SlideShare a Scribd company logo
Dragonflow
TLV OpenStack meet-up
Eran Gampel
Gal Sagie
What is Dragonflow?
Full Implementation of OpenStack Neutron API
Lightweight Distributed SDN Controller with pluggable
database
Project mission
To Implement advanced networking services in a manner
that is efficient, elegant and resource-nimble
Page 2
Dragonflow Highlights
Page 3
• Integral part of OpenStack
• Fully Open Source
• Scale, Performance and Latency
• Lightweight and Simple
• Easily Extendable
• Distributed SDN Control Plane
• Sync Policy Level abstraction to the CN
Dragonflow - Distributed SDN
Neutron-Server
Dragonflow Plugin
DB
OVS
Dragonflow
DB
Driver
Compute Node
OVS
Dragonflow
DB
Driver
Compute Node
OVS
Dragonflow
DB
Driver
Compute Node
OVS
Dragonflow
DB
Driver
Compute Node
DB
VM VM
..
VM VM
..
VM VM
.. VM VM
..
Compute NodeCompute NodeCompute Node
Dragonflow
Network DB
OVS
Neutron
Server
OVSDB
OVSDB-Server
ETCD RethinkDBRAMCloud
Kernel Datapath Module
NIC
User Space
Kernel Space
Dragonflow DB Drivers
OVSDB ETCD RethinkDBRMC
Future
Dragonflow Plugin
Route
Core
API
SG
vswitchd
Container
VM Dragonflow Controller
Abstraction Layer
L2 App L3 App DHCP App
Fault
Detection
SG
LBaaS …FWaaS
Pluggable DB
Layer
NBDBDrivers
SB DB Drivers
smartNIC OVSDB
OVSDB
ETCD
RMC
RethinkDB
OpenFlow
Dragonflow – Under The Hood
Current Release Features (Liberty)
L2 core API, IPv4, IPv6
GRE/VxLAN/Geneve tunneling protocols
Distributed L3 Virtual Router
Hybrid proactive + reactive flow installation
North-South traffic is still centralized
Distributed DHCP
(with just 500 lines of code!)
Pluggable Distributed Database
ETCD, RethinkDB, RAMCloud, OVSDB
Dragonflow
Distributed DHCP
Page 7
Network Node
DHCP
namespace
DHCP
namespace
DHCP
namespace
DHCP
namespace
Neutron DHCP Implementation
DHCP
namespace
dnsmasq
DHCP
Agent
Neutron Server
Message Queue
Example
• 100 Tenants
• 3 vNet / tenant
= 300 DHCP Servers
1 VM Send DHCP_DISCOVER
2 Classify Flow as DHCP, Forward to Controller
3 DHCP App sends DHCP_OFFER back to VM
4 VM Send DHCP_REQUEST
5 Classify Flow as DHCP, Forward to Controller
6 DHCP App populates DHCP_OPTIONS from DB/CFG and send
DHCP_ACK
Dragonflow Distributed DHCP
VM DHCP SERVER
1
3
4
6
7
Compute Node
Dragonflow
VM
OVS
VM
1 2
br-int
qvoXXX qvoXXX
OpenFlow
1
4
2
5
7
Dragonflow Controller
Abstraction Layer
L2
App
L3
App
DHCP
App
SG
36
Pluggable DB
Layer
DB
Dragonflow Distributed DHCP
Match:
Broadcast +UDP +S_Port=68 +D_Port=67
Action:
Send to DHCP table
Service Table
DHCP Table
Match: in_port => Action:
Set metadata with port unique key
SEND TO CONTROLLER
(for every local port that its network has DHCP
enabled)
Default:
goto “L2 Lookup Table”
Compute Node
VM
OVS
br-int
qvoXXX
VM
qvoXXX
1 2
Dragonflow
Dragonflow Local Controller
Abstraction Layer
L2
App
L3
App
DHCP
App
SG
DB
OpenFlow
Ingress Port Security
Ingress Classification
Dispatch to Ports
Dragonflow
Pluggable DB
Page 11
Database Framework
Requirements
• HA + Scalability
• Different Environments have different requirements
• Performance, Latency, Scalability, etc.
Why Pluggable?
• Long time to productize
• Mature Open Source alternatives
• Allow us to focus on the networking services only
DB Driver API
Implementations
RAMCloud
ETCD
RethinkDB
Zookeeper
Dragonflow Pluggable Database
Compute NodeCompute NodeCompute Node
Dragonflow
Local
Controller
Pluggable
DB Layer
Applicative
DB Layer
Adapter
DB
Driver
API
Expose DB
Features
Neutron Server
Dragonflow
Neutron
Plugin
DB Operations
Database
Server
DB Adapter
DB Adapter
DB Adapter
Distributed
Database
DB Data 3
DB Data 2
DB Data 1
Full Distribution
Compute Node 1
Dragonflow
Local Cache
OVS
Compute Node N
Dragonflow
OVS
Local Cache
Dragonflow DB Drivers
OVSDB ETCD RethinkDBRMC
DB Data 3
DB Data 2
DB Data 1
DB Data 3
DB Data 2
DB Data 1
Distributed
Database
DB Data 3
DB Data 2
DB Data 1
Selective Proactive Distribution
Compute Node 1
Dragonflow
Local Cache
OVS
DB Data 1
Compute Node N
Dragonflow
OVS
Local Cache
DB Data 3
DB Data 2
Dragonflow DB Drivers
OVSDB ETCD RethinkDBRMC
Selective Proactive Distribution
Compute Node 1
Dragonflow
Local Cache
OVS
Net1 – VM1, VM2
Compute Node 2
Dragonflow
OVS
Local Cache
Net2 – VM3, VM4
VM1 VM2 VM3 VM4
RethinkDB
Net2 – VM3, VM4
Net1 – VM1, VM2
Dragonflow Pipeline
Page 17
Dragonflow Pipeline
Installed in every OVS
Service
Traffic
Classification
Ingress Processing
(NAT, BUM)
ARP DHCP
L2
Lookup
L3
Lookup
DVR
Egress
Dispatching outgoing
traffic to external
nodes or local ports
Ingress
Port
Security
(ARP spoofing , SG, …)
Egress
Port
Security
Egress
Processing
(NAT)
Fully Proactive
Has Reactive Flows to Controller
Security Groups
…
Outgoing from local
port Classification and
tagging
Dispatching Incoming
traffic from external
nodes to local ports
Dragonflow Roadmap
Page 19
Roadmap
 Additional DBs Drivers ZooKeeper, Redis …
Selective Proactive DB
 Hierarchical Port Binding (SDN ToR) move to ML2
 Pluggable Pub/Sub Mechanism
 DB Consistency
 Distributed DNAT
 Security Group
 Containers (Kuryr plugin and nested VM support)
 Topology Service Injection / Service Chaining
 Inter Cloud Connectivity (Border Gateway / L2GW)
 …
Hierarchical Port Binding (SDN ToR)
move to ML2
Rack n
ToR
VLAN
Segmentation
Rack 1
ToR
Rack 2
ToR
Rack 3
ToR
Vxlan
Segmentation
Dargonflow Hierarchical Port Binding (SDN ToR)
Neutron Server
REST API
Neutron Core plugins
ML2
Cisco(Nexus,
N1Kv)
OVN
Morevendor
plugins
Type Drivers Mechanism Drivers
VLAN
GRE
VXLAN
ONOS
Dragonflow
TOR
Neutron Service plugins
DragonflowDB
Rack n
ToR
VLAN
Segmentation
Vxlan
Segmentation
Compute Node
Dragonflow
VM
OVS
VM
br-int
qvoXXX qvoXXX
OpenFlow
Dragonflow Controller
Abstraction Layer
Vlan
L2
App
L3
App
DHCP
App
SG
Pluggable DB
Layer
DBDB
ToRMach
Driver
OpenDayLight
Pluggable Pub/Sub Mechanism
Neutron-Server
Dragonflow Plugin
DB
OVS
Dragonflow
DB
Driver
Compute Node
OVS
Dragonflow
DB
Driver
Compute Node
OVS
Dragonflow
DB
Driver
Compute Node
OVS
Dragonflow
DB
Driver
Compute Node
DB
VM VM
..VM VM
..
VM VM
.. VM VM
..
Pub/Sub
if the DB internally supports
Pub sub then we use it
Pluggable Pub/Sub Mechanism
Neutron-Server
Dragonflow Plugin
DB
OVS
Dragonflow
DB
Driver
Compute Node
OVS
Dragonflow
DB
Driver
Compute Node
OVS
Dragonflow
DB
Driver
Compute Node
OVS
Dragonflow
DB
Driver
Compute Node
DB
VM VM
..
VM VM
..
VM VM
..
VM VM
..
Pub/Sub
Why do we need it ?
 Not all DBs support pub-sub (e.g. RamCloud)
 We need to be able to customize
Performance, Latency, Scalability, etc.
DB Consistency Common Problem to all SDN Solution
SDN Controller
North-bound Interface (REST?)
