Cisco Connect Dubrovnik27.-29.3.2019.

Global vision.Local knowledge.

Dejan JaksicSytstems Engineer Service Providers28-3-2019

EVPN in Service Provider network

Agenda

• Planet EVPN motivation

• EVPN Basics

• Network Fabric Architecture

• EVPN-VPNv4 interconnect

• EVPN and VPLS seamless integration

• EVPN positioning in SP network

• Conclusion

Create New Revenue Streams• Network as a service through fabric

designs• E-LAN, E-LINE, E-TREE, L3, IRB

Services

Protect Investments• Unified Networks on single overlay• Simplify protocols and operations• Industry adoption and standardization

Deploy with Ease• Seamless Brownfield Integration• Same principles and operational

experience as IP VPNs

Increase Availability• Workload Mobility• Optimal forwarding• All-Active Redundancy with Fast

Convergence

Fast, Resilient, Flexible Unified Services

EVPN: Value Proposition

EVPN

EVPN Unified Services Attributes

Per-Flow Redundancy

and load-balancing

Fast Convergence

Optimal East-West traffic

delivery

Seamless Deployment

Workload Mobility

All-Active Multi-Homing

Access Services

EVPN

EVPN external Hub: https://e-vpn.io/

1973 1982 1983 1998 2006 2007 2008 2011 2015

Ethernet Begins

Ethernet II

IEEE802.3

IEEE802.1Q

EthernetOver MPLS

IETFVPLS

IEEE802.1ah(PBB)

IETFTRILL

IETFEVPN

Evolution of Ethernet

Stolen Data Center requirements J

• EVPN with a choice of data plane encapsulation (MPLS/SR, VxLAN, SRv6) is the designed technology to address these requirements.

§ Flexible service/workload placement

§ Multi-tenancy with L2 and L3 VPN

§ Optimal Forwarding, Workload mobility

§ Fast Convergence

§ Efficient bandwidth utilization

EVPN family introduces next generation solutions for Ethernet services• BGP control-plane for Ethernet Segment

and MAC distribution learning over MPLS and VXLAN data-plane

• Same principles and operational experience as in IP VPNs

No use of Pseudowires

Multi-vendor solutions

Cisco leader in industry standardization efforts (RFCs/Drafts)

What is EVPN?

EVPN-VPWS

P2P Multipoint

EVPN

EVPN RFC 7432

RFC 8214

RFC 7432

draft-ietf-bess-evpn-inter-subnet-forwarding

EVPN-IRB

ConceptsEthernet Segment

• Represents a ‘site’ connected to one or more PEs

• Uniquely identified by a 10-byte global Ethernet Segment Identifier (ESI)

• Could be a single device or an entire networkSingle-Homed Device (SHD)Multi-Homed Device (MHD)Single-Homed Network (SHN)Multi-Homed Network (MHN)

BGP Routes

• New SAFI [70]

• Routes serve control plane purposes, including:MAC address reachabilityMAC mass withdrawalSplit-Horizon label adv.AliasingMulticast endpoint discoveryRedundancy group discoveryDesignated forwarder electionIP address reachabilityL2/L3 Integration

EVPN Instance (EVI)

• EVI identifies a VPN in the network

• Encompass one or more bridge-domains, depending on service interface typePort-basedVLAN-based (shown above)VLAN-bundling

BGP Route Attributes

• New BGP extended communities defined

• Expand information carried in BGP routes, including:MAC address movesRedundancy modeMAC / IP bindings of a GWSplit-horizon label encodingData plane Encapsulation

