SlideShare uma empresa Scribd logo
1 de 21
Baixar para ler offline
ESPRES:
Easy Scheduling and
Prioritization for SDN
Peter Perešíni @ EPFL
Maciej Kuzniar @ EPFL, Marco Canini @ UCLouvain,
Dejan Kostić @ IMDEA Networks
Big Networks = Big Updates
Topology changes
Traffic Engineering
Policy changes
2
Network updates touch many flows
Meanwhile:
Control plane is not fast enough
Big Updates Take Time
Big updates (many rule changes)
+
Slow rule installation speed
=
Updates take a long time
3
Our Work - ESPRES
• Transparent layer between ctrl and switches
• Goal = optimize big update installation
o Complete installing majority of flows sooner
o Decrease mid-update switch rule overhead
o Minimize flow disruption times
4
Challenges
5
2
3
54
Switch message queue
1
Flow update requires all
its rules to complete
1.
Dependencies might
hold rule installation
2.
Cannot reorder
message queue
3.
Switches have different
rule installation speeds
Ingress rule depends on
core being installed
ingress
ESPRES Design
Two key ideas:
Manage switch message queue
Schedule rule installations carefully
6
Managing Switch Message Queue
Switch message queue In-ESPRES “queue”
● Short
● Fixed order
● Arbitrary long
● Flexible ordering
7
Managing Switch Message Queue
When switch message queue has room
1. Determine ready operations
2. Send according to the scheduler
8
Scheduling from 10,000 feet
Main highlights:
1. Group operations per flow
2. Order flows
9
Grouping Rule Installations
Arbitrary rule installation Grouping rules by flows
Switch 1:
Switch 2:
Flow completion
determined by last op
Group related operations and schedule
them in parallel
10
Time Time
TimeTime
Ordering Flows
4 + 5
avg=4.5
1 + 5
avg=3
11
Schedule 1: Schedule 2:
Finish times:
Ordering makes a difference
Different goals ⇒ different ordering
• Finishing flows sooner ⇒ Shorter updates first
• Reduce rule overhead ⇒ Updates removing rules
first
Evaluation
• ESPRES prototype in Python
• Use Mininet to emulate network
o Patch Reference Switch implementation ---
Throttle to 40 updates/sec [1]
[1] Danny Yuxing Huang, Kenneth Yocum, and Alex C. Snoeren.
High-Fidelity Switch Models for Software-Defined Network Emulation.
In HotSDN 2013.
12
1000 new flows; 18 switches (IBM backbone topo)
Finishing Majority of Flows Sooner
We reduce completion time
for ½ of flows by ≥40%
ESPRES near optimal
schedule (ILP)
13
ESPRES Summary
• Big updates + slow rule installations need
update installation strategy
• ESPRES based on two ideas
o Managing switch message queue
o Scheduling rule installation
Schedule, you should!
14
Backup slides
Is next switch generation fine?
PICA8
• circa 1000 flows/sec
• “4000 flows not enough for some clients”
• announced P-3297
o 8000 flows
o if performance similar to previous models:
➢ at least 8-10s to install full flow table
Is next switch generation fine?
• Big flow churn may still require scheduling
Work in progress
• Stream of big network updates
o some are high-priority (failure recovery)
Evaluation: Lowering Rule Overhead
FatTree (20 switches), update 370 flows,
use consistent update (2-phase commit)
Method Max per-switch rule overhead
Naive 76%
Incremental Consistent Updates [1] 16.5% (for 4 rounds)
ESPRES/overhead 10.2%
Dependencies
sw 1 sw 2 sw 3 h 2h 1
Fwd:
h2
Single flow update touches
many switches
Inter-switch dependencies
(e.g. core before ingress)
Fwd:
sw 3
Fwd:
sw 2

Mais conteúdo relacionado

Mais procurados

Real time network monitoring
Real time network monitoringReal time network monitoring
Real time network monitoring
Sumit Rajpal
 
CopperLan broadcast
CopperLan broadcastCopperLan broadcast
CopperLan broadcast
dherscovitch
 
