Up the Dirt of the Nineties - How New Protocols are Modernizing the WebCleaning Up the Dirt of the Nineties - How New Protocols are Modernizing the Web
Cleaning Up the Dirt of the Nineties - How New Protocols are Modernizing the WebSteffen Gebert
 
Der Weg zu TYPO3 CMS 6.0 und Einblicke in die TYPO3-Entwicklung
Der Weg zu TYPO3 CMS 6.0 und Einblicke in die TYPO3-EntwicklungDer Weg zu TYPO3 CMS 6.0 und Einblicke in die TYPO3-Entwicklung
Der Weg zu TYPO3 CMS 6.0 und Einblicke in die TYPO3-EntwicklungSteffen Gebert
 
Official typo3.org infrastructure &
the TYPO3 Server Admin Team
Official typo3.org infrastructure &
the TYPO3 Server Admin TeamOfficial typo3.org infrastructure &
the TYPO3 Server Admin Team
Official typo3.org infrastructure &
the TYPO3 Server Admin TeamSteffen Gebert
 
Neuigkeiten aus dem TYPO3-Projekt
Neuigkeiten aus dem TYPO3-ProjektNeuigkeiten aus dem TYPO3-Projekt
Neuigkeiten aus dem TYPO3-ProjektSteffen Gebert
 
The TYPO3 Server Admin Team
The TYPO3 Server Admin TeamThe TYPO3 Server Admin Team
The TYPO3 Server Admin TeamSteffen Gebert
 

Mehr von Steffen Gebert (20)

Building an IoT SuperNetwork on top of the AWS Global Infrastructure
Building an IoT SuperNetwork on top of the AWS Global InfrastructureBuilding an IoT SuperNetwork on top of the AWS Global Infrastructure
Building an IoT SuperNetwork on top of the AWS Global Infrastructure
 
Wenn selbst ‘erlaube allen Verkehr von 0.0.0.0/0’ nicht hilft - Verbindungspr...
Wenn selbst ‘erlaube allen Verkehr von 0.0.0.0/0’ nicht hilft - Verbindungspr...Wenn selbst ‘erlaube allen Verkehr von 0.0.0.0/0’ nicht hilft - Verbindungspr...
Wenn selbst ‘erlaube allen Verkehr von 0.0.0.0/0’ nicht hilft - Verbindungspr...
 
Feature Management Platforms
Feature Management PlatformsFeature Management Platforms
Feature Management Platforms
 
Serverless Networking - How We Provide Cloud-Native Connectivity for IoT Devices
Serverless Networking - How We Provide Cloud-Native Connectivity for IoT DevicesServerless Networking - How We Provide Cloud-Native Connectivity for IoT Devices
Serverless Networking - How We Provide Cloud-Native Connectivity for IoT Devices
 
How our Cloudy Mindsets Approached Physical Routers
How our Cloudy Mindsets Approached Physical RoutersHow our Cloudy Mindsets Approached Physical Routers
How our Cloudy Mindsets Approached Physical Routers
 
Jenkins vs. AWS CodePipeline (AWS User Group Berlin)
Jenkins vs. AWS CodePipeline (AWS User Group Berlin)Jenkins vs. AWS CodePipeline (AWS User Group Berlin)
Jenkins vs. AWS CodePipeline (AWS User Group Berlin)
 
Jenkins vs. AWS CodePipeline
Jenkins vs. AWS CodePipelineJenkins vs. AWS CodePipeline
Jenkins vs. AWS CodePipeline
 
Monitoring Akka with Kamon 1.0
Monitoring Akka with Kamon 1.0Monitoring Akka with Kamon 1.0
Monitoring Akka with Kamon 1.0
 
(Declarative) Jenkins Pipelines
(Declarative) Jenkins Pipelines(Declarative) Jenkins Pipelines
(Declarative) Jenkins Pipelines
 
An Open-Source Chef Cookbook CI/CD Implementation Using Jenkins Pipelines
An Open-Source Chef Cookbook CI/CD Implementation Using Jenkins PipelinesAn Open-Source Chef Cookbook CI/CD Implementation Using Jenkins Pipelines
An Open-Source Chef Cookbook CI/CD Implementation Using Jenkins Pipelines
 
Continuous Delivery
Continuous DeliveryContinuous Delivery
Continuous Delivery
 
Jenkins Pipelines
Jenkins PipelinesJenkins Pipelines
Jenkins Pipelines
 
Let's go HTTPS-only! - More Than Buying a Certificate
Let's go HTTPS-only! - More Than Buying a CertificateLet's go HTTPS-only! - More Than Buying a Certificate
Let's go HTTPS-only! - More Than Buying a Certificate
 
Cleaning Up the Dirt of the Nineties - How New Protocols are Modernizing the Web
Cleaning Up the Dirt of the Nineties - How New Protocols are Modernizing the WebCleaning Up the Dirt of the Nineties - How New Protocols are Modernizing the Web
Cleaning Up the Dirt of the Nineties - How New Protocols are Modernizing the Web
 