PE

BD

BD

EVIEVI PE1

PE2

CE1

CE2

SHD

MHD

ESI1

ESI2

Route Types

[1] Ethernet Auto-Discovery (AD) Route

[2] MAC/IP Advertisement Route

[3] Inclusive Multicast Route

[4] Ethernet Segment Route

[5] IP Prefix Advertisement Route

Extended Communities

ESI MPLS Label

ES-Import

MAC Mobility

Default Gateway

Encapsulation

LDP

RSVP

IGP

BGP-LU

LDP

Intra-Domain CP

FRR or TE

Inter-Domain CP

L2/L3VPN Services

IGP with SR

BGP

IGP with SR

BGP-LU

Unified MPLS EPN 5.0Compass

Metro Fabric

Programmability

Provisioning NETCONFYANG

NETCONFYANG

Service Provider Network - Simplification Journey

BGP LDP BGP

https://xrdocs.io/design/

Bag of existing Protocols

NETCONF/YANGSSH

Next Gen. Protocols

SRv6 SR (MPLS)PCEPISISBGP (TE, LS)IP OAMEthernet OAMEVPN

Reducing operations complexity§ Simpler automation§ Simpler to repair§ Simpler integration§ Foundation for service Orchestration

Ethernet802.1Q, 802.1adIPv4PPPoEIPv6MPLSL2TPPWE3ISISOSPFRSVP-TELACPMC-LACP

MP-BGPLDPLDP-TEIP OAMMPLS OAMEthernet OAMSTPG.8032RADIUSSNMPSyslogNetflowSSH CLI/XMLHSRP/VRRP

Key enabler for

Drastic Network Protocols Reduction

Why was EVPN needed?Solving VPLS challenges for per-flow Redundancy

• Existing VPLS solutions do not offer an All-Active per-flow redundancy – VPLS technology lacks the capability of preventing L2 loops

• Looping of Traffic Flooded from PE (BUM)

• Duplicate Frames from Floods from the Core

• MAC Flip-Flopping over Pseudowire• E.g. Port-Channel Load-Balancing does not

produce a consistent hash-value for a frame with the same source MAC (e.g. non MAC based Hash-Schemes)

CE1

Echo !

M1 M2

PE1

PE2

PE3

PE4

CE2

CE1

M1 M2

PE1

PE2

PE3

PE4

CE2

Duplicate !

CE1

M1 M2

PE1

PE2

PE3

PE4

CE2MAC

Flip-Flop

• Network Operators have emerging needs in their network:• Data center interconnect operation (DCI)• Cloud and Services virtualization (DC)• Remove protocols and Network Simplification (ICCP, HSRP/VRRP) • Integration of Layer 2 and Layer 3 VPN Services

• What about PBB-EVPN?• ASR9k only and slowing down investments

Why was EVPN needed in 2016+?

EVPN is all about BGP Control Plane and Services!!!

A1 WAN/CoreAccess

PE1

PE2 DCI2

DCI1

Spin

eSp

ine

Leaf

Leaf

Leaf

VM

VM

VM

IP, IGP, MPLS (LDP), RSVP-TE, BGP-LU

L2/L3VPN (BGP,T-LDP) - VPLS, EoMPLS

IP, MPLS, L2

VPLS, OTV

L2, STP, VLAN

Trill, Fabric-Path

Service Provider NetworkData Center Network

IP, MPLS IP,MPLS,VXLAN IP,MPLS,VXLAN

Common BGP Control PlaneL3VPN – VPNv4/6, L2/3VPN – EVPN, EVPN-VPWS

Existing Solution:

Evolution:

overlap

Overlay

Underlay

Overlay

Underlay

From MAC Bridging to MAC Routing

MPLS Transport & BGP Service

PE2

PE1

CE1 MPLS

BGP Signaling

Data Plane

Transport MPLS Label

IP Packet

Service BGP Label

BGP Signaling BGP Signaling

PE4

PE3

CE2

PE2

PE1

CE1 MPLS

BGP Signaling

Data Plane

Transport MPLS Label

L2 Frame

Service BGP Label

PE4

PE3

CE2

BGP L3VPN BGP EVPN

IP Packet IP Packet L2 Frame L2 Frame

EVPN - Ethernet VPN• Concepts are same!!! Pick your side!

SP1 SP2

C1 C2VM VM VM VM

L1 L2 L3 L4

CE1

PE2

PE1

PE4

PE3

Pick your side!

