Verified Scalability Limits

This document describes the Cisco NX-OS configuration limits for the Cisco Nexus 9000 Series switches.

Introduction

The values provided in this guide should not be interpreted as theoretical system limits for Cisco Nexus 9000 Series hardware or Cisco NX-OS software. These limits refer to values that have been validated by Cisco. They can increase over time as more testing and validation is done.

Verified Scalability Limits

The tables in this section list the verified scalability limits for Cisco NX-OS Release 7.0(3)I4(4). These limits are validated with a unidimensional configuration. The values provided in these tables focus on the scalability of one particular feature at a time.

Each number is the absolute maximum currently supported by this Cisco NX-OS release for the corresponding feature. If the hardware is capable of a higher scale, future software releases might increase this verified maximum limit. Results might differ from the values listed here when trying to achieve maximum scalability with multiple features enabled.


Note

In the following tables, the Cisco Nexus 93108TC-EX and 93180YC-EX are Leaf Spine Engine (LSE) top-of-rack (TOR) switches, and the Cisco Nexus 9508 with an N9K-X9732C-EX line card is an LSE end-of-row (EOR) switch.

Table 1. Cisco Nexus 2000 Series Fabric Extenders (FEX) Verified Scalability Limits (Unidimensional)

Feature

9500 Series Verified Limit1

9300 Series Verified Limit2

9200 Series Verified Limit

Leaf Spine Engine (LSE) TOR and EOR Verified Limit

Fabric Extenders and Fabric Extender server interfaces

32 and 1536

16 and 768

Not applicable

Not applicable

VLANs across all Fabric Extenders

2000

2000

Not applicable

Not applicable

VLANs per Fabric Extender server interface3

75

75

Not applicable

Not applicable

Port channels

426

378

Not applicable

Not applicable

Unique Fabric Extenders per Cisco Nexus 9500 Series supported line card

12

Not applicable

Not applicable

Not applicable

1 The Cisco Nexus 2200 Series and B22 Series Fabric Extenders are supported with X9464PX and X9564PX line cards on Cisco Nexus 9500 Series switches. The Cisco Nexus 2300 Series Fabric Extenders are supported with X9432PQ, X9464PX, X9464TX, X9536PQ, X9564PX, X9564TX, and X9636PQ line cards on Cisco Nexus 9500 Series switches.
2 The Cisco Nexus 2200 Series and B22 Series Fabric Extenders are supported with the Cisco Nexus 9396PX, 9372PX, and 9372PX-E chassis. The Cisco Nexus 2300 Series Fabric Extenders are supported with the Cisco Nexus 9332PQ, 9396PX, 9372PX, and 9372PX-E chassis.
3 For FEX HIF port channels, Cisco recommends that you enable STP port type edge using the spanning tree port type edge [trunk] command.
Table 2. FCoE Verified Scalability Limits (Unidimensional)

Feature

9500 Series Verified Limit

9300 Series Verified Limit

9200 Series Verified Limit

Leaf Spine Engine (LSE) TOR and EOR Verified Limit

Number of FLOGI per port

255

255

Not applicable

Not applicable

Number of FLOGI per switch

512

512

Not applicable

Not applicable

Number of port channels

8

8

Not applicable

Not applicable

Maximum number of member ports in a port channel

16

16

Not applicable

Not applicable

Number of VFCs

64

64

Not applicable

Not applicable

Number of VSANs

8

8

Not applicable

Not applicable


Note

To achieve these FCoE verified scalability numbers, you must disable FIP keep-alive messages (FKAs) on the NPIV core switch (FCF).