Git Power-Workshop
Git Power-WorkshopGit Power-Workshop
Git Power-Workshop
 
Der Weg zu TYPO3 CMS 6.0 und Einblicke in die TYPO3-Entwicklung
Der Weg zu TYPO3 CMS 6.0 und Einblicke in die TYPO3-EntwicklungDer Weg zu TYPO3 CMS 6.0 und Einblicke in die TYPO3-Entwicklung
Der Weg zu TYPO3 CMS 6.0 und Einblicke in die TYPO3-Entwicklung
 
Official typo3.org infrastructure &
the TYPO3 Server Admin Team
Official typo3.org infrastructure &
the TYPO3 Server Admin TeamOfficial typo3.org infrastructure &
the TYPO3 Server Admin Team
Official typo3.org infrastructure &
the TYPO3 Server Admin Team
 
Neuigkeiten aus dem TYPO3-Projekt
Neuigkeiten aus dem TYPO3-ProjektNeuigkeiten aus dem TYPO3-Projekt
Neuigkeiten aus dem TYPO3-Projekt
 
The TYPO3 Server Admin Team
The TYPO3 Server Admin TeamThe TYPO3 Server Admin Team
The TYPO3 Server Admin Team
 
Gerrit Workshop
Gerrit WorkshopGerrit Workshop
Gerrit Workshop
 

Kürzlich hochgeladen

Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piececharlottematthew16
 
What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024Stephanie Beckett
 
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Mark Simos
 
"Federated learning: out of reach no matter how close",Oleksandr Lapshyn
"Federated learning: out of reach no matter how close",Oleksandr Lapshyn"Federated learning: out of reach no matter how close",Oleksandr Lapshyn
"Federated learning: out of reach no matter how close",Oleksandr LapshynFwdays
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsMark Billinghurst
 
DevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenDevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenHervé Boutemy
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machinePadma Pradeep
 
Vector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesVector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesZilliz
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLScyllaDB
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Wonjun Hwang
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationSafe Software
 
Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyAlfredo García Lavilla
 
The Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfThe Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfSeasiaInfotech2
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationRidwan Fadjar
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupFlorian Wilhelm
 
"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr BaganFwdays
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii SoldatenkoFwdays
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxhariprasad279825
 
Vertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsVertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsMiki Katsuragi
 
Gen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfGen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfAddepto
 

Kürzlich hochgeladen (20)

Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piece
 
What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024
 
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
 
"Federated learning: out of reach no matter how close",Oleksandr Lapshyn
"Federated learning: out of reach no matter how close",Oleksandr Lapshyn"Federated learning: out of reach no matter how close",Oleksandr Lapshyn
"Federated learning: out of reach no matter how close",Oleksandr Lapshyn
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
 
DevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenDevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache Maven
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
 
Vector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesVector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector Databases
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQL
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
 
Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easy
 
The Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfThe Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdf
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project Setup
 
"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptx
 
Vertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsVertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering Tips
 
Gen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfGen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdf
 

Investigating the Impact of Network Topology on the Processing Times of SDN Controllers