• BGP integrates services with programmable SR transport• Common across L2 / L3 services

• Services Control Plane is BGP with different AF / SAFI

• Single Service Control Plane is easy to manage and troubleshoot• HUGE investment in existing VPNv4/6• EVPN doesn’t replace L3VPN VPNv4/6 - no technical reason to do it!

EVPN vs VPNv4/6 or BGP Control Plane?

• Multi-Homed All-Active Ethernet Access• Replacement of: mLACP, STP, T-LDP, BGP-AD, etc.

• Standards-based Multi-chassis / Cluster Control Plane• Replacement of: vPC, VSS, nVCluster, etc.• Replacement of: HSRP, VRRP, etc.

• Carrier Ethernet Today• E-LINE - 80% of SP’s L2VPN portfolio (PWs)• E-LAN - Smaller # of L2 Multipoint VPN services

• There isn’t other standard technology with Ethernet All-Active Multi-homing

EVPN Flavors

• EVPN should be door-opener for IOS XR in Next Generation CO (Network Fabric)• EVPN L2/L3 multipoint brings optimal forwarding, MAC mobility, all-active

MH access• BGP Control-Plane

• Simplifies DCI/Border-Leaf configuration and service provisioning• Provides common SLA signaling

EVPN - Positioning

EVPN - Ethernet VPN• Leafs run Multi-Protocol BGP to advertise & learn MAC/IP addresses over the

Network Fabric• MAC/IP addresses are advertised to rest of Leafs

SP SP

C CVM VM VM VM

L L L L

MAC/IP advertisement & learning via BGP EVPN NLRI

Data Plane learning from the hosts

All Active multi-homingEthernet Segment

Step 1: Step 2:XR CLI:

EVPN - Ethernet-Segment for Multi-Homing

SP1 SP2

C1 C2VM VM VM VM

L1 L2 L3 L4

Unique 10-byte global identifier per Ethernet Segment

The bundle on the Leafs connecting to a node should have Identical ES identifier (ESI)

Ethernet Segment represents a node connected multiple Leafs

EVPN – Designated Forwarder (DF)

SP1 SP2

C1 C2

L1 L2 L3 L4

Duplicate

Challenge: How to prevent duplicate copies of flooded traffic from being delivered to a multi-homed Ethernet Segment (BUM traffic)?

DFNDF

EVPN – Split Horizon

SP1 SP2

C1VM VM

L1 L2

Challenge: How to prevent flooded traffic from echoing back to a multi-homed Ethernet Segment?

Echo !

BUM Label

SH Label

Transport Label

EVPN – MAC Mass-Withdraw

SP1 SP2

C1 C2VM VM VM VM

L1 L2 L3 L4

ESI1 MAC1

MAC1 can NOT be reached via ESI1

MAC1 can be reached via ESI1

MAC1 à ESI1 à Leaf1 + Leaf2

Challenge: How to inform other Leafs of a failure affecting many MAC addresses quickly while the control-plane re-converges?

EVPN – BUM Ingress Replication

SP1 SP2

C1 C2VM VM VM VM

L1 L2 L3 L4

BUM

BUM BUM

BUM

BUM = Broadcast, Unknown unicast, Multicast

R35

R34

R39

R38

R37

R36

H2

H1

LACP

LACP

R36, R37, R38, R39 - EVPN StartupR36 - Example1. RT4: DF Election & Multi-Homed Ethernet

Segment Auto-Discovery

Service Carving: 100 modulo 2 = 0R36 is DF for EVI-100

ESI: 0036.3700.0000.0000.1100

RD: 1.1.1.36:1

Ext-Com: 3637.0000.0000 (RT)

RT-4 - DF Election

R35

R34

R39

R38

R37

R36

H2

H1

LACP

LACP

R36, R37, R38, R39 - EVPN StartupR36 - Example1. RT4: DF Election & Multi-Homed Ethernet

Segment Auto-Discovery2. RT1: Per ESI Ethernet Auto-Discovery

(Split-Horizon, Mass-Withdraw)