Table 3. Interfaces Verified Scalability Limits (Unidimensional)

Feature

9500 Series Verified Limit

9300 Series Verified Limit

9200 Series Verified Limit

Leaf Spine Engine (LSE) TOR and EOR Verified Limit

DHCP clients per switch

10 (IPv4) + 10 (IPv6)

10 (IPv4) + 10 (IPv6)

10 (IPv4) + 10 (IPv6)

10 (IPv4) + 10 (IPv6)

IP DHCP relay addresses (helper addresses) per switch

32 (IPv4) + 32 (IPv6)

32 (IPv4) + 32 (IPv6)

32 (IPv4) + 32 (IPv6)

Not applicable

Generic routing encapsulation (GRE) tunnels

8

8

8

8

Port channel links

32

32

32

32

SVIs

490 (with HSRP), 1500 (without HSRP)

450 (with HSRP)

490

LSE TOR: 450 (with HSRP)

LSE EOR: 490 (with HSRP), 1500 (without HSRP)

vPCs

300

48

48

LSE TOR: 48

LSE EOR: 300

Static network address translation (NAT)

Not applicable

1023

1023

1023

Dynamic network address translation (NAT)

Not applicable

1023

1023

1023

Static twice network address translation (NAT)

Not applicable

768

768

768

Dynamic twice network address translation (NAT)

Not applicable

1023

1023

1023

Table 4. Label Switching Verified Scalability Limits (Unidimensional)

Feature

9500 Series Verified Limit

9300 Series Verified Limit

9200 Series Verified Limit

Leaf Spine Engine (LSE) TOR and EOR Verified Limit

Forwarding Equivalence Classes (FECs)

128

128

128

Not applicable

Equal-cost multipaths (ECMPs)

32

16

32

Not applicable

FECs * ECMPs

1000

1000

1000

Not applicable

Flex counters for static MPLS in egress direction

4000

4000

4000

Not applicable

Flex counters per adjacency

2

2

2

Not applicable

Adjacencies

1024

1024

1024

Not applicable


Note

For network scalability, Cisco recommends using a hierarchical routing design with multi-hop BGP for advertising the attached prefixes from a top-of-rack (TOR) or border leaf switch.

Table 5. Layer 2 Switching Verified Scalability Limits (Unidimensional)

Feature

9500 Series Verified Limit

9300 Series Verified Limit

9200 Series Verified Limit

Leaf Spine Engine (LSE) TOR and EOR Verified Limit

MAC addresses

90,000

90,000

92,000

92,000

MST instances

64

64

64

64

MST virtual ports

85,000

48,000

48,000

LSE TOR: 48,000

LSE EOR: 85,000

RPVST virtual ports

22,000

12,000

12,000

LSE TOR: 12,000

LSE EOR: 22,000

VLANs

3967 (the remaining 127 VLANs are reserved)

3967 (the remaining 127 VLANs are reserved)

3967 (the remaining 127 VLANs are reserved)

3967 (the remaining 127 VLANs are reserved)

VLANs in RPVST mode

500

500

500

500

Total number of VLANs × ports with switchport isolated (3967 VLANs x 48 ports)

190,000

190,000

190,000

190,000

Private VLANs (PVLANs)

Primary VLANs

16

16

Not applicable

16

Secondary VLANs

20

20

Not applicable

20

Ports in Community host mode

40

40

Not applicable

40

Ports in isolated host mode

20

40

Not applicable

40

Ports in isolated trunk host mode

22

40

Not applicable

40

Ports in promiscuous mode

48

5

Not applicable

5

Ports in promiscuous trunk mode

80

5

Not applicable

5

PVLANs allowed on a PVLAN port

16

16

Not applicable

16


Note

The number of supported VLANs per vPC should be within the MST or RPVST virtual port count specified in this table, depending on the topology.


Note

The number of supported STP VLAN port instances, for Fabric Extender host interface ports, should be less than 13,000.

Table 6. Multicast Routing Verified Scalability Limits (Unidimensional)

Feature

9500 Series Verified Limit

9300 Series Verified Limit

9200 Series Verified Limit

Leaf Spine Engine (LSE) TOR and EOR Verified Limit

IPv4 multicast routes

32,000 (Layer 2 + Layer 3)

8000 (Layer 2 + Layer 3)

8000 (Layer 2 + Layer 3)

LSE TOR: 8000 (Layer 2 + Layer 3)

LSE EOR: 32,000 (Layer 2 + Layer 3)

Outgoing interfaces (OIFs)

40 (SVI + physical Layer 3) or 256 (physical Layer 3)

40 (SVI + physical Layer 3)

40 (SVI + physical Layer 3)

LSE TOR: 40 (SVI + physical Layer 3)

LSE EOR: 40 (SVI + physical Layer 3) or 256 (physical Layer 3)

IGMP snooping groups

32,000

8000

8000

LSE TOR: 8000

LSE EOR: 32,000

PIM neighbors

500

250

250

LSE TOR: 250

LSE EOR: 500


Note

The IPv4 multicast routes and the IPv4/IPv6 host routes share the same hardware table. Limits are provided for both the default line card mode and the max host line card mode.