  • 1. Institute  of  Computer  Science Chair  of  Communication  Networks Prof.  Dr.-­Ing.  P.  Tran-­Gia Investigating  the  Impact  of  Network  Topology   on  the  Processing  Times  of  SDN  Controllers Christopher  Metter,  Steffen  Gebert,  Stanislav  Lange,   Thomas  Zinner,  Phuoc Tran-­Gia,  Michael  Jarschel* www3.informatik.uni-­wuerzburg.de * Nokia GmbH, Munich
  • 2. Steffen  Gebert 2Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers Metadata u This  paper  was  presented  at  the  “The  Seventh  IFIP/IEEE   International  Workshop  on  Management  of  the  Future  Internet”   (ManFI),  colocated with  IM`2015  in  Ottawa,  Canada. u http://manfi.org u Paper  author  version: http://www3.informatik.uni-­wuerzburg.de/staff/steffen.gebert/author_versions/conf_510_author_versions.pdf
  • 3. Steffen  Gebert 3Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers Motivation  – SDN  Architecture Data PlaneData Plane C D C D Control Plane
  • 4. Steffen  Gebert 4Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers Impact  of  Controller  Implementations  on  Message  Processing  Times u Analysis  of  processing  times  for  a  L2  learning  switch § Switching  modules  of  OpenDayLight (ODL)  and  Ryu § Packet-­in/packet-­out  pairs – e.g.,  in  case  of  unconnected  switches  or  requests  for  non-­reachable   destinations   § Packet-­in-­packet-­out/flow-­mod  pairs – e.g.,  to  set  a  forwarding  path  in  the  flow  table  to  a  certain  destination   u Topology  emulation  and  packet-­in  traffic  generation § Controlled,  low-­rate  traffic  emulation  to  reduce  waiting  times § Impact  of  network  topology  on  the  service  times   – Unconnected  topology  (baseline) – Fat-­tree  topology  (data  center) – WAN  topologies  (261  topologies  from  the  Internet  Topology  Zoo)
  • 5. Steffen  Gebert 5Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers OFCProbe github.com/lsinfo3/ofcprobe u Controller  performance  analysis  tool § Emulation  of  virtual  OpenFlow-­enabled  switches § Generate  OF  messages  per  switch § PCAP  file  playback § Configurable  payload  IAT  distributions § Per-­switch  statistics  collection C
  • 6. Steffen  Gebert 6Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers Topology Discovery u Switches  are configured to connect to the controller u Controller  has no information about links  between switches ?? ? 1 2 3 C
  • 7. Steffen  Gebert 7Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers ? Topology Discovery u Controller  injects LLDP  packets (Link  Layer  Discovery  Protocol) u Switches  flood them to all  ports ? LLDP LLDP PacketOut (flood) LLDP ? 1 2 3 C
  • 8. Steffen  Gebert 8Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers C Topology Discovery u Other  switches receive packets and  report packet-­in  to controller u Controller  learns topology from information about incoming ports ? PacketOut (flood) LLDP PacketIn (port 7) LLDP PacketIn (port 2) LLDP Switch Port To Switch 2 2 1 3 7 1 1 2 3 2 7
  • 9. Steffen  Gebert 9Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers Processing  Steps of  a  Controller  for Packet-­Ins 1. Data  reception 2. Decoding  of  OpenFlow message  and  classification 3. Attribute  extraction  and  delegation  to  application  module 4. Module-­specific  processing   5. Response  generation  (e.g.,  packet-­out  /  flow-­mod  message  …) 6. Data  transmission Impact  of  controller  implementation  on  module-­specific  processing?
  • 10. Steffen  Gebert 10Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers Controller C Baseline  Topology u 20  switches,  1  host  per  switch,  no  connections  between  switches u Resulting  action  in  packet-­out  message:  flood
  • 11. Steffen  Gebert 11Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers Baseline  Performance u Emulated switches generate 16  packet-­ins  one after  another u Controller  responds with packet-­out  message u Periodic behavior in  response times (per  switch) u Higher  variation of  processing times for ODL  (JVM?) u ODL  outperforms Ryu
  • 12. Steffen  Gebert 12Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers C u 20  switches,  represents data center topology Fat-­Tree
  • 13. Steffen  Gebert 13Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers Packet-­Out:  Fat-­Tree u Bursts  of 16  packets are sent from one host  to another u Increased processing times compared to unconnected switches u Similar average processing times,  but  high  variation and outliers when using ODL
  • 14. Steffen  Gebert 14Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers Flow-­Mod:  Fat-­Tree u Flow-­mods by utilizing Layer  2  learning switch module u Processing  times for flow-­mods 2-­8  times higher than packet-­outs u ODL  processing times stabilize after  ~20  seconds u Ryu has nearly constant processing times and outperforms ODL
  • 15. Steffen  Gebert 15Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers u Experiment  performed on  261  WAN  topologies u CDF  shows all  measured processing times u Ryu:  Topology has no effect on  processing times,  flow-­mod processing slightly slower than packet-­out  processing u ODL:  High  variation for packet-­out,  flow-­mods significantly longer Internet  Topology Zoo  – Aggregated Results
  • 16. Steffen  Gebert 16Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers Topology  Investigation u Analysis  of  the  topology  impact  on  the  processing  times u Ryu:  rather  constant  behavior,  similar  processing  times  for  flow-­ mods  and  packet-­outs u ODL:  high  impact  of  message  type,  varying  processing  times  
  • 17. Steffen  Gebert 17Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers u Investigation  of  controller  processing  times § Emulated  different  network  topologies  using  OFCProbe § Time  measurements  done  directly  within  the  controllers § Using Layer  2  switch modules (as starting point) u Benchmarking  results: § Ryu shows  very  constant  response  times § ODL  faster  for  packet-­out,  Ryu for  flow-­mod u Main  take-­aways: § Test  your  own  forwarding  algorithms  using  different  topologies! § OFCProbe now  allows  importing  GraphML topologies § Measurement  probes  available  for  ODL  and  Ryu controllers1 1 source code on  github.com/lsinfo3  (manfi2015-­perf-­*  repos) Summary
  • 18. Steffen  Gebert 18Impact  of  Network  Topology  on  the  Processing  Times  of  SDN  Controllers OFCProbe Architecture OF  Controller Best  Effort   Traffic  Generator OpenFlow  Connection  Handler  Module Configuration Communication   Layer Traffic  Generation   Layer Traffic  Generation  Module Statistics Statistics:   PPS  Module Statistics:  RTT   Module Statistics:   Custom   Module Interface Configuration Packets Statistics Implementation Payload OF Message OF Reply OF ReplyOF Reply