South-bound Interface (Openflow)
SDN Apps
SDN DB
Neutron
DB
Neutron-server
ML2-Core-Plugin
ML2.Drivers.Mechanism.XXX
Services-Plugin
Service
Network
Neutron API Nova API
CLI / Dashboard (Horizon) / Orchestration Tool (Heat)
HW Switch
Nova
Nova Compute
VM VM
Nova Compute
VM VM
Virtual Switch (OVS?) Virtual Switch (OVS?)
Neutron
Plugin Agent
Neutron
Plugin Agent
Vendor-specific API
Message Queue (AMQP)
Neutron-L3-Agent
Neutron-DHCP-Agent
LoadBalancer
Firewall
VPN
L3Services
TopologyMgr.
OverlayMgr.
Security
Dragonflow DB Consistency
Neutron-Server
Dragonflow Plugin
DB
OVS
Dragonflow
DB
Driver
Compute Node
OVS
Dragonflow
DB
Driver
Compute Node
DB
VM VM
..
VM VM
..
Neutron
DB
 The Neutron DB is the master Database
 Introduce a full-sync diff based mechanism
NDB DDB
 Introduce a virtual transaction mechanism
NDB DDB
 Key DB Requirement from multi production
environments
 Optimized for Read, multiple read request in
very high volume from nova, Horizon …
 Multi Neutron server API running on
different hosts
Neutron-Server
Dragonflow Plugin
DB
Neutron-Server
Dragonflow Plugin
DB
Join the project Dragonflow
• Documentation
https://wiki.openstack.org/wiki/Dragonflow
• Bugs & blueprints
https://launchpad.net/dragonflow
• DF IRC channel
#openstack-dragonflow
Weekly on Monday at 0900 UTC in #openstack-meeting-4 (IRC)
Break
10:30 - 11:00 - Break and Networking
Security Groups in Neutron
Security Groups Problems
• Data plane performance
• Additional Linux Bridge on the Path
• Control plane performance
• Rules needs to be re-compiled on port changes
• Many rules due to security group capabilities
• Iptable commands issued by CLI process
• RPC bulks
Security Groups in Dragonflow
Security Groups Translations
 Direction:Egress, Type:IPv4, IP Protocol:TCP, Port Range:Any, Remote IP
Prefix:0.0.0.0/0
 match:ct_state=+new+trk,tcp,reg6=X
actions=ct(commit,zone=NXM_NX_REG6[0..15]),resubmit(,<next_table>)
 Direction:Egress, Type:IPv4, IP Protocol:TCP, Port Range:Any, Remote
Security Group: Y
 match:ct_state=+new+trk,tcp,reg6=X,reg5=Y,
actions=ct(commit,zone=NXM_NX_REG6[0..15]),resubmit(,<next_table>)
Distributed DNAT (Floating IP)
OVS
VM
Compute Node
Public
network
OVS
VM
Compute Node
Public
network
OVS
Network Node
Router
Namespace
Dragonflow and Containers
Neutron and libnetwork
A Docker
Container
Network
Sandbox
Endpoint
A Docker Container
Network Sandbox
Endpoint
A Docker
Container
Network
Sandbox
Endpoint
Frontend
Network
Endpoint
Backend
Network
Tenant A Net1
192.168.1.0/0
Tenant A Net2
192.168.5.0/0
VM1
192.168.1.5
VM2
192.168.1.7
192.168.5.2
Kuryr Project Overview
• Open source
• Part of OpenStack Neutron’s big stadium
• Under OpenStack big tent from next release!!!
• Brings the Neutron networking model as a provider for the Docker
CNM
• Aims to support different Container Orchestration Engines
• E.g. Kubernetes, Mesos, Docker Swarm
• Weekly IRC meetings
• Working together with OpenStack community
• Neutron, Magnum, Kolla
Kuryr And Dragonflow
Dragonflow and Kuryr plans
• Dragonflow to support containers networking use cases
• Nested containers inside VMs support
• Containers can leverage all of Dragonflow features
• Distributed DHCP
• Security and QoS
• Containers performance and fault management
• Port forwarding
• Dragonflow distributed load balancer
• DNS as a Service in Dragonflow
• Integration with Kubernetes
• Full Integration of Dragonflow and Kuryr
• Containerized image of Dragonflow
• VIF Binding to Dragonflow
• OVS, IPVLAN
Mixed OpenStack Environments
Neutron network 1
Neutron network 2
Neutron network 3
Compute Node
VM
Dragonflow OVS(Controller: Dragonflow)
IPVLAN / OVS
VM
Dragonflow Service Injection
Dragonflow Pipeline
Tenant/Admin Added Services
Dynamic
Routing
VPN Firewall
DPI QoS LB
Service
Discovery
Inter
Cloud
Public Network
10.50.50.0/24
Router Router
Tenant A Net1
192.168.1.0/0
Tenant A Net2
192.168.5.0/0
Tenant B Net1
192.168.1.0/0
Tenant B Net2
192.168.9.0/0
VM1
192.168.1.5
VM2
192.168.1.7
192.168.5.2
VM3
192.168.1.9
VM1
192.168.1.3
VM2
192.168.1.3
192.168.9.5
VM3
192.168.9.7
Neutron Abstractions
SDN Solutions
OVN
L2
Lookup
L3
Routing
Egress
Security
Groups
Simple But Extendable
• Various special services and behavior's
• VPN
• QoS (DSCP tagging)
• Dynamic Routing
• Inter clouds connectivity
• And so much more…
• External applications
• Centralized “SDN” applications
• New distributed networking services
• Networking as a Service to NFV
Classic Service Chaining
OVS OVS
Classic Service Chaining
• Chain of ports the traffic traverses
• Classifier for entry point
• Different types of chains
• Static or dynamic
• Different underlying technologies
• NSH
• MPLS
• App ports
• End points of various kinds
• VMs
• Containers
• User space applications
• Physical devices
Topology-based Service Injection
Compute Node
VM 1 VM 2
Table 0 Table 1 Table N
…
External
Application
External
Application
Table
OpenFlow / Other API
OVS
Service Injection Hooks
Logical Router
Logical Switch Logical Switch
VM 1 VM 2 VM 3
DSCP
Marking
DPI
Distributed
Load
Balancing
Topology Service Injection
• Interact with base OpenFlow pipeline
• Leverage classification metadata
• Distributed network services
• Flow based
• Compatible with SDN Applications
• Can use OpenFlow
• Expose virtual topology
• Inject services in specific hooks
• Easily extendable
• No code modifications
Service Injection Example – IPS
Compute Node
VM 1 IPS
Table 0 Service
Chains
Table N…
IPS Manager
Data Path App
IPS recognizes infected VM
Service Injection Example – IPS
Compute Node
VM 1 IPS
Table 0 Service
Chains
Table N…
IPS Manager
Data Path App
IPS app manager installs
blocking flows for VM1
traffic (Quarantine)
Use Cases
• Security Appliance
• Send specific traffic for inspection
• Traffic Mirroring
• Implement TAP on various different locations in the path
• Applicative Load Balancing
• Receive first packets of a connection and wire connection in flows
• Tenants Differentiate service between clouds
• Inter Cloud connectivity
• Border Gateway / L2GW
Server Server
Detect Elephant Flows
0 1 … 64
Flow Table
Test 1
10.0.0.3
Test 2
10.0.0.4
0 1 … 64
Flow Table
Elephant
detector
Detect elephant flow:
10.0.0.3  10.0.0.4 TCP port 1234
Write flows to
tableDSCP=64
slow path
fast path
Collect
sFlow
stats
Dragonflow Inter Cloud Connectivity (Border Gateway)
CN
CN
CN
NN
CN
CN
CN
NN
Data Center B
GW-GW
Tunnel
Data Center A
Intra-Cloud
Tunnels
Intra-Cloud
Tunnels
Connecting
Bare-Metal
Servers as
before
192.168.10.2
192.168.10.3
192.168.10.8
We are
Hiring!
jobs@toganetworks.com
www.toganetworks.com