Note
High availability (graceful restart and stateful switchover) is not supported when unicast or multicast aggressive timers are configured at any scale.
Table 7. Programmability Verified Scalability Limits (Unidimensional)

Feature

9500 Series Verified Limit

9300 Series Verified Limit

9200 Series Verified Limit

Leaf Spine Engine (LSE) TOR and EOR Verified Limit

OpenFlow

OpenFlow ports

Not applicable

96

Not applicable

Not applicable

OpenFlow Layer 2 flows

Not applicable

32,000

Not applicable

Not applicable

OpenFlow Layer 3 flows

Not applicable

3000

Not applicable

Not applicable

Table 8. Security Verified Scalability Limits (Unidimensional)

Feature

9500 Series Verified Limit

9300 Series Verified Limit

9200 Series Verified Limit

Leaf Spine Engine (LSE) TOR and EOR Verified Limit

DHCP snooping bindings

2048

2048

2048

2048

IPv4 ingress access control entries (ACEs)

3072 (per network forwarding engine)

3072 (per network forwarding engine)

3582 (per slice of the forwarding engine)

3582 (per slice of the forwarding engine)

IPv4 egress access control entries (ACEs)

768 (per network forwarding engine)

768 (per network forwarding engine)

1792 (per slice of the forwarding engine)

1792 (per slice of the forwarding engine)

IPv6 ingress access control entries (ACEs)

1536 (per network forwarding engine)

1536 (per network forwarding engine)

1792 (per slice of the forwarding engine)

1792 (per slice of the forwarding engine)

IPv6 egress access control entries (ACEs)

256 (per network forwarding engine)

256 (per network forwarding engine)

896 (per slice of the forwarding engine)

896 (per slice of the forwarding engine)


Note

The ACE scalability limits also apply to policy-based ACLs (PBACLs).

Table 9. System Management Verified Scalability Limits (Unidimensional)

Feature

9500 Series Verified Limit

9300 Series Verified Limit

9200 Series Verified Limit

Leaf Spine Engine (LSE) TOR and EOR Verified Limit

MPLS Stripping

Labels

12,000

12,000

No limit

Not applicable

Ingress interfaces

400

48

48

Not applicable

Egress interfaces

64

16

16

Not applicable

PTP

10G physical ports enabled for PTP

44

44

44

44

sFlow

sFlow ports

256

64

Not applicable

Not applicable

SPAN and ERSPAN

Configurable SPAN or ERSPAN sessions

32

32

32

32

Active SPAN or ERSPAN sessions4

4 to 32, based on the number of line cards and the session configuration

4

4

LSE TOR: 4

LSE EOR: 4 to 32, based on the number of line cards and the session configuration

Active localized SPAN or ERSPAN sessions per line card5

4

4

4

4

Source interfaces per SPAN or ERSPAN session (Rx and Tx, Rx, or Tx)

48

48

48

48

Destination interfaces per SPAN session

1 (physical/PO interface)

1 (physical/PO interface)

1 (physical/PO interface)

LSE TOR: 1 (physical/PO interface)

LSE EOR: 1 (physical interface)

Source VLANs per SPAN or ERSPAN session

32

32

32

32

TAP aggregation

Redirect interfaces in the redirect port list

12

12

12

LSE TOR: 12

LSE EOR: Not applicable

Redirect port lists (or fan outs) per system

100

100

50

LSE TOR: 50

LSE EOR: Not applicable

4 A single forwarding engine instance supports four SPAN or ERSPAN sessions. For Cisco Nexus 9300 Series switches, if the first three sessions have bidirectional sources, the fourth session has hardware resources only for Rx sources. This limitation might also apply to Cisco Nexus 9500 Series switches, depending on the SPAN or ERSPAN source's forwarding engine instance mappings.
5 The number of SPAN or ERSPAN sessions per line card reduces to two if the same interface is configured as the bidirectional source in more than one session.