Ric presentation
Ric presentationRic presentation
Ric presentation
IvyKuo1
 
REMOWZ - Realtime Water Quality Monitoring using ZigBee based WSN (Part I)
REMOWZ - Realtime Water Quality Monitoring using ZigBee based WSN (Part I)REMOWZ - Realtime Water Quality Monitoring using ZigBee based WSN (Part I)
REMOWZ - Realtime Water Quality Monitoring using ZigBee based WSN (Part I)
Nitin Balakrishnan
 

Mais procurados (20)

The Role of Inter-Controller Traffic in SDN Controllers Placement
The Role of Inter-Controller Traffic in SDN Controllers PlacementThe Role of Inter-Controller Traffic in SDN Controllers Placement
The Role of Inter-Controller Traffic in SDN Controllers Placement
 
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
 
Real time network monitoring
Real time network monitoringReal time network monitoring
Real time network monitoring
 
Free training on NCM - Discovery & Disaster recovery
Free training on NCM - Discovery & Disaster recovery Free training on NCM - Discovery & Disaster recovery
Free training on NCM - Discovery & Disaster recovery
 
Understanding firewall-policies-their-effectiveness-in-defending-against-netw...
Understanding firewall-policies-their-effectiveness-in-defending-against-netw...Understanding firewall-policies-their-effectiveness-in-defending-against-netw...
Understanding firewall-policies-their-effectiveness-in-defending-against-netw...
 
Free OpManager training_Part 1- Discovery & classification
Free OpManager training_Part 1- Discovery & classificationFree OpManager training_Part 1- Discovery & classification
Free OpManager training_Part 1- Discovery & classification
 
Free OpManager training_ Part 2-server monitoring
Free OpManager training_ Part 2-server monitoringFree OpManager training_ Part 2-server monitoring
Free OpManager training_ Part 2-server monitoring
 
Monitoring network performance- Part 3_Free OpManager training
Monitoring network performance- Part 3_Free OpManager training Monitoring network performance- Part 3_Free OpManager training
Monitoring network performance- Part 3_Free OpManager training
 
Season 4 [Free OpManager training] Part1- Discovery and classification
Season 4 [Free OpManager training] Part1- Discovery and classificationSeason 4 [Free OpManager training] Part1- Discovery and classification
Season 4 [Free OpManager training] Part1- Discovery and classification
 
2015 COSCUP SDN Workshop -- SDN Quick Start
2015 COSCUP SDN Workshop -- SDN Quick Start2015 COSCUP SDN Workshop -- SDN Quick Start
2015 COSCUP SDN Workshop -- SDN Quick Start
 
Ncm Middle East Workshop
Ncm Middle East WorkshopNcm Middle East Workshop
Ncm Middle East Workshop
 
CopperLan broadcast
CopperLan broadcastCopperLan broadcast
CopperLan broadcast
 
Network entry success rate
Network entry success rateNetwork entry success rate
Network entry success rate
 
Ric presentation
Ric presentationRic presentation
Ric presentation
 
Network Maps & Reporting [Free OpManager Training - Part 5]
Network Maps & Reporting [Free OpManager Training - Part 5]Network Maps & Reporting [Free OpManager Training - Part 5]
Network Maps & Reporting [Free OpManager Training - Part 5]
 
OWF12/Open Standards for Cloud - Cs owf
OWF12/Open Standards for Cloud - Cs owfOWF12/Open Standards for Cloud - Cs owf
OWF12/Open Standards for Cloud - Cs owf
 
REMOWZ - Realtime Water Quality Monitoring using ZigBee based WSN (Part I)
REMOWZ - Realtime Water Quality Monitoring using ZigBee based WSN (Part I)REMOWZ - Realtime Water Quality Monitoring using ZigBee based WSN (Part I)
REMOWZ - Realtime Water Quality Monitoring using ZigBee based WSN (Part I)
 
Best Network Performance Monitoring Tool
Best Network Performance Monitoring ToolBest Network Performance Monitoring Tool
Best Network Performance Monitoring Tool
 
Academic Experiment 2 - IP forwarding
Academic Experiment 2 - IP forwarding Academic Experiment 2 - IP forwarding
Academic Experiment 2 - IP forwarding
 
IT307 Exploring the Networks.pptx
IT307 Exploring the Networks.pptxIT307 Exploring the Networks.pptx
IT307 Exploring the Networks.pptx
 