More Related Content

What's hot

Docker 1.12 networking deep dive
Docker 1.12 networking deep diveDocker 1.12 networking deep dive
Docker 1.12 networking deep dive
Madhu Venugopal
 
Overview of Distributed Virtual Router (DVR) in Openstack/Neutron
Overview of Distributed Virtual Router (DVR) in Openstack/NeutronOverview of Distributed Virtual Router (DVR) in Openstack/Neutron
Overview of Distributed Virtual Router (DVR) in Openstack/Neutron
vivekkonnect
 
Hybrid Cloud and Hyper Cloud
Hybrid Cloud and Hyper CloudHybrid Cloud and Hyper Cloud
Hybrid Cloud and Hyper Cloud
Eshed Gal-Or
 
OpenStack HA
OpenStack HAOpenStack HA
OpenStack HA
tcp cloud
 
Troubleshooting Tracebacks
Troubleshooting TracebacksTroubleshooting Tracebacks
Troubleshooting Tracebacks
James Denton
 
Cumulus Linux 2.2 Overview
Cumulus Linux 2.2 OverviewCumulus Linux 2.2 Overview
Cumulus Linux 2.2 Overview
Cumulus Networks
 
Neutron high availability open stack architecture openstack israel event 2015
Neutron high availability  open stack architecture   openstack israel event 2015Neutron high availability  open stack architecture   openstack israel event 2015
Neutron high availability open stack architecture openstack israel event 2015
Arthur Berezin
 
OpenStack HA
OpenStack HAOpenStack HA
OpenStack HA
Kenneth Hui
 
No Surprises Geo Replication - Pulsar Virtual Summit Europe 2021
No Surprises Geo Replication - Pulsar Virtual Summit Europe 2021No Surprises Geo Replication - Pulsar Virtual Summit Europe 2021
No Surprises Geo Replication - Pulsar Virtual Summit Europe 2021
StreamNative
 
Network Automation (Bay Area Juniper Networks Meetup)
Network Automation (Bay Area Juniper Networks Meetup)Network Automation (Bay Area Juniper Networks Meetup)
Network Automation (Bay Area Juniper Networks Meetup)
Alejandro Salinas
 
Running Neutron at Scale - Gal Sagie & Eran Gampel - OpenStack Day Israel 2016
Running Neutron at Scale - Gal Sagie & Eran Gampel - OpenStack Day Israel 2016Running Neutron at Scale - Gal Sagie & Eran Gampel - OpenStack Day Israel 2016
Running Neutron at Scale - Gal Sagie & Eran Gampel - OpenStack Day Israel 2016
Cloud Native Day Tel Aviv
 
Open stack HA - Theory to Reality
Open stack HA -  Theory to RealityOpen stack HA -  Theory to Reality
Open stack HA - Theory to Reality
Sriram Subramanian
 
Advanced Data Retrieval and Analytics with Apache Spark and Openstack Swift
Advanced Data Retrieval and Analytics with Apache Spark and Openstack SwiftAdvanced Data Retrieval and Analytics with Apache Spark and Openstack Swift
Advanced Data Retrieval and Analytics with Apache Spark and Openstack Swift
Daniel Krook
 
Topologies of OpenStack
Topologies of OpenStackTopologies of OpenStack
Topologies of OpenStack
haribabu kasturi
 
Netflix at-disney-09-26-2014
Netflix at-disney-09-26-2014Netflix at-disney-09-26-2014
Netflix at-disney-09-26-2014
Monal Daxini
 
OpenStack Networks the Web-Scale Way - Scott Laffer, Cumulus Networks
OpenStack Networks the Web-Scale Way - Scott Laffer, Cumulus NetworksOpenStack Networks the Web-Scale Way - Scott Laffer, Cumulus Networks
OpenStack Networks the Web-Scale Way - Scott Laffer, Cumulus Networks
OpenStack
 
Stacking up with OpenStack: Building for High Availability
Stacking up with OpenStack: Building for High AvailabilityStacking up with OpenStack: Building for High Availability
Stacking up with OpenStack: Building for High Availability
OpenStack Foundation
 
Getting up to speed with MirrorMaker 2 | Mickael Maison, IBM and Ryanne Dolan...
Getting up to speed with MirrorMaker 2 | Mickael Maison, IBM and Ryanne Dolan...Getting up to speed with MirrorMaker 2 | Mickael Maison, IBM and Ryanne Dolan...
Getting up to speed with MirrorMaker 2 | Mickael Maison, IBM and Ryanne Dolan...
HostedbyConfluent
 
Docker network performance in the public cloud
Docker network performance in the public cloudDocker network performance in the public cloud
Docker network performance in the public cloud
Arjan Schaaf
 
How Splunk Is Using Pulsar IO
How Splunk Is Using Pulsar IOHow Splunk Is Using Pulsar IO
How Splunk Is Using Pulsar IO
StreamNative
 

What's hot (20)

Docker 1.12 networking deep dive
Docker 1.12 networking deep diveDocker 1.12 networking deep dive
Docker 1.12 networking deep dive
 
Overview of Distributed Virtual Router (DVR) in Openstack/Neutron
Overview of Distributed Virtual Router (DVR) in Openstack/NeutronOverview of Distributed Virtual Router (DVR) in Openstack/Neutron
Overview of Distributed Virtual Router (DVR) in Openstack/Neutron
 
Hybrid Cloud and Hyper Cloud
Hybrid Cloud and Hyper CloudHybrid Cloud and Hyper Cloud
Hybrid Cloud and Hyper Cloud
 
OpenStack HA
OpenStack HAOpenStack HA
OpenStack HA
 
Troubleshooting Tracebacks
Troubleshooting TracebacksTroubleshooting Tracebacks
Troubleshooting Tracebacks
 
Cumulus Linux 2.2 Overview
Cumulus Linux 2.2 OverviewCumulus Linux 2.2 Overview
Cumulus Linux 2.2 Overview
 
Neutron high availability open stack architecture openstack israel event 2015
Neutron high availability  open stack architecture   openstack israel event 2015Neutron high availability  open stack architecture   openstack israel event 2015
Neutron high availability open stack architecture openstack israel event 2015
 
OpenStack HA
OpenStack HAOpenStack HA
OpenStack HA
 
No Surprises Geo Replication - Pulsar Virtual Summit Europe 2021
No Surprises Geo Replication - Pulsar Virtual Summit Europe 2021No Surprises Geo Replication - Pulsar Virtual Summit Europe 2021
No Surprises Geo Replication - Pulsar Virtual Summit Europe 2021
 
Network Automation (Bay Area Juniper Networks Meetup)
Network Automation (Bay Area Juniper Networks Meetup)Network Automation (Bay Area Juniper Networks Meetup)
Network Automation (Bay Area Juniper Networks Meetup)
 
Running Neutron at Scale - Gal Sagie & Eran Gampel - OpenStack Day Israel 2016
Running Neutron at Scale - Gal Sagie & Eran Gampel - OpenStack Day Israel 2016Running Neutron at Scale - Gal Sagie & Eran Gampel - OpenStack Day Israel 2016
Running Neutron at Scale - Gal Sagie & Eran Gampel - OpenStack Day Israel 2016
 
Open stack HA - Theory to Reality
Open stack HA -  Theory to RealityOpen stack HA -  Theory to Reality
Open stack HA - Theory to Reality
 
Advanced Data Retrieval and Analytics with Apache Spark and Openstack Swift
Advanced Data Retrieval and Analytics with Apache Spark and Openstack SwiftAdvanced Data Retrieval and Analytics with Apache Spark and Openstack Swift
Advanced Data Retrieval and Analytics with Apache Spark and Openstack Swift
 
Topologies of OpenStack
Topologies of OpenStackTopologies of OpenStack
Topologies of OpenStack
 
Netflix at-disney-09-26-2014
Netflix at-disney-09-26-2014Netflix at-disney-09-26-2014
Netflix at-disney-09-26-2014
 
OpenStack Networks the Web-Scale Way - Scott Laffer, Cumulus Networks
OpenStack Networks the Web-Scale Way - Scott Laffer, Cumulus NetworksOpenStack Networks the Web-Scale Way - Scott Laffer, Cumulus Networks
OpenStack Networks the Web-Scale Way - Scott Laffer, Cumulus Networks
 
Stacking up with OpenStack: Building for High Availability
Stacking up with OpenStack: Building for High AvailabilityStacking up with OpenStack: Building for High Availability
Stacking up with OpenStack: Building for High Availability
 
Getting up to speed with MirrorMaker 2 | Mickael Maison, IBM and Ryanne Dolan...
Getting up to speed with MirrorMaker 2 | Mickael Maison, IBM and Ryanne Dolan...Getting up to speed with MirrorMaker 2 | Mickael Maison, IBM and Ryanne Dolan...
Getting up to speed with MirrorMaker 2 | Mickael Maison, IBM and Ryanne Dolan...
 