Note

Beginning with Cisco NX-OS Release 7.0(3)I1(2), PTP is supported for all Cisco Nexus 9000 Series hardware except for the 100G 9408PC line card and the 100G M4PC generic expansion module (GEM).

Table 10. Unicast Routing Verified Scalability Limits (Unidimensional)

Feature

9500 Series Verified Limit

9300 Series Verified Limit

9200 Series Verified Limit

Leaf Spine Engine (LSE) TOR and EOR Verified Limit

Unicast Routing

BFD sessions (echo mode)

512

256

256

LSE TOR: 256

LSE EOR: 512

BGP neighbors

2000

512

512 (IPv4), 512 (IPv6), or 256 (IPv4 + IPv6)

512

EIGRP routes

20,000

20,000

20,000

20,000

EIGRP neighbors

512

256

256

LSE TOR: 256

LSE EOR: 512

HSRP groups

490

490

490

490

IPv4 ARP

48,000

48,000

32,000

LSE TOR: 32,000

LSE EOR: 32,000

IPv4 host routes

6

208,000 (hash table and there will be more collisions after 80%)

208,000 (hash table and there will be more collisions after 80%)

96,000 (hash table and there will be more collisions after 80%)

LSE TOR: 524,000 (default)

LSE EOR: 655,000 (default)

IPv6 host routes

7

104,000 (hash table and there will be more collisions after 80%)

104,000 (hash table and there will be more collisions after 80%)

48,000 (hash table and there will be more collisions after 80%)

LSE TOR: 24,000 (protocol learned host)

LSE EOR: 32,000 (shared between IPv6 ND and protocol learned host)

IPv6 ND

48,000

48,000

32,000

LSE TOR: 16,000

LSE EOR: 32,000

IPv4 unicast routes (LPM)*

128,000 (default system routing mode)

16,000 (max-host routing mode)

128,000 with no IPv6 routes (64-bit ALPM routing mode)

12,000 (default system routing mode)

128,000 (ALPM routing mode)

Default values: 6000 (IPv4), 1900 (IPv6), and 2000 (multicast)

With hardware profile multicast max-limit lpm-entries 0 configured: 8000 (IPv4), 1900 (IPv6), and 0 (multicast)

With hardware profile ipv6 lpm-entries maximum 0 configured: 14,000 (IPv4), 0 (IPv6), and 2000 (multicast)

With hardware profile ipv6 lpm-entries maximum 4096 and hardware profile multicast max-limit lpm-entries 0 configured: 0 (IPv4), 4096 (IPv6), and 0 (multicast)

Note

 

When you allocate the entire table for IPv4 or IPv6 LPM unicast routes, the other address family cannot be used.

LSE TOR: 524,000 (default)

LSE EOR: 655,000 (default)

IPv6 unicast routes (LPM)*

20,000 (default system routing mode)

4000 (max-host routing mode)

80,000 with no IPv4 routes (64-bit ALPM routing mode)

7000 (6000 routes < /64, 1000 routes > /64) (default system routing mode)

20,000 (ALPM routing mode)

LSE TOR: 235,000 (/64 prefix length); 1900 (non /64 prefix length)

LSE EOR: 176,000 (/64 prefix length); 3900 (non /64 prefix length)

IPv4 and IPv6 unicast routes (LPM) in 64-bit ALPM routing mode

128,000 (IPv4)

80,000 (IPv6)

Not applicable

Not applicable

Not applicable

IPv4 host routes (LPM heavy mode)

Not applicable

Not applicable

Cisco Nexus 9236C, 9272Q, and 92304QC switches: 262,000

Cisco Nexus 92160YC-X switches: 650,000

LSE TOR: 786,000

LSE EOR: 786,000

IPv6 host routes (LPM heavy mode)

Not applicable

Not applicable

16,000

LSE TOR: 24,000 (protocol learned host)

LSE EOR: 32,000 (shared between IPv6 ND and protocol learned host)

IPv4 LPM routes (LPM heavy mode)