Destaque

Cloud computing for enterprise
Cloud computing for enterpriseCloud computing for enterprise
Cloud computing for enterprise
Pravin Asar
 

Destaque (7)

Software Defined Networking: Enabling The Mobile Workplace
Software Defined Networking: Enabling The Mobile WorkplaceSoftware Defined Networking: Enabling The Mobile Workplace
Software Defined Networking: Enabling The Mobile Workplace
 
Control Exchange Points: Providing QoS-en abled End-to-End Services via SDN-b...
Control Exchange Points: Providing QoS-en abled End-to-End Services via SDN-b...Control Exchange Points: Providing QoS-en abled End-to-End Services via SDN-b...
Control Exchange Points: Providing QoS-en abled End-to-End Services via SDN-b...
 
Cloud computing for enterprise
Cloud computing for enterpriseCloud computing for enterprise
Cloud computing for enterprise
 
Learnings from Carrier SDN Deployments
Learnings from Carrier SDN DeploymentsLearnings from Carrier SDN Deployments
Learnings from Carrier SDN Deployments
 
Introduction to CORD project
Introduction to CORD projectIntroduction to CORD project
Introduction to CORD project
 
OPNFV Webinar – No Time to Wait: Accelerating NFV Time to Market Through Open...
OPNFV Webinar – No Time to Wait: Accelerating NFV Time to Market Through Open...OPNFV Webinar – No Time to Wait: Accelerating NFV Time to Market Through Open...
OPNFV Webinar – No Time to Wait: Accelerating NFV Time to Market Through Open...
 
CORD: Central Office Re-architected as a Datacenter
CORD: Central Office Re-architected as a DatacenterCORD: Central Office Re-architected as a Datacenter
CORD: Central Office Re-architected as a Datacenter
 

Semelhante a ESPRES: Easy Scheduling and Prioritization for SDN

Using OpenStack In a Traditional Hosting Environment
Using OpenStack In a Traditional Hosting EnvironmentUsing OpenStack In a Traditional Hosting Environment
Using OpenStack In a Traditional Hosting Environment
OpenStack Foundation
 
Data Center Network Trends - Lin Nease
Data Center Network Trends - Lin NeaseData Center Network Trends - Lin Nease
Data Center Network Trends - Lin Nease
HPDutchWorld
 
RedisConf18 - Application of Redis in IOT Edge Devices
RedisConf18 - Application of Redis in IOT Edge DevicesRedisConf18 - Application of Redis in IOT Edge Devices
RedisConf18 - Application of Redis in IOT Edge Devices
Redis Labs
 
Linac Coherent Light Source (LCLS) Data Transfer Requirements
Linac Coherent Light Source (LCLS) Data Transfer RequirementsLinac Coherent Light Source (LCLS) Data Transfer Requirements
Linac Coherent Light Source (LCLS) Data Transfer Requirements
inside-BigData.com
 
Transforming a traditional home gateway into a hardwareaccelerated SDN switch
Transforming a traditional home gateway into a hardwareaccelerated SDN switchTransforming a traditional home gateway into a hardwareaccelerated SDN switch
Transforming a traditional home gateway into a hardwareaccelerated SDN switch
IJECEIAES
 
Four Ways to Improve Linux Performance IEEE Webinar, R2.0
Four Ways to Improve Linux Performance IEEE Webinar, R2.0Four Ways to Improve Linux Performance IEEE Webinar, R2.0
Four Ways to Improve Linux Performance IEEE Webinar, R2.0
Michael Christofferson
 
Remote core locking-Andrea Lombardo
Remote core locking-Andrea LombardoRemote core locking-Andrea Lombardo
Remote core locking-Andrea Lombardo
Andrea Lombardo
 
참여기관_발표자료-국민대학교 201301 정기회의
참여기관_발표자료-국민대학교 201301 정기회의참여기관_발표자료-국민대학교 201301 정기회의
참여기관_발표자료-국민대학교 201301 정기회의
DzH QWuynh
 

Semelhante a ESPRES: Easy Scheduling and Prioritization for SDN (20)

Traffic Engineering in Software-Defined Networks
Traffic Engineering in Software-Defined NetworksTraffic Engineering in Software-Defined Networks
Traffic Engineering in Software-Defined Networks
 