Docker network performance in the public cloud
Docker network performance in the public cloudDocker network performance in the public cloud
Docker network performance in the public cloud
 
How Splunk Is Using Pulsar IO
How Splunk Is Using Pulsar IOHow Splunk Is Using Pulsar IO
How Splunk Is Using Pulsar IO
 

Viewers also liked

Cloud nativeday duane (final)
Cloud nativeday duane (final)Cloud nativeday duane (final)
Cloud nativeday duane (final)
DuaneDeC7
 
20150525 open flow1.3_ryu_sdn_link aggregation 1_김지은
20150525 open flow1.3_ryu_sdn_link aggregation 1_김지은20150525 open flow1.3_ryu_sdn_link aggregation 1_김지은
20150525 open flow1.3_ryu_sdn_link aggregation 1_김지은
jieun kim
 
20150509 unix v6로 배우는 커널의 원리와 구조 4 김지은
20150509 unix v6로 배우는 커널의 원리와 구조 4 김지은20150509 unix v6로 배우는 커널의 원리와 구조 4 김지은
20150509 unix v6로 배우는 커널의 원리와 구조 4 김지은
jieun kim
 
20150502 unix v6로 배우는 커널의 원리와 구조 1 김지은
20150502 unix v6로 배우는 커널의 원리와 구조 1 김지은20150502 unix v6로 배우는 커널의 원리와 구조 1 김지은
20150502 unix v6로 배우는 커널의 원리와 구조 1 김지은
jieun kim
 
KrDAG 오픈소스를 활용하여 웹블로그 만들기_김지은_201603
KrDAG 오픈소스를 활용하여 웹블로그 만들기_김지은_201603 KrDAG 오픈소스를 활용하여 웹블로그 만들기_김지은_201603
KrDAG 오픈소스를 활용하여 웹블로그 만들기_김지은_201603
jieun kim
 
150416 OpenStack Networking with Neutron Jieun, Kim
150416 OpenStack Networking with Neutron Jieun, Kim150416 OpenStack Networking with Neutron Jieun, Kim
150416 OpenStack Networking with Neutron Jieun, Kim
jieun kim
 
Build the OpenStack Cloud with Neutron Networing, IceHouse
Build the OpenStack Cloud with Neutron Networing, IceHouseBuild the OpenStack Cloud with Neutron Networing, IceHouse
Build the OpenStack Cloud with Neutron Networing, IceHouse
jieun kim
 
집단지성프로그래밍 05. 최적화(optimization) 김지은_20150522
집단지성프로그래밍 05. 최적화(optimization) 김지은_20150522집단지성프로그래밍 05. 최적화(optimization) 김지은_20150522
집단지성프로그래밍 05. 최적화(optimization) 김지은_20150522
jieun kim
 
Kernel vm#9 powerkvm-dist-20131208
Kernel vm#9 powerkvm-dist-20131208Kernel vm#9 powerkvm-dist-20131208
Kernel vm#9 powerkvm-dist-20131208Manabu Ori
 
150625 마이크로커널 운영체제 김지은
150625 마이크로커널 운영체제 김지은150625 마이크로커널 운영체제 김지은
150625 마이크로커널 운영체제 김지은
jieun kim
 
20150509 unix v6로 배우는 커널의 원리와 구조 3 김지은
20150509 unix v6로 배우는 커널의 원리와 구조 3 김지은20150509 unix v6로 배우는 커널의 원리와 구조 3 김지은
20150509 unix v6로 배우는 커널의 원리와 구조 3 김지은
jieun kim
 
Ryu with OpenFlow 1.3, Traffic Monitor
Ryu with OpenFlow 1.3, Traffic MonitorRyu with OpenFlow 1.3, Traffic Monitor
Ryu with OpenFlow 1.3, Traffic Monitor
jieun kim
 
Pgcon2012 ori-20120224
Pgcon2012 ori-20120224Pgcon2012 ori-20120224
Pgcon2012 ori-20120224
Manabu Ori
 
resource on openstack
 resource on openstack resource on openstack
resource on openstack
jieun kim
 
150326 openstack, glance 김지은
150326 openstack, glance 김지은150326 openstack, glance 김지은
150326 openstack, glance 김지은
jieun kim
 
Ryu with OpenFlow 1.3, REST API
Ryu with OpenFlow 1.3, REST APIRyu with OpenFlow 1.3, REST API
Ryu with OpenFlow 1.3, REST API
jieun kim
 
Multi tier-app-network-topology-neutron-final
Multi tier-app-network-topology-neutron-finalMulti tier-app-network-topology-neutron-final
Multi tier-app-network-topology-neutron-final
Sadique Puthen
 
L3HA-VRRP-20141201
L3HA-VRRP-20141201L3HA-VRRP-20141201
L3HA-VRRP-20141201
Manabu Ori
 
OpenStack Neutronの機能概要 - OpenStack最新情報セミナー 2014年12月
OpenStack Neutronの機能概要 - OpenStack最新情報セミナー 2014年12月OpenStack Neutronの機能概要 - OpenStack最新情報セミナー 2014年12月
OpenStack Neutronの機能概要 - OpenStack最新情報セミナー 2014年12月
VirtualTech Japan Inc.
 
OpenStack Neutron Havana Overview - Oct 2013
OpenStack Neutron Havana Overview - Oct 2013OpenStack Neutron Havana Overview - Oct 2013
OpenStack Neutron Havana Overview - Oct 2013
Edgar Magana
 

Viewers also liked (20)

Cloud nativeday duane (final)
Cloud nativeday duane (final)Cloud nativeday duane (final)
Cloud nativeday duane (final)
 
20150525 open flow1.3_ryu_sdn_link aggregation 1_김지은
20150525 open flow1.3_ryu_sdn_link aggregation 1_김지은20150525 open flow1.3_ryu_sdn_link aggregation 1_김지은
20150525 open flow1.3_ryu_sdn_link aggregation 1_김지은
 
20150509 unix v6로 배우는 커널의 원리와 구조 4 김지은
20150509 unix v6로 배우는 커널의 원리와 구조 4 김지은20150509 unix v6로 배우는 커널의 원리와 구조 4 김지은
20150509 unix v6로 배우는 커널의 원리와 구조 4 김지은
 
20150502 unix v6로 배우는 커널의 원리와 구조 1 김지은
20150502 unix v6로 배우는 커널의 원리와 구조 1 김지은20150502 unix v6로 배우는 커널의 원리와 구조 1 김지은
20150502 unix v6로 배우는 커널의 원리와 구조 1 김지은
 
KrDAG 오픈소스를 활용하여 웹블로그 만들기_김지은_201603
KrDAG 오픈소스를 활용하여 웹블로그 만들기_김지은_201603 KrDAG 오픈소스를 활용하여 웹블로그 만들기_김지은_201603
KrDAG 오픈소스를 활용하여 웹블로그 만들기_김지은_201603
 
150416 OpenStack Networking with Neutron Jieun, Kim
150416 OpenStack Networking with Neutron Jieun, Kim150416 OpenStack Networking with Neutron Jieun, Kim
150416 OpenStack Networking with Neutron Jieun, Kim
 
Build the OpenStack Cloud with Neutron Networing, IceHouse
Build the OpenStack Cloud with Neutron Networing, IceHouseBuild the OpenStack Cloud with Neutron Networing, IceHouse
Build the OpenStack Cloud with Neutron Networing, IceHouse
 
집단지성프로그래밍 05. 최적화(optimization) 김지은_20150522
집단지성프로그래밍 05. 최적화(optimization) 김지은_20150522집단지성프로그래밍 05. 최적화(optimization) 김지은_20150522
집단지성프로그래밍 05. 최적화(optimization) 김지은_20150522
 