ESI: 0036.3700.0000.0000.1100

RD: 1.1.1.36:1

Ext-Com: Flag:0x00 All-ActiveSplit-Horizon Label: 64005

RT-1 - Per ESI Ethernet AD

Ext-Com: 1:100 (RT)

R35

R34

R39

R38

R37

R36

H2

H1

LACP

LACP

R36, R37, R38, R39 - EVPN StartupR36 - Example1. RT4: DF Election & Multi-Homed Ethernet

Segment Auto-Discovery2. RT1: Per ESI Ethernet Auto-Discovery

(Split-Horizon, Mass-Withdraw)3. RT3: Inclusive Multicast

RD: 1.1.1.36:100

RT-3 - Inclusive Multicast

Ext-Com: 1:100 (RT)

Ext-Com: Type 6 Ingress-ReplicationMulticast(BUM) Label: 64120

R35

R34

R39

R38

R37

R36

H2

H1

LACP

LACP

BUM Forwarding

1. RT4: DF Election & Multi-Homed Ethernet Segment Auto-Discovery

2. RT1: Per ESI Ethernet Auto-Discovery (Split-Horizon, Mass-Withdraw)

3. RT3: Inclusive Multicast

IRBUM - Traffic

Transport Label R38-9

BUM - Traffic

BUM Label R38-9/EVI100

R35

R34

R39

R38

R37

R36

H2

H1

LACP

LACP

BUM Forwarding

1. RT4: DF Election & Multi-Homed Ethernet Segment Auto-Discovery

2. RT1: Per ESI Ethernet Auto-Discovery (Split-Horizon, Mass-Withdraw)

3. RT3: Inclusive Multicast

IRBUM - Traffic

Transport Label R37

BUM - Traffic

BUM Label R37/EVI100

SH Label R37/ESIx

X

R35

R34

R39

R38

R37

R36

H2

H1

LACP

LACP

R36, R37, R38, R39 - EVPN StartupR36 - Example1. RT4: DF Election & Multi-Homed Ethernet

Segment Auto-Discovery2. RT1: Per ESI Ethernet Auto-Discovery

(Split-Horizon, Mass-Withdraw)3. RT3: Inclusive Multicast4. RT2: MAC Advertisement

ESI: 0036.3700.0000.0000.1100

RD: 1.1.1.36:100

RT-2 - MAC Advertisement

Ext-Com: 1:100 (RT)

MAC: 0062.ec71.fbd7

Label: 64004

L2 Frame SMAC: 0062.ec71.fbd7

per EVPN Bridge Domain label

R35

R34

R39

R38

R37

R36

H2

H1

LACP

LACP

Unicast Forwarding (looks familiar?)

1. RT4: DF Election & Multi-Homed Ethernet Segment Auto-Discovery

2. RT1: Per ESI Ethernet Auto-Discovery (Split-Horizon, Mass-Withdraw)

3. RT3: Inclusive Multicast4. RT2: MAC Advertisement

L2 Frame Flow1DMAC: H1

Transport Label R36

RT-2 MAC Label/EVI

L2 Frame Flow1DMAC: H1

L2 Frame Flow1DMAC: H1

EVPN Routes – Cheat Sheet

BD1

BVI1

VRF1

EVI1

BD2

BVI2

EVI2

Vlan1BE

1.1

BE1.2

BE1 - ESI1

Vlan2

PE1

PE1 – Advertises:RT-4 Ethernet Segment Route• I have ESI1 in case when someone needs this information for Designated Forwarder(DF) Election

RT-1 Per ESI Ethernet Auto-Discovery (AD) Route• I have ESI1• ESI1 is All-Active• AC with ESI1 is connected to EVI1 and EVI2• My Split Horizon Label for ESI1 is

RT-1 Per EVI Ethernet Auto-Discovery (AD) Route(s)• EVI1 per-EVI (Aliasing) Label is • EVI2 per-EVI (Aliasing) Label is