Using OpenStack In a Traditional Hosting Environment
Using OpenStack In a Traditional Hosting EnvironmentUsing OpenStack In a Traditional Hosting Environment
Using OpenStack In a Traditional Hosting Environment
 
Fast switching of threads between cores - Advanced Operating Systems
Fast switching of threads between cores - Advanced Operating SystemsFast switching of threads between cores - Advanced Operating Systems
Fast switching of threads between cores - Advanced Operating Systems
 
Hyper Threading technology
Hyper Threading technologyHyper Threading technology
Hyper Threading technology
 
Data Center Network Trends - Lin Nease
Data Center Network Trends - Lin NeaseData Center Network Trends - Lin Nease
Data Center Network Trends - Lin Nease
 
RedisConf18 - Application of Redis in IOT Edge Devices
RedisConf18 - Application of Redis in IOT Edge DevicesRedisConf18 - Application of Redis in IOT Edge Devices
RedisConf18 - Application of Redis in IOT Edge Devices
 
Linac Coherent Light Source (LCLS) Data Transfer Requirements
Linac Coherent Light Source (LCLS) Data Transfer RequirementsLinac Coherent Light Source (LCLS) Data Transfer Requirements
Linac Coherent Light Source (LCLS) Data Transfer Requirements
 
lect4_SDNbasic_openflow.pptx
lect4_SDNbasic_openflow.pptxlect4_SDNbasic_openflow.pptx
lect4_SDNbasic_openflow.pptx
 
4_SDN.pdf
4_SDN.pdf4_SDN.pdf
4_SDN.pdf
 
Software defined network
Software defined network Software defined network
Software defined network
 
Transforming a traditional home gateway into a hardwareaccelerated SDN switch
Transforming a traditional home gateway into a hardwareaccelerated SDN switchTransforming a traditional home gateway into a hardwareaccelerated SDN switch
Transforming a traditional home gateway into a hardwareaccelerated SDN switch
 
L-2 (Computer Performance).ppt
L-2 (Computer Performance).pptL-2 (Computer Performance).ppt
L-2 (Computer Performance).ppt
 
How we scaled Rudder to 10k, and the road to 50k
How we scaled Rudder to 10k, and the road to 50kHow we scaled Rudder to 10k, and the road to 50k
How we scaled Rudder to 10k, and the road to 50k
 
Insider operating system
Insider   operating systemInsider   operating system
Insider operating system
 
Four Ways to Improve Linux Performance IEEE Webinar, R2.0
Four Ways to Improve Linux Performance IEEE Webinar, R2.0Four Ways to Improve Linux Performance IEEE Webinar, R2.0
Four Ways to Improve Linux Performance IEEE Webinar, R2.0
 
Remote core locking-Andrea Lombardo
Remote core locking-Andrea LombardoRemote core locking-Andrea Lombardo
Remote core locking-Andrea Lombardo
 
The Challenges of SDN/OpenFlow in an Operational and Large-scale Network
The Challenges of SDN/OpenFlow in an Operational and Large-scale NetworkThe Challenges of SDN/OpenFlow in an Operational and Large-scale Network
The Challenges of SDN/OpenFlow in an Operational and Large-scale Network
 
SDN Introduction
SDN IntroductionSDN Introduction
SDN Introduction
 
tcp.pptx
tcp.pptxtcp.pptx
tcp.pptx
 
참여기관_발표자료-국민대학교 201301 정기회의
참여기관_발표자료-국민대학교 201301 정기회의참여기관_발표자료-국민대학교 201301 정기회의
참여기관_발표자료-국민대학교 201301 정기회의
 

Mais de Open Networking Summits

[Webinar Slides] Programming the Network Dataplane in P4
[Webinar Slides] Programming the Network Dataplane in P4[Webinar Slides] Programming the Network Dataplane in P4
[Webinar Slides] Programming the Network Dataplane in P4
Open Networking Summits
 

Mais de Open Networking Summits (20)

[Webinar Slides] Programming the Network Dataplane in P4
[Webinar Slides] Programming the Network Dataplane in P4[Webinar Slides] Programming the Network Dataplane in P4
[Webinar Slides] Programming the Network Dataplane in P4
 
Application Driven SDN
Application Driven SDNApplication Driven SDN
Application Driven SDN
 