Kernel vm#9 powerkvm-dist-20131208
Kernel vm#9 powerkvm-dist-20131208Kernel vm#9 powerkvm-dist-20131208
Kernel vm#9 powerkvm-dist-20131208
 
150625 마이크로커널 운영체제 김지은
150625 마이크로커널 운영체제 김지은150625 마이크로커널 운영체제 김지은
150625 마이크로커널 운영체제 김지은
 
20150509 unix v6로 배우는 커널의 원리와 구조 3 김지은
20150509 unix v6로 배우는 커널의 원리와 구조 3 김지은20150509 unix v6로 배우는 커널의 원리와 구조 3 김지은
20150509 unix v6로 배우는 커널의 원리와 구조 3 김지은
 
Ryu with OpenFlow 1.3, Traffic Monitor
Ryu with OpenFlow 1.3, Traffic MonitorRyu with OpenFlow 1.3, Traffic Monitor
Ryu with OpenFlow 1.3, Traffic Monitor
 
Pgcon2012 ori-20120224
Pgcon2012 ori-20120224Pgcon2012 ori-20120224
Pgcon2012 ori-20120224
 
resource on openstack
 resource on openstack resource on openstack
resource on openstack
 
150326 openstack, glance 김지은
150326 openstack, glance 김지은150326 openstack, glance 김지은
150326 openstack, glance 김지은
 
Ryu with OpenFlow 1.3, REST API
Ryu with OpenFlow 1.3, REST APIRyu with OpenFlow 1.3, REST API
Ryu with OpenFlow 1.3, REST API
 
Multi tier-app-network-topology-neutron-final
Multi tier-app-network-topology-neutron-finalMulti tier-app-network-topology-neutron-final
Multi tier-app-network-topology-neutron-final
 
L3HA-VRRP-20141201
L3HA-VRRP-20141201L3HA-VRRP-20141201
L3HA-VRRP-20141201
 
OpenStack Neutronの機能概要 - OpenStack最新情報セミナー 2014年12月
OpenStack Neutronの機能概要 - OpenStack最新情報セミナー 2014年12月OpenStack Neutronの機能概要 - OpenStack最新情報セミナー 2014年12月
OpenStack Neutronの機能概要 - OpenStack最新情報セミナー 2014年12月
 
OpenStack Neutron Havana Overview - Oct 2013
OpenStack Neutron Havana Overview - Oct 2013OpenStack Neutron Havana Overview - Oct 2013
OpenStack Neutron Havana Overview - Oct 2013
 

Similar to Dragonflow 01 2016 TLV meetup

Scaling OpenStack Networking Beyond 4000 Nodes with Dragonflow - Eshed Gal-Or...
Scaling OpenStack Networking Beyond 4000 Nodes with Dragonflow - Eshed Gal-Or...Scaling OpenStack Networking Beyond 4000 Nodes with Dragonflow - Eshed Gal-Or...
Scaling OpenStack Networking Beyond 4000 Nodes with Dragonflow - Eshed Gal-Or...
Cloud Native Day Tel Aviv
 
Tungsten Fabric Overview
Tungsten Fabric OverviewTungsten Fabric Overview
Tungsten Fabric Overview
Michelle Holley
 
Osnug meetup-tungsten fabric - overview.pptx
Osnug meetup-tungsten fabric - overview.pptxOsnug meetup-tungsten fabric - overview.pptx
Osnug meetup-tungsten fabric - overview.pptx
M.Qasim Arham
 
FlowER Erlang Openflow Controller
FlowER Erlang Openflow ControllerFlowER Erlang Openflow Controller
FlowER Erlang Openflow Controller
Holger Winkelmann
 
Harmonia open iris_basic_v0.1
Harmonia open iris_basic_v0.1Harmonia open iris_basic_v0.1
Harmonia open iris_basic_v0.1
Yongyoon Shin
 
Simplify Networking for Containers
Simplify Networking for ContainersSimplify Networking for Containers
Simplify Networking for Containers
LinuxCon ContainerCon CloudOpen China
 
An Introduce of OPNFV (Open Platform for NFV)
An Introduce of OPNFV (Open Platform for NFV)An Introduce of OPNFV (Open Platform for NFV)
An Introduce of OPNFV (Open Platform for NFV)
Mario Cho
 
SDN/OpenFlow #lspe
SDN/OpenFlow #lspeSDN/OpenFlow #lspe
SDN/OpenFlow #lspe
Chris Westin
 
DCUS17 : Docker networking deep dive
DCUS17 : Docker networking deep diveDCUS17 : Docker networking deep dive
DCUS17 : Docker networking deep dive
Madhu Venugopal
 
LinuxCon 2015 Stateful NAT with OVS
LinuxCon 2015 Stateful NAT with OVSLinuxCon 2015 Stateful NAT with OVS
LinuxCon 2015 Stateful NAT with OVS
Thomas Graf
 
OpenStack Networking and Automation
OpenStack Networking and AutomationOpenStack Networking and Automation
OpenStack Networking and Automation
Adam Johnson
 
Nvp deep dive_session_cee-day
Nvp deep dive_session_cee-dayNvp deep dive_session_cee-day
Nvp deep dive_session_cee-day
yfauser
 
Pushing Packets - How do the ML2 Mechanism Drivers Stack Up
Pushing Packets - How do the ML2 Mechanism Drivers Stack UpPushing Packets - How do the ML2 Mechanism Drivers Stack Up
Pushing Packets - How do the ML2 Mechanism Drivers Stack Up
James Denton
 
Quantum - Virtual networks for Openstack
Quantum - Virtual networks for OpenstackQuantum - Virtual networks for Openstack
Quantum - Virtual networks for Openstack
salv_orlando
 
Development, test, and characterization of MEC platforms with Teranium and Dr...
Development, test, and characterization of MEC platforms with Teranium and Dr...Development, test, and characterization of MEC platforms with Teranium and Dr...
Development, test, and characterization of MEC platforms with Teranium and Dr...
Michelle Holley
 
OpenFlow tutorial
OpenFlow tutorialOpenFlow tutorial
OpenFlow tutorial
openflow
 
OpenStack cloud for ConoHa, Z.com and GMO AppsCloud in okinawa opendays 2015 ...
OpenStack cloud for ConoHa, Z.com and GMO AppsCloud in okinawa opendays 2015 ...OpenStack cloud for ConoHa, Z.com and GMO AppsCloud in okinawa opendays 2015 ...
OpenStack cloud for ConoHa, Z.com and GMO AppsCloud in okinawa opendays 2015 ...
Naoto Gohko
 
Discovery Day 2019 Sofia - Big data clusters
Discovery Day 2019 Sofia - Big data clustersDiscovery Day 2019 Sofia - Big data clusters
Discovery Day 2019 Sofia - Big data clusters
Ivan Donev
 
Neutron Advanced Services - Akanda - Astara 201 presentation
Neutron Advanced Services - Akanda - Astara 201 presentationNeutron Advanced Services - Akanda - Astara 201 presentation
Neutron Advanced Services - Akanda - Astara 201 presentation
Eric Lopez
 
Building a sdn solution for the deployment of web application stacks in docker
Building a sdn solution for the deployment of web application stacks in dockerBuilding a sdn solution for the deployment of web application stacks in docker
Building a sdn solution for the deployment of web application stacks in docker
Jorge Juan Mendoza
 

Similar to Dragonflow 01 2016 TLV meetup (20)

Scaling OpenStack Networking Beyond 4000 Nodes with Dragonflow - Eshed Gal-Or...
Scaling OpenStack Networking Beyond 4000 Nodes with Dragonflow - Eshed Gal-Or...Scaling OpenStack Networking Beyond 4000 Nodes with Dragonflow - Eshed Gal-Or...
Scaling OpenStack Networking Beyond 4000 Nodes with Dragonflow - Eshed Gal-Or...
 