RT-3 Inclusive Multicast Route(s)• EVI1 Label for BUM traffic is • EVI2 Label for BUM traffic is

RT-2 MAC/IP Advertisement Route(s)• MAC-A/IP-A in EVI1 and IP-A in VRF1 via label • MAC-B/IP-B in EVI2 and IP-B in VRF1 via label

RT-5 Prefix Advertisement Route(s)• IPv4/6 prefix of BVI1 in VRF1 via label VRF1-AGGL• IPv4/6 prefix of BVI2 in VRF1 via label VRF1-AGGL

BE1-SHL

BE1-SHL

EVI1-L

EVI2-L

EVI1-BUML

EVI2-BUML

PE2

PE1

CE1 MPLS

BGP Signaling

Data Plane

Transport MPLS Label

L2 Frame

Service BGP Label

PE4

PE3

CE2

L2 Frame L2 Frame

MAC-AIP-A

MAC-BIP-B

VRF1 ARPIP-A MAC-A -> BVI1IP-B MAC-B -> BVI2

BD1 MACMAC-A -> BE1.1

BD2 MACMAC-B -> BE1.2

VRF1-AGGL

EVI1-BUML

EVI2-BUML

EVI1-L

EVI2-L

BE1-SHL

EVI1-L

EVI2-L

EVPN – Distributed Anycast Gateway

SP1 SP2

C3VM

L1 L2 L3 L4

C4VM

C1 C2VM VM

BVIGW

BVIGW

BVIGW

BVIGW

Identical Anycast Gateway Virtual IP and MAC address are configured on all the Leafs

Distributed Anycast Gateway serves as the gateway for connected hosts

All the BVIs perform active forwarding in contrast to active/standby like First-hop routing protocol

Purpose: Optimal intra and inter-subnet connectivity with seamless workload mobility

EVPN – IRB in Network Fabric

40

SP SP

CVM

L L L L

CVM

C CVM VM

GWGWGWGW

Inter-subnet Forwarding L3

--L2

Intra-subnet Forwarding

Subnet 1

Subnet 2

Routing CORE

Centralized vs. Distributed Routing

Distributed Routing (IOS XR)

Fabric

Leaf

BoarderLeaf

• Optimized forwarding of east-west traffic• ARP/MAC state localized to Leafs• Helps with horizontal scaling of DC

Centralized Routing

• All east<->west routed traffic traverses to centralized gateways• Centralized gateways have full ARP/MAC state in the DC• Scale challenge• We do NOT support this design!

Fabric

Leaf

Centralized GW

Subnet 1 Subnet 2 VLAN 1 VLAN 2

L2

L3

L2

L3

Symmetric IRB Asymmetric IRB

• Flexible workload placement – any subnet anywhere

• Bridge->Route/Route->Bridge (symmetric VNI in both directions)

• ARP/MAC state localized to Leafs• Helps with horizontal scaling of DC• Cisco supports ONLY this mode

• Egress subnet must be local• Bridge->Route->Bridge (Different (Asymmetric) VNI

depending on directions)• Ingress Leaf needs ARP/MAC state for every egress

leaf• Limits scale

FabricLeaf

BoarderLeaf

FabricLeaf

BoarderLeaf

Integrated Routing and Bridging

Symmetric IRB and Asymmetric IRB are NOT interoperable!

R35

R34

R39

R38

R37

R36

H2

H1

LACP

LACP

LR36, R37, R38, R39 - EVPN StartupR36 – Example L2/L3 service1. RT4: DF Election & Multi-Homed Ethernet

Segment Auto-Discovery2. RT1: Per ESI Ethernet Auto-Discovery

(Split-Horizon, Mass-Withdraw)3. RT3: Inclusive Multicast4. RT2: MAC/IP Advertisement

ESI: 0036.3700.0000.0000.1100

RD: 1.1.1.36:100

RT-2 - MAC Advertisement

Ext-Com: 1:100 (RT) + VRF RT

MAC: 0062.ec71.fbd7

Label: 64004(BD) + 64008(VRF)