Not applicable

Not applicable

Cisco Nexus 9236C, 9272Q, and 92304QC switches: 262,000

Cisco Nexus 92160YC-X switches: 650,000

LSE TOR: 786,000

LSE EOR: 786,000

IPv6 LPM routes (LPM heavy mode)

Not applicable

Not applicable

Cisco Nexus 9236C, 9272Q, and 92304QC switches: 131,000 (/64 prefix length); 1900 (non /64 LPM scale)

Cisco Nexus 92160YC-X switches: 294,000 (/64 prefix length); 1900 (non /64 LPM scale)

LSE TOR: 353,000 (/64 prefix length); 1900 (non /64 prefix length)

LSE EOR: 235,000 (/64 prefix length); 3900 (non /64 prefix length)

IS-ISv4 adjacencies (either L1, L2, or sum of L1 and L2 with default timers)

255

255

255

255

IS-ISv4 BFD sessions (with default timers)

255

255

Not applicable

255

IS-ISv4 routes

10,000

10,000

10,000

10,000

IS-ISv4 network type

Point to point, broadcast

Point to point, broadcast

Point to point, broadcast

Point to point, broadcast

OSPFv2 neighbors

1000

256

256

LSE TOR: 256

LSE EOR: 1000

OSPFv3 neighbors

1000

256

256

LSE TOR: 256

LSE EOR: 1000

OSPF/OSPFv3 LSA/LSDB size

100,000

100,000

100,000

100,000

OSPF/OSPFv3 areas

100

100

100

100

VRFs

1000

1000

1000

1000

VRRP groups per interface or I/O module

250

250

490

250

Policy-based routing (PBR)

Configured sequences per policy

256

256

Not applicable

Not applicable

Next-hop addresses per policy

32

32

Not applicable

Not applicable

IPv4 ACEs (unidimensional)

3072 (per network forwarding engine)

3072 (per network forwarding engine)

Not applicable

Not applicable

IPv6 ACEs (unidimensional)

1536 (per network forwarding engine)

1536 (per network forwarding engine)

Not applicable

Not applicable

IPv4 and IPv6s ACEs

2048 IPv4 + 256 IPv6

2048 IPv4 + 256 IPv6

Not applicable

Not applicable

Interfaces with PBR policy

512

512

Not applicable

Not applicable

VRRPv3

VRRPv3 groups per interface

255

255

255

255

VRRPv3 groups with default timers (1 s)

490

490

490

490

VRRPv3 groups with relaxed timers (3 s)

490

490

490

490

Pathways with one VRRPv3 group with default timer (1 s)

489

489

489

489

VRRPv3 groups and pathways combined

490

490

490

490

6 The hash table is subject to collisions. Depending on the host route pattern, collisions might occur.
7 The hash table is subject to collisions. Depending on the host route pattern, collisions might occur.

*For the Cisco Nexus 9200 Series switches, the default value for LPM unicast routes is 6000 (IPv4) or 1900 (IPv6). You can use the hardware profile multicast max-limit lpm-entries 0 command to increase the number of IPv4 LPM unicast routes to 8000. The hardware profile ipv6 lpm-entries maximum 0 command reserves the entire LPM table for IPv4. With this configuration, the IPv4 LPM scale is 14,000 (with 2000 reserved for multicast by default). This value can be increased to 16,000 with the hardware profile multicast max-limit lpm-entries 0 command. The hardware profile ipv6 lpm-entries maximum 4096 command reserves the entire LPM table for IPv6. With this configuration, the IPv6 LPM scale is 3900. When you allocate the entire table for IPv4 or IPv6 LPM unicast routes, the other address family cannot be used.


Note

The IPv4/IPv6 host routes and the IPv4 multicast routes share the same hardware table. Limits are provided for both the default line card mode and the max host line card mode.


Note

The IPv4 and IPv6 unicast routes share the same hardware table. Limits are provided for both the default line card mode and the max host line card mode.


Note
High availability (graceful restart and stateful switchover) is not supported when unicast or multicast aggressive timers are configured at any scale.

Guidelines and Limitations for OSPF Verified Scalability Limits

  • To achieve the highest scale, we recommend that you use a single OSPF instance instead of multiple instances.