Tungsten Fabric Overview
Tungsten Fabric OverviewTungsten Fabric Overview
Tungsten Fabric Overview
 
Osnug meetup-tungsten fabric - overview.pptx
Osnug meetup-tungsten fabric - overview.pptxOsnug meetup-tungsten fabric - overview.pptx
Osnug meetup-tungsten fabric - overview.pptx
 
FlowER Erlang Openflow Controller
FlowER Erlang Openflow ControllerFlowER Erlang Openflow Controller
FlowER Erlang Openflow Controller
 
Harmonia open iris_basic_v0.1
Harmonia open iris_basic_v0.1Harmonia open iris_basic_v0.1
Harmonia open iris_basic_v0.1
 
Simplify Networking for Containers
Simplify Networking for ContainersSimplify Networking for Containers
Simplify Networking for Containers
 
An Introduce of OPNFV (Open Platform for NFV)
An Introduce of OPNFV (Open Platform for NFV)An Introduce of OPNFV (Open Platform for NFV)
An Introduce of OPNFV (Open Platform for NFV)
 
SDN/OpenFlow #lspe
SDN/OpenFlow #lspeSDN/OpenFlow #lspe
SDN/OpenFlow #lspe
 
DCUS17 : Docker networking deep dive
DCUS17 : Docker networking deep diveDCUS17 : Docker networking deep dive
DCUS17 : Docker networking deep dive
 
LinuxCon 2015 Stateful NAT with OVS
LinuxCon 2015 Stateful NAT with OVSLinuxCon 2015 Stateful NAT with OVS
LinuxCon 2015 Stateful NAT with OVS
 
OpenStack Networking and Automation
OpenStack Networking and AutomationOpenStack Networking and Automation
OpenStack Networking and Automation
 
Nvp deep dive_session_cee-day
Nvp deep dive_session_cee-dayNvp deep dive_session_cee-day
Nvp deep dive_session_cee-day
 
Pushing Packets - How do the ML2 Mechanism Drivers Stack Up
Pushing Packets - How do the ML2 Mechanism Drivers Stack UpPushing Packets - How do the ML2 Mechanism Drivers Stack Up
Pushing Packets - How do the ML2 Mechanism Drivers Stack Up
 
Quantum - Virtual networks for Openstack
Quantum - Virtual networks for OpenstackQuantum - Virtual networks for Openstack
Quantum - Virtual networks for Openstack
 
Development, test, and characterization of MEC platforms with Teranium and Dr...
Development, test, and characterization of MEC platforms with Teranium and Dr...Development, test, and characterization of MEC platforms with Teranium and Dr...
Development, test, and characterization of MEC platforms with Teranium and Dr...
 
OpenFlow tutorial
OpenFlow tutorialOpenFlow tutorial
OpenFlow tutorial
 
OpenStack cloud for ConoHa, Z.com and GMO AppsCloud in okinawa opendays 2015 ...
OpenStack cloud for ConoHa, Z.com and GMO AppsCloud in okinawa opendays 2015 ...OpenStack cloud for ConoHa, Z.com and GMO AppsCloud in okinawa opendays 2015 ...
OpenStack cloud for ConoHa, Z.com and GMO AppsCloud in okinawa opendays 2015 ...
 
Discovery Day 2019 Sofia - Big data clusters
Discovery Day 2019 Sofia - Big data clustersDiscovery Day 2019 Sofia - Big data clusters
Discovery Day 2019 Sofia - Big data clusters
 
Neutron Advanced Services - Akanda - Astara 201 presentation
Neutron Advanced Services - Akanda - Astara 201 presentationNeutron Advanced Services - Akanda - Astara 201 presentation
Neutron Advanced Services - Akanda - Astara 201 presentation
 
Building a sdn solution for the deployment of web application stacks in docker
Building a sdn solution for the deployment of web application stacks in dockerBuilding a sdn solution for the deployment of web application stacks in docker
Building a sdn solution for the deployment of web application stacks in docker
 

Recently uploaded

Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
University of Maribor
 
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Sinan KOZAK
 
Engine Lubrication performance System.pdf
Engine Lubrication performance System.pdfEngine Lubrication performance System.pdf
Engine Lubrication performance System.pdf
mamamaam477
 
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming PipelinesHarnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
Christina Lin
 
132/33KV substation case study Presentation
132/33KV substation case study Presentation132/33KV substation case study Presentation
132/33KV substation case study Presentation
kandramariana6
 
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEMTIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
HODECEDSIET
 
Redefining brain tumor segmentation: a cutting-edge convolutional neural netw...
Redefining brain tumor segmentation: a cutting-edge convolutional neural netw...Redefining brain tumor segmentation: a cutting-edge convolutional neural netw...
Redefining brain tumor segmentation: a cutting-edge convolutional neural netw...
IJECEIAES
 
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
insn4465
 
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.pptUnit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
KrishnaveniKrishnara1
 
Heat Resistant Concrete Presentation ppt
Heat Resistant Concrete Presentation pptHeat Resistant Concrete Presentation ppt
Heat Resistant Concrete Presentation ppt
mamunhossenbd75
 
Textile Chemical Processing and Dyeing.pdf
Textile Chemical Processing and Dyeing.pdfTextile Chemical Processing and Dyeing.pdf
Textile Chemical Processing and Dyeing.pdf
NazakatAliKhoso2
 
ML Based Model for NIDS MSc Updated Presentation.v2.pptx
ML Based Model for NIDS MSc Updated Presentation.v2.pptxML Based Model for NIDS MSc Updated Presentation.v2.pptx
ML Based Model for NIDS MSc Updated Presentation.v2.pptx
JamalHussainArman
 
Generative AI leverages algorithms to create various forms of content
Generative AI leverages algorithms to create various forms of contentGenerative AI leverages algorithms to create various forms of content
Generative AI leverages algorithms to create various forms of content
Hitesh Mohapatra
 
Advanced control scheme of doubly fed induction generator for wind turbine us...
Advanced control scheme of doubly fed induction generator for wind turbine us...Advanced control scheme of doubly fed induction generator for wind turbine us...
Advanced control scheme of doubly fed induction generator for wind turbine us...
IJECEIAES
 
basic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdfbasic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdf
NidhalKahouli2
 
A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...
nooriasukmaningtyas
 
Casting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdfCasting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdf
zubairahmad848137
 
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODELDEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
gerogepatton
 
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressionsKuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
Victor Morales
 
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdfBPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
MIGUELANGEL966976
 

Recently uploaded (20)

Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
 
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
 
Engine Lubrication performance System.pdf
Engine Lubrication performance System.pdfEngine Lubrication performance System.pdf
Engine Lubrication performance System.pdf
 
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming PipelinesHarnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
 
132/33KV substation case study Presentation
132/33KV substation case study Presentation132/33KV substation case study Presentation
132/33KV substation case study Presentation
 
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEMTIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
 
Redefining brain tumor segmentation: a cutting-edge convolutional neural netw...
Redefining brain tumor segmentation: a cutting-edge convolutional neural netw...Redefining brain tumor segmentation: a cutting-edge convolutional neural netw...
Redefining brain tumor segmentation: a cutting-edge convolutional neural netw...
 
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
 
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.pptUnit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
 
Heat Resistant Concrete Presentation ppt
Heat Resistant Concrete Presentation pptHeat Resistant Concrete Presentation ppt
Heat Resistant Concrete Presentation ppt
 
Textile Chemical Processing and Dyeing.pdf
Textile Chemical Processing and Dyeing.pdfTextile Chemical Processing and Dyeing.pdf
Textile Chemical Processing and Dyeing.pdf
 
ML Based Model for NIDS MSc Updated Presentation.v2.pptx
ML Based Model for NIDS MSc Updated Presentation.v2.pptxML Based Model for NIDS MSc Updated Presentation.v2.pptx
ML Based Model for NIDS MSc Updated Presentation.v2.pptx
 
Generative AI leverages algorithms to create various forms of content
Generative AI leverages algorithms to create various forms of contentGenerative AI leverages algorithms to create various forms of content
Generative AI leverages algorithms to create various forms of content
 
Advanced control scheme of doubly fed induction generator for wind turbine us...
Advanced control scheme of doubly fed induction generator for wind turbine us...Advanced control scheme of doubly fed induction generator for wind turbine us...
Advanced control scheme of doubly fed induction generator for wind turbine us...
 
basic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdfbasic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdf
 
A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...
 