L2 Frame SMAC: 0062.ec71.fbd7

IP: 192.168.1.10

IP Header SurceIP: 192.168.1.10

IRB

IRB

IRB

Anycast IRB 192.168.1.1/24

Anycast IRB 192.168.2.1/24

EVPN - load-balancing modes

CE

PE1 PE2

All-Active(per flow)

Single LAG at the CEVLAN goes to both PETraffic hashed per flowBenefits: Bandwidth, ConvergenceCan replace: vPC, VSS, nV cluster

CE

PE1 PE2

Single-Active(per VLAN)

Multiple LAGs at the CEVLAN active on single PETraffic hashed per VLANBenefits: QoS, Billing, PolicingCan replace: HSRP/VRRP

CE

PE1 PE2

Port-Active(per port)

Multiple LAGs at the CEPort active on single PETraffic hashed per portBenefits: Protocol SimplificationCan replace: ICCP MC-LAG

V1V1 V2V1 V1, V2

EVPN - load-balancing modes

CE1

PE1 PE2

Single-Flow-Active(access L2 GW)

Single LAG at the CEVLAN goes to both PEAccess takes care of L2 loopBenefits: Legacy support for STP,

REP, G.8032

H2H1

CE3CE2

EVPN-VPWS

CE

PE

PE

PE

PE

CE

• Benefits of EVPN applied to point-to-point services• No signaling of PWs. Instead signals MP2P LSPs instead

(like L3VPN)• All-active CE multi-homing (per-flow LB) - today• Single-active CE multi-homing (per-service LB) - roadmap

• Relies on a sub-set of EVPN routes to advertise Ethernet Segment and AC reachability• PE discovery & signaling via a single protocol – BGP• Per-EVI Ethernet Auto-Discovery route (RT1 only)!!!

EVPN – Flexible Cross-Connect ServiceChallenge: How to bring multiple access services from different sources using a single EVPN E-LINE tunnel?

A1

A2

CE2

CEn

CE1

PE

CE1

CE2

CEn

EVPN Forwarder PE

MUXDEMUX

VLAN translationNormalized VLANover unique tunnel

A1

Single MPLS label used

Flexible Cross-Connect Service: Head-End

A1

A2

CE2

PE1

PE2

PWHE

PWHECEn

CE1

VLANs VRFs

ELINE(single tunnel)

Purpose: Bring multiple ELINE services into Pseudo-wire Head-end termination

EVPN - L3 Multi-Homing using EVLAG

CE1

CE2

CE3

CE4

PE1

PE2

PE3

PE4

L3VPNor

EVPN

EVLAG EVLAG

AccessAccess

EVPN

Core / Metro Fabric

No ICCP!HSRP/VRRP/MC-LAG Replacement

BVI GW

BVI GW

Multi-homing is provided by EVPN (EVLAG) Determine DF/ NDF PESynchronization (ARP, IGMP, etc.)

EVPN - Anycast-PW

A1CE Anycast-PWA3 CE

Anycast-PW

AG

AG

AG

AG

EVPNAll-Active

Anycast-SID Anycast-SID

Core / Metro Fabric AccessAccess

EVPN - access VPWS (H-EVPN)Multi-Homed EVPN-VPWS - Roadmap

A1

A2

CE

A3

A4

CE

AG

AG

AG

AG

Core / Metro Fabric AccessAccess

EVPN-VPWS EVPN EVPN-VPWS

EVPN and VPNv4/6 Interconnect

Access CoreA

A

A

ABR

P

P

CE2 DC/CODCI/BL

LEAF

LEAF

CE1

BGP - L3VPN VPNv4/6 BGP - EVPN

S

S

• DCI/BL provides EVPN to VPNv4/6 stitching• DCI/BL participates in L3 Routing not in L2 Bridging• DCI/BL is mandatory, because of summarization!!!