Software Defined Networks Network Function Virtualization Pivotal Technologies
Software Defined Networks Network Function Virtualization Pivotal TechnologiesSoftware Defined Networks Network Function Virtualization Pivotal Technologies
Software Defined Networks Network Function Virtualization Pivotal Technologies
 
NFV & SDN Customer Deployments
NFV & SDN Customer DeploymentsNFV & SDN Customer Deployments
NFV & SDN Customer Deployments
 
Automation of end-to-end QOS
Automation of end-to-end QOSAutomation of end-to-end QOS
Automation of end-to-end QOS
 
Building a Digital Telco
Building a Digital TelcoBuilding a Digital Telco
Building a Digital Telco
 
Spreading NFV through the Network: the ETSI NFV use cases
Spreading NFV through the Network: the ETSI NFV use casesSpreading NFV through the Network: the ETSI NFV use cases
Spreading NFV through the Network: the ETSI NFV use cases
 
BeHop : SDN for Dense WiFi Networks
BeHop : SDN for Dense WiFi NetworksBeHop : SDN for Dense WiFi Networks
BeHop : SDN for Dense WiFi Networks
 
Ranges & Cross-Entrance Consistency with OpenFlow
Ranges & Cross-Entrance Consistency with OpenFlowRanges & Cross-Entrance Consistency with OpenFlow
Ranges & Cross-Entrance Consistency with OpenFlow
 
On the Necessity of Time-based Updates in SDN
On the Necessity of Time-based Updates in SDNOn the Necessity of Time-based Updates in SDN
On the Necessity of Time-based Updates in SDN
 
SDN & OPTICAL FLOW STEERING FOR NETWORK FUNCTION VIRTUALIZATION
SDN & OPTICAL FLOW STEERING FOR NETWORK FUNCTION VIRTUALIZATIONSDN & OPTICAL FLOW STEERING FOR NETWORK FUNCTION VIRTUALIZATION
SDN & OPTICAL FLOW STEERING FOR NETWORK FUNCTION VIRTUALIZATION
 
SoftMoW: A Dynamic and Scalable Software Defined Architecture for Cellular WANs
SoftMoW: A Dynamic and Scalable Software Defined Architecture for Cellular WANsSoftMoW: A Dynamic and Scalable Software Defined Architecture for Cellular WANs
SoftMoW: A Dynamic and Scalable Software Defined Architecture for Cellular WANs
 
RadioVisor - A Slicing Plane for Radio Access Networks
RadioVisor - A Slicing Plane for Radio Access NetworksRadioVisor - A Slicing Plane for Radio Access Networks
RadioVisor - A Slicing Plane for Radio Access Networks
 
Enabling SDN in old school networks with Software-Controlled Routing Protocols
Enabling SDN in old school networks with Software-Controlled Routing ProtocolsEnabling SDN in old school networks with Software-Controlled Routing Protocols
Enabling SDN in old school networks with Software-Controlled Routing Protocols
 
Accelerating SDN/NFV with transparent offloading architecture
Accelerating SDN/NFV with transparent offloading architectureAccelerating SDN/NFV with transparent offloading architecture
Accelerating SDN/NFV with transparent offloading architecture
 
Serial Composition of Heterogeneous Control Planes
Serial Composition of Heterogeneous Control PlanesSerial Composition of Heterogeneous Control Planes
Serial Composition of Heterogeneous Control Planes
 
SFA: Stateful Forwarding Abstraction in SDN Data Plane
SFA: Stateful Forwarding Abstraction in SDN Data PlaneSFA: Stateful Forwarding Abstraction in SDN Data Plane
SFA: Stateful Forwarding Abstraction in SDN Data Plane
 
Towards a Reliable SDN Firewall
Towards a Reliable SDN FirewallTowards a Reliable SDN Firewall
Towards a Reliable SDN Firewall
 
Coherent SDN Forwarding Plane Programming
Coherent SDN Forwarding Plane ProgrammingCoherent SDN Forwarding Plane Programming
Coherent SDN Forwarding Plane Programming
 
Proof-based Verification of Software Defined Networks
Proof-based Verification of Software Defined NetworksProof-based Verification of Software Defined Networks
Proof-based Verification of Software Defined Networks
 

Último

Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
WSO2
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
?#DUbAI#??##{{(☎️+971_581248768%)**%*]'#abortion pills for sale in dubai@
 
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Victor Rentea
 

Último (20)

Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Exploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusExploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with Milvus
 
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
 
Platformless Horizons for Digital Adaptability
Platformless Horizons for Digital AdaptabilityPlatformless Horizons for Digital Adaptability
Platformless Horizons for Digital Adaptability
 
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
 
WSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering DevelopersWSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering Developers
 
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf
 
Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..
 