Casting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdfCasting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdf
 
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODELDEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
 
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressionsKuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
 
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdfBPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
 

Dragonflow 01 2016 TLV meetup

  • 2. What is Dragonflow? Full Implementation of OpenStack Neutron API Lightweight Distributed SDN Controller with pluggable database Project mission To Implement advanced networking services in a manner that is efficient, elegant and resource-nimble Page 2
  • 3. Dragonflow Highlights Page 3 • Integral part of OpenStack • Fully Open Source • Scale, Performance and Latency • Lightweight and Simple • Easily Extendable • Distributed SDN Control Plane • Sync Policy Level abstraction to the CN
  • 4. Dragonflow - Distributed SDN Neutron-Server Dragonflow Plugin DB OVS Dragonflow DB Driver Compute Node OVS Dragonflow DB Driver Compute Node OVS Dragonflow DB Driver Compute Node OVS Dragonflow DB Driver Compute Node DB VM VM .. VM VM .. VM VM .. VM VM ..
  • 5. Compute NodeCompute NodeCompute Node Dragonflow Network DB OVS Neutron Server OVSDB OVSDB-Server ETCD RethinkDBRAMCloud Kernel Datapath Module NIC User Space Kernel Space Dragonflow DB Drivers OVSDB ETCD RethinkDBRMC Future Dragonflow Plugin Route Core API SG vswitchd Container VM Dragonflow Controller Abstraction Layer L2 App L3 App DHCP App Fault Detection SG LBaaS …FWaaS Pluggable DB Layer NBDBDrivers SB DB Drivers smartNIC OVSDB OVSDB ETCD RMC RethinkDB OpenFlow Dragonflow – Under The Hood
  • 6. Current Release Features (Liberty) L2 core API, IPv4, IPv6 GRE/VxLAN/Geneve tunneling protocols Distributed L3 Virtual Router Hybrid proactive + reactive flow installation North-South traffic is still centralized Distributed DHCP (with just 500 lines of code!) Pluggable Distributed Database ETCD, RethinkDB, RAMCloud, OVSDB
  • 8. Network Node DHCP namespace DHCP namespace DHCP namespace DHCP namespace Neutron DHCP Implementation DHCP namespace dnsmasq DHCP Agent Neutron Server Message Queue Example • 100 Tenants • 3 vNet / tenant = 300 DHCP Servers
  • 9. 1 VM Send DHCP_DISCOVER 2 Classify Flow as DHCP, Forward to Controller 3 DHCP App sends DHCP_OFFER back to VM 4 VM Send DHCP_REQUEST 5 Classify Flow as DHCP, Forward to Controller 6 DHCP App populates DHCP_OPTIONS from DB/CFG and send DHCP_ACK Dragonflow Distributed DHCP VM DHCP SERVER 1 3 4 6 7 Compute Node Dragonflow VM OVS VM 1 2 br-int qvoXXX qvoXXX OpenFlow 1 4 2 5 7 Dragonflow Controller Abstraction Layer L2 App L3 App DHCP App SG 36 Pluggable DB Layer DB
  • 10. Dragonflow Distributed DHCP Match: Broadcast +UDP +S_Port=68 +D_Port=67 Action: Send to DHCP table Service Table DHCP Table Match: in_port => Action: Set metadata with port unique key SEND TO CONTROLLER (for every local port that its network has DHCP enabled) Default: goto “L2 Lookup Table” Compute Node VM OVS br-int qvoXXX VM qvoXXX 1 2 Dragonflow Dragonflow Local Controller Abstraction Layer L2 App L3 App DHCP App SG DB OpenFlow Ingress Port Security Ingress Classification Dispatch to Ports
  • 12. Database Framework Requirements • HA + Scalability • Different Environments have different requirements • Performance, Latency, Scalability, etc. Why Pluggable? • Long time to productize • Mature Open Source alternatives • Allow us to focus on the networking services only
  • 13. DB Driver API Implementations RAMCloud ETCD RethinkDB Zookeeper Dragonflow Pluggable Database Compute NodeCompute NodeCompute Node Dragonflow Local Controller Pluggable DB Layer Applicative DB Layer Adapter DB Driver API Expose DB Features Neutron Server Dragonflow Neutron Plugin DB Operations Database Server DB Adapter DB Adapter DB Adapter
  • 14. Distributed Database DB Data 3 DB Data 2 DB Data 1 Full Distribution Compute Node 1 Dragonflow Local Cache OVS Compute Node N Dragonflow OVS Local Cache Dragonflow DB Drivers OVSDB ETCD RethinkDBRMC DB Data 3 DB Data 2 DB Data 1 DB Data 3 DB Data 2 DB Data 1
  • 15. Distributed Database DB Data 3 DB Data 2 DB Data 1 Selective Proactive Distribution Compute Node 1 Dragonflow Local Cache OVS DB Data 1 Compute Node N Dragonflow OVS Local Cache DB Data 3 DB Data 2 Dragonflow DB Drivers OVSDB ETCD RethinkDBRMC
  • 16. Selective Proactive Distribution Compute Node 1 Dragonflow Local Cache OVS Net1 – VM1, VM2 Compute Node 2 Dragonflow OVS Local Cache Net2 – VM3, VM4 VM1 VM2 VM3 VM4 RethinkDB Net2 – VM3, VM4 Net1 – VM1, VM2
  • 18. Dragonflow Pipeline Installed in every OVS Service Traffic Classification Ingress Processing (NAT, BUM) ARP DHCP L2 Lookup L3 Lookup DVR Egress Dispatching outgoing traffic to external nodes or local ports Ingress Port Security (ARP spoofing , SG, …) Egress Port Security Egress Processing (NAT) Fully Proactive Has Reactive Flows to Controller Security Groups … Outgoing from local port Classification and tagging Dispatching Incoming traffic from external nodes to local ports
  • 20. Roadmap  Additional DBs Drivers ZooKeeper, Redis … Selective Proactive DB  Hierarchical Port Binding (SDN ToR) move to ML2  Pluggable Pub/Sub Mechanism  DB Consistency  Distributed DNAT  Security Group  Containers (Kuryr plugin and nested VM support)  Topology Service Injection / Service Chaining  Inter Cloud Connectivity (Border Gateway / L2GW)  …
  • 21. Hierarchical Port Binding (SDN ToR) move to ML2 Rack n ToR VLAN Segmentation Rack 1 ToR Rack 2 ToR Rack 3 ToR Vxlan Segmentation
  • 22. Dargonflow Hierarchical Port Binding (SDN ToR) Neutron Server REST API Neutron Core plugins ML2 Cisco(Nexus, N1Kv) OVN Morevendor plugins Type Drivers Mechanism Drivers VLAN GRE VXLAN ONOS Dragonflow TOR Neutron Service plugins DragonflowDB Rack n ToR VLAN Segmentation Vxlan Segmentation Compute Node Dragonflow VM OVS VM br-int qvoXXX qvoXXX OpenFlow Dragonflow Controller Abstraction Layer Vlan L2 App L3 App DHCP App SG Pluggable DB Layer DBDB ToRMach Driver OpenDayLight
  • 23. Pluggable Pub/Sub Mechanism Neutron-Server Dragonflow Plugin DB OVS Dragonflow DB Driver Compute Node OVS Dragonflow DB Driver Compute Node OVS Dragonflow DB Driver Compute Node OVS Dragonflow DB Driver Compute Node DB VM VM ..VM VM .. VM VM .. VM VM .. Pub/Sub if the DB internally supports Pub sub then we use it
  • 24. Pluggable Pub/Sub Mechanism Neutron-Server Dragonflow Plugin DB OVS Dragonflow DB Driver Compute Node OVS Dragonflow DB Driver Compute Node OVS Dragonflow DB Driver Compute Node OVS Dragonflow DB Driver Compute Node DB VM VM .. VM VM .. VM VM .. VM VM .. Pub/Sub Why do we need it ?  Not all DBs support pub-sub (e.g. RamCloud)  We need to be able to customize Performance, Latency, Scalability, etc.
  • 25. DB Consistency Common Problem to all SDN Solution SDN Controller North-bound Interface (REST?) South-bound Interface (Openflow) SDN Apps SDN DB Neutron DB Neutron-server ML2-Core-Plugin ML2.Drivers.Mechanism.XXX Services-Plugin Service Network Neutron API Nova API CLI / Dashboard (Horizon) / Orchestration Tool (Heat) HW Switch Nova Nova Compute VM VM Nova Compute VM VM Virtual Switch (OVS?) Virtual Switch (OVS?) Neutron Plugin Agent Neutron Plugin Agent Vendor-specific API Message Queue (AMQP) Neutron-L3-Agent Neutron-DHCP-Agent LoadBalancer Firewall VPN L3Services TopologyMgr. OverlayMgr. Security
  • 26. Dragonflow DB Consistency Neutron-Server Dragonflow Plugin DB OVS Dragonflow DB Driver Compute Node OVS Dragonflow DB Driver Compute Node DB VM VM .. VM VM .. Neutron DB  The Neutron DB is the master Database  Introduce a full-sync diff based mechanism NDB DDB  Introduce a virtual transaction mechanism NDB DDB  Key DB Requirement from multi production environments  Optimized for Read, multiple read request in very high volume from nova, Horizon …  Multi Neutron server API running on different hosts Neutron-Server Dragonflow Plugin DB Neutron-Server Dragonflow Plugin DB
  • 27. Join the project Dragonflow • Documentation https://wiki.openstack.org/wiki/Dragonflow • Bugs & blueprints https://launchpad.net/dragonflow • DF IRC channel #openstack-dragonflow Weekly on Monday at 0900 UTC in #openstack-meeting-4 (IRC)
  • 28. Break 10:30 - 11:00 - Break and Networking
  • 30. Security Groups Problems • Data plane performance • Additional Linux Bridge on the Path • Control plane performance • Rules needs to be re-compiled on port changes • Many rules due to security group capabilities • Iptable commands issued by CLI process • RPC bulks
  • 31. Security Groups in Dragonflow
  • 32. Security Groups Translations  Direction:Egress, Type:IPv4, IP Protocol:TCP, Port Range:Any, Remote IP Prefix:0.0.0.0/0  match:ct_state=+new+trk,tcp,reg6=X actions=ct(commit,zone=NXM_NX_REG6[0..15]),resubmit(,<next_table>)  Direction:Egress, Type:IPv4, IP Protocol:TCP, Port Range:Any, Remote Security Group: Y  match:ct_state=+new+trk,tcp,reg6=X,reg5=Y, actions=ct(commit,zone=NXM_NX_REG6[0..15]),resubmit(,<next_table>)
  • 33. Distributed DNAT (Floating IP) OVS VM Compute Node Public network OVS VM Compute Node Public network OVS Network Node Router Namespace
  • 35. Neutron and libnetwork A Docker Container Network Sandbox Endpoint A Docker Container Network Sandbox Endpoint A Docker Container Network Sandbox Endpoint Frontend Network Endpoint Backend Network Tenant A Net1 192.168.1.0/0 Tenant A Net2 192.168.5.0/0 VM1 192.168.1.5 VM2 192.168.1.7 192.168.5.2
  • 36. Kuryr Project Overview • Open source • Part of OpenStack Neutron’s big stadium • Under OpenStack big tent from next release!!! • Brings the Neutron networking model as a provider for the Docker CNM • Aims to support different Container Orchestration Engines • E.g. Kubernetes, Mesos, Docker Swarm • Weekly IRC meetings • Working together with OpenStack community • Neutron, Magnum, Kolla
  • 38. Dragonflow and Kuryr plans • Dragonflow to support containers networking use cases • Nested containers inside VMs support • Containers can leverage all of Dragonflow features • Distributed DHCP • Security and QoS • Containers performance and fault management • Port forwarding • Dragonflow distributed load balancer • DNS as a Service in Dragonflow • Integration with Kubernetes • Full Integration of Dragonflow and Kuryr • Containerized image of Dragonflow • VIF Binding to Dragonflow • OVS, IPVLAN
  • 39. Mixed OpenStack Environments Neutron network 1 Neutron network 2 Neutron network 3 Compute Node VM Dragonflow OVS(Controller: Dragonflow) IPVLAN / OVS VM
  • 40. Dragonflow Service Injection Dragonflow Pipeline Tenant/Admin Added Services Dynamic Routing VPN Firewall DPI QoS LB Service Discovery Inter Cloud
  • 41. Public Network 10.50.50.0/24 Router Router Tenant A Net1 192.168.1.0/0 Tenant A Net2 192.168.5.0/0 Tenant B Net1 192.168.1.0/0 Tenant B Net2 192.168.9.0/0 VM1 192.168.1.5 VM2 192.168.1.7 192.168.5.2 VM3 192.168.1.9 VM1 192.168.1.3 VM2 192.168.1.3 192.168.9.5 VM3 192.168.9.7 Neutron Abstractions
  • 43. Simple But Extendable • Various special services and behavior's • VPN • QoS (DSCP tagging) • Dynamic Routing • Inter clouds connectivity • And so much more… • External applications • Centralized “SDN” applications • New distributed networking services • Networking as a Service to NFV
  • 45. Classic Service Chaining • Chain of ports the traffic traverses • Classifier for entry point • Different types of chains • Static or dynamic • Different underlying technologies • NSH • MPLS • App ports • End points of various kinds • VMs • Containers • User space applications • Physical devices
  • 46. Topology-based Service Injection Compute Node VM 1 VM 2 Table 0 Table 1 Table N … External Application External Application Table OpenFlow / Other API OVS
  • 47. Service Injection Hooks Logical Router Logical Switch Logical Switch VM 1 VM 2 VM 3 DSCP Marking DPI Distributed Load Balancing
  • 48. Topology Service Injection • Interact with base OpenFlow pipeline • Leverage classification metadata • Distributed network services • Flow based • Compatible with SDN Applications • Can use OpenFlow • Expose virtual topology • Inject services in specific hooks • Easily extendable • No code modifications
  • 49. Service Injection Example – IPS Compute Node VM 1 IPS Table 0 Service Chains Table N… IPS Manager Data Path App IPS recognizes infected VM
  • 50. Service Injection Example – IPS Compute Node VM 1 IPS Table 0 Service Chains Table N… IPS Manager Data Path App IPS app manager installs blocking flows for VM1 traffic (Quarantine)
  • 51. Use Cases • Security Appliance • Send specific traffic for inspection • Traffic Mirroring • Implement TAP on various different locations in the path • Applicative Load Balancing • Receive first packets of a connection and wire connection in flows • Tenants Differentiate service between clouds • Inter Cloud connectivity • Border Gateway / L2GW
  • 52. Server Server Detect Elephant Flows 0 1 … 64 Flow Table Test 1 10.0.0.3 Test 2 10.0.0.4 0 1 … 64 Flow Table Elephant detector Detect elephant flow: 10.0.0.3  10.0.0.4 TCP port 1234 Write flows to tableDSCP=64 slow path fast path Collect sFlow stats
  • 53. Dragonflow Inter Cloud Connectivity (Border Gateway) CN CN CN NN CN CN CN NN Data Center B GW-GW Tunnel Data Center A Intra-Cloud Tunnels Intra-Cloud Tunnels Connecting Bare-Metal Servers as before 192.168.10.2 192.168.10.3 192.168.10.8