EVPN and VPNv4/6 Interconnect

Access CoreA

A

A

ABR

P

P

CE2 CODCI/BL

LEAF

LEAF

CE1

BGP - L3VPN VPNv4/6 BGP - EVPN

S

S

• DCI/BL provides EVPN to VPNv4/6 stitching• DCI/BL participates in L3 Routing not in L2 Bridging• DCI/BL is mandatory, because of summarization!!!

RT2 MAC/IP = CE1/32

RT5 Prefix = prefix-CE1/24 RT5 Prefix = prefix-CE1/24

X

RT5 Prefix = prefix-CE2/24 RT5 Prefix = prefix-CE2/24

Interconnect

EVPN and VPNv4/6 Interconnect

DCI/BL

BGP - L3VPN VPNv4/6 BGP - EVPN

RT2 MAC/IP = CE1/32

RT5 Prefix = prefix-CE1/24 RT5 Prefix = prefix-CE1/24

X

RT5 Prefix = prefix-CE2/24 RT5 Prefix = prefix-CE2/24

VRF ARD DCI:0RT import/export: VRF A StitchingRT import/export: VRF A

RT: VRF A Stitching

RT: VRF A Stitching

RT: VRF A Stitching

RT: VRF A

RT: VRF A

EVPN and VPNv4/6 InterconnectEVPN to VPNv4/6 Re-Advertise

DCI/BL

BGP - L3VPN VPNv4/6 BGP - EVPN

RT2 MAC/IP = CE1/32

RT5 Prefix = prefix-CE1/24 RT5 Prefix = prefix-CE1/24

X

VRF ARD DCI:0RT import/export: VRF A StitchingRT import/export: VRF A

RT: VRF A Stitching

RT: VRF A StitchingRT: VRF A

router bgp 1address-family l2vpn evpnimport stitching-rt re-originateadvertise vpnv4 unicast re-originated stitching-rt!address-family vpnv4 unicastimport re-originate stitching-rtroute-policy rt2-filter outadvertise vpnv4 unicast re-originated!

DCI/BL - BGP Configuration

3. Filter RT2 => /32 Router

1. Import: RT: VRF A Stitching

2. Advertise to vpnv4: VRF A

EVPN and VPNv4/6 InterconnectVPNv4/6 to EVPN Re-Advertise

DCI/BL

BGP - L3VPN VPNv4/6 BGP - EVPN

RT5 Prefix = prefix-CE2/24 RT5 Prefix = prefix-CE2/24

VRF ARD DCI:0RT import/export: VRF A StitchingRT import/export: VRF A

RT: VRF A StitchingRT: VRF A

router bgp 1address-family l2vpn evpnimport stitching-rt re-originateadvertise vpnv4 unicast re-originated stitching-rt!address-family vpnv4 unicastimport re-originate stitching-rtroute-policy rt2-filter outadvertise vpnv4 unicast re-originated!

DCI/BL - BGP Configuration

1. Import: VRF A

2. Advertise to EVPN: RT: VRF A Stitching

VPLS & EVPN Seamless Integration - Migration

PE2

CE1

PE4

PE3

CE4CE2

CE3

VFI1PW_PE2 UPPW_PE3 UPPW_PE4 UP

BD1

MPLSPE1

VFI1 is by default in Split Horizon Group 1• SHG1 protects loops in MPLS Core• Full Mesh of pseudowires (PW) is required for Any-to-Any forwarding

Migrate VPLS Network to EVPN Network through Seamless Integration

VPLS & EVPN Seamless Integration - Migration

PE2

CE1

PE4

PE3

CE4CE2

CE3EVI1

VFI1PW_PE2 UPPW_PE3 UPPW_PE4 UPXSHG1

BD1

MPLSPE1

VFI1 is by default in Split Horizon Group 1• SHG1 protects loops in MPLS Core• Full Mesh of pseudowires(PW) is required for Any-to-Any forwarding

EVI1 is also by default in Split Horizon Group 1• PE1 doesn’t forward data between VFI1 and EVI1