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
 
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In PakistanCNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistan
 
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...
 

ESPRES: Easy Scheduling and Prioritization for SDN

  • 1.
  • 2.
  • 3. ESPRES: Easy Scheduling and Prioritization for SDN Peter Perešíni @ EPFL Maciej Kuzniar @ EPFL, Marco Canini @ UCLouvain, Dejan Kostić @ IMDEA Networks
  • 4. Big Networks = Big Updates Topology changes Traffic Engineering Policy changes 2 Network updates touch many flows
  • 5. Meanwhile: Control plane is not fast enough Big Updates Take Time Big updates (many rule changes) + Slow rule installation speed = Updates take a long time 3
  • 6. Our Work - ESPRES • Transparent layer between ctrl and switches • Goal = optimize big update installation o Complete installing majority of flows sooner o Decrease mid-update switch rule overhead o Minimize flow disruption times 4
  • 7. Challenges 5 2 3 54 Switch message queue 1 Flow update requires all its rules to complete 1. Dependencies might hold rule installation 2. Cannot reorder message queue 3. Switches have different rule installation speeds Ingress rule depends on core being installed ingress
  • 8. ESPRES Design Two key ideas: Manage switch message queue Schedule rule installations carefully 6
  • 9. Managing Switch Message Queue Switch message queue In-ESPRES “queue” ● Short ● Fixed order ● Arbitrary long ● Flexible ordering 7
  • 10. Managing Switch Message Queue When switch message queue has room 1. Determine ready operations 2. Send according to the scheduler 8
  • 11. Scheduling from 10,000 feet Main highlights: 1. Group operations per flow 2. Order flows 9
  • 12. Grouping Rule Installations Arbitrary rule installation Grouping rules by flows Switch 1: Switch 2: Flow completion determined by last op Group related operations and schedule them in parallel 10 Time Time
  • 13. TimeTime Ordering Flows 4 + 5 avg=4.5 1 + 5 avg=3 11 Schedule 1: Schedule 2: Finish times: Ordering makes a difference Different goals ⇒ different ordering • Finishing flows sooner ⇒ Shorter updates first • Reduce rule overhead ⇒ Updates removing rules first
  • 14. Evaluation • ESPRES prototype in Python • Use Mininet to emulate network o Patch Reference Switch implementation --- Throttle to 40 updates/sec [1] [1] Danny Yuxing Huang, Kenneth Yocum, and Alex C. Snoeren. High-Fidelity Switch Models for Software-Defined Network Emulation. In HotSDN 2013. 12
  • 15. 1000 new flows; 18 switches (IBM backbone topo) Finishing Majority of Flows Sooner We reduce completion time for ½ of flows by ≥40% ESPRES near optimal schedule (ILP) 13
  • 16. ESPRES Summary • Big updates + slow rule installations need update installation strategy • ESPRES based on two ideas o Managing switch message queue o Scheduling rule installation Schedule, you should! 14
  • 18. Is next switch generation fine? PICA8 • circa 1000 flows/sec • “4000 flows not enough for some clients” • announced P-3297 o 8000 flows o if performance similar to previous models: ➢ at least 8-10s to install full flow table
  • 19. Is next switch generation fine? • Big flow churn may still require scheduling Work in progress • Stream of big network updates o some are high-priority (failure recovery)
  • 20. Evaluation: Lowering Rule Overhead FatTree (20 switches), update 370 flows, use consistent update (2-phase commit) Method Max per-switch rule overhead Naive 76% Incremental Consistent Updates [1] 16.5% (for 4 rounds) ESPRES/overhead 10.2%
  • 21. Dependencies sw 1 sw 2 sw 3 h 2h 1 Fwd: h2 Single flow update touches many switches Inter-switch dependencies (e.g. core before ingress) Fwd: sw 3 Fwd: sw 2