Editor's Notes

  1. Why is this a good thing? Common Applicative DB Adapter Layer Same layer is used by all clients Dragonflow Neutron plugin Dragonflow local controller External/Internal applications Expressed in terms of the schema model Converts model to “Key / Value” Calls the DB Driver API for DB Operations Leverage DB advance features Knows to receive and wait for DB changes According to a pre defined generic API with the driver Selective publish-subscribe
  2. Each local controller sync only relevant data according to its local ports Depends on the virtual topology Local controller gets all local ports information DB framework must support waiting for changes on specific entry column values The plugin tags the related objects with a special column value Reduce the sync load and change rate Each local controller only gets the subset of the data that is relevant for it
  3. Each local controller sync only relevant data according to its local ports Depends on the virtual topology Local controller gets all local ports information DB framework must support waiting for changes on specific entry column values The plugin tags the related objects with a special column value Reduce the sync load and change rate Each local controller only gets the subset of the data that is relevant for it
  4. Each local controller sync only relevant data according to its local ports Depends on the virtual topology Local controller gets all local ports information DB framework must support waiting for changes on specific entry column values The plugin tags the related objects with a special column value Reduce the sync load and change rate Each local controller only gets the subset of the data that is relevant for it
  5.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  6.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  7.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  8.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  9.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  10.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  11.   <voice note: Here I'd stop to thank Neutron drivers for welcoming us into the    big stadium/>   <voice note: Talk about how this may be straight away support or by the    plugins for this platforms that we can incorporate in our repository/> <voice note: Here tell the people to join us and contribute/>
  12.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  13.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  14.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  15.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  16.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  17.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  18.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  19.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  20.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  21.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>
  22.   <voice note: Here we'd explain the part about them being vendor specific    makes that each Neutron vendor would have to make its own implementation    of libnetwork or cni reinventing the wheel and without the ability to share    the common parts./>