VPLS & EVPN Seamless Integration - Migration

PE2

CE1

PE4

PE3

CE4CE2

CE3EVI1

VFI1PW_PE2 UPPW_PE3 DOWNPW_PE4 UPXSHG1

BD1

BGP EVPN

MPLSPE1

VFI1 is by default in Split Horizon Group 1• SHG1 protects loops in MPLS Core• Full Mesh of pseudowires(PW) is required for Any-to-Any forwarding

EVI1 is also by default in Split Horizon Group 1• PE1 doesn’t forward data between VFI1 and EVI1

PE1&PE3 run BGP EVPN• PW_PE3 goes DOWN• Data Forwarding between PE1 and PE3 via EVI1

EVPN – MVPN in the Network Fabric (in progress)

SP SP

L L L L

State syncin EVPN

IGMP Join / Leave

Receiver

C C

L3---L2

BL BLMVPN

IRB vrf(x)

EVI-y

EVI-x

evpnmcast Receiver

SourceVRF(x)

XR 6.6.1 - IGMP L2 EVPN state sync

EVPN – Service Layering

L L L

CE CEC C

L L L

SP SP

L L L

SP SP

L L L L

L L L

SP SP

L L L

SP SP

L L L L

SP SP

L L L L

SP SP

L L L

SP SP

L L L

SP SP

L L L L

SP SP

L L L L

Core

PE

PE

P

P

BL BL

PE

PE

AG

AG

A

A

CE

AggregationAccess

MulticastFXC

EVPN-HEE-TREE

P2P L2 Bridging

L3 RoutingIRB

“Shared or single tenant”

Service Provider NetworkCore

PE/P

PE/P

P

P

PE/P

PE/P

AG

AG

A

A

CE

Metro

A

A

AG

AG

A

A

CE

A

A

Metro

BGP/EVPN

o Distributed Anycast Gateway EVPN-IRB

o All-active Multi-Homing

o EVPN symmetric IRB

o EVPN L2 for east-west traffic

o EVPN L3 for north-south traffic

ü Seamless mobility

ü Optimal forwarding (east-west & north-south)

ü All-active multi-homing load-balancing

ü Allow virtualization of appliances

ü EVPN multi-services (E-LAN, E-LINE, IRB)

ü Optimal bandwidth utilization within fabric

SP Routing Use Case Representation

Access Core

Public Cloud

Internet

Direct Content

Edge

MOBILE EDGEMobile FH

Mobile BH

5G

4G

RPD

Enterprise

Residential

Network FabricAggregation

CIN

SP PEERINGMETRO AGG

SPINE

LEAF

TOR

IP or MPLS CORE

DCI

CLOUD PEERING

Peering

CABLE AGG

B2B EDGE

CABLE EDGE

RESIDENTIAL EDGE

BUS ACCESS

CELL SITE ROUTER

PRE-AGG

EVPN Applicability

Integrated Services

EVPN Advantages:

• All-active Multi-homing & PE load-balancing (ECMP)

• Fast convergence (link, node, MAC moves)

• Control-Place (BGP) learning. PWs are no longer used.

• Optimized Broadcast, Unknown-unicast, Multicast traffic delivery

Network Efficiency

• Choice of MPLS, VxLAN or SRv6 data plane encapsulation

• Support existing and new services types (E-LAN, E-Line, E-TREE)

• Peer PE auto-discovery. Redundancy group auto-sensing

• Fully support IPv4 and IPv6 in the data plane and control plane

Service Flexibility

Investment Protection • Open-Standard and Multi-vendor support

• Integrated Layer 2 and Layer 3 VPN services

• L3VPN-like principals and operational experience for scalability and control

• EVPN is an very important complement to BGP based services

• BGP is Unified Services Control Plane across SP Network

• EVPN All-Active Multihomed Service with Distributed Anycast Gateway & Integration to L3VPN simplifies SPDC/NextGen-MEC/WAN Integration

• Cisco 5G xHaul vision: Segment Routing for underlay data plane + BGP EVPN for control plane end to end!

Conclusion

Cisco xHaul whitepaper

EVPN in Service Provider network