  • Each OSPFv2 and OSPFv3 scale value might vary when combined with other parameters.

  • The graceful restart timeout value might need to be increased in multi-dimensional scenarios.

Table 11. VXLAN Verified Scalability Limits (Unidimensional)

Feature

9500 Series Verified Limit

9300 Series Verified Limit

9200 Series Verified Limit

Leaf Spine Engine (LSE) TOR and EOR Verified Limit

VXLAN Flood and Learn

Virtual network identifiers (VNIs) or VXLAN-mapped VLANs

1000

1000

1000

1000

Underlay multicast groups

128

128

128

128

Overlay MAC addresses

64,000

64,000

64,000

90,000

Remote VXLAN tunnel endpoints (VTEPs)

256

256

256

256

Ingress replication peers

256

256

256

256

Ingress replication Layer 2 VNIs

1000

1000

1000

1000

MAC addresses for ingress replication

64,000

64,000

64,000

90,000

Port VLAN translations under an interface

100

100

Not applicable

Not applicable

Port VLAN translations in a switch

2000

2000

Not applicable

Not applicable

Static MAC addresses pointing to a remote VTEP

1000

1000

1000

1000

VXLAN VLAN logical port VP count

7000

7000

Not applicable

Not applicable

VXLAN VLANs per FEX port (host interface)

75

75

Not applicable

Not applicable

Layer 2 routed VNIs for vPC-centralized gateway

450

450

450

450

VXLAN BGP eVPN

Layer 2 VNIs

1000

1000

1000

1000

Layer 3 VNIs / VRFs8

750

900

900

LSE TOR: 900

LSE EOR: 750

Underlay multicast groups

128

128

128

128

VTEPs

256

256

256

256

MAC addresses

64,000

64,000

64,000

90,000

IPv4 host routes

60,000

60,000

60,000

60,000

IPv6 host routes

7000

7000

7000

7000

Overlay IPv4 LPM routes

12,000

12,000

8000

6000

Overlay IPv6 LPM routes

7000

7000

2000

1900

VXLAN VLAN logical port VP count

7000

7000

Not applicable

Not applicable

VXLAN VLANs per FEX port (host interface)

75

75

Not applicable

Not applicable

VXLAN BGP eVPN Ingress Replication

Layer 2 VNIs

1000

1000

1000

1000

Layer 3 VNIs / VRFs9

750

900

900

LSE TOR: 900

LSE EOR: 750

VTEPs

128

128

128

128

MAC addresses

64,000

64,000

64,000

90,000

IPv4 host routes

32,000

32,000

32,000

32,000

IPv6 host routes

7000

7000

7000

7000

Overlay IPv4 LPM routes

12,000

12,000

8000

6000

Overlay IPv6 LPM routes

7000

7000

2000

1900

VXLAN VLAN logical port VP count

7000

7000

Not applicable

Not applicable

VXLAN VLANs per FEX port (host interface)

75

75

Not applicable

Not applicable

8 ECMP objects are not shared across multiple VRFs.
9 ECMP objects are not shared across multiple VRFs.

Deployment Case Studies

This section provides sample topologies for some common deployments. For each topology, the scalability numbers are the limits with all of the listed features enabled at the same time.


Attention

These numbers are not the maximum verified values if each feature is viewed in isolation. For these numbers, see the "Verified Scalability Limits" section.

Layer 2/Layer 3 Aggregation Topology (Max-Host Routing Mode)

This Layer 2/Layer 3 aggregation topology consists of Cisco Nexus 9508 switches as virtual port channel (vPC) aggregation pairs. These aggregation nodes are fully loaded with N9K-X9564TX, N9K-X9564PX, and N9K-X9636PQ line cards. The N9K-X9636PQ line cards are used in normal mode and breakout mode. Cisco Nexus 9396PX and 93128TX switches are used as top-of-rack units with Cisco Nexus 3000 Series switches to achieve the desired vPC scale.

The Cisco Nexus 9508 switch is also used as a core Layer 3 node that connects to a pair of vPC aggregation nodes. The focus of the topology is to test IPv4 ARP, IPv6 neighbor discovery (ND), and Layer 2 scalability and other routing, switching, and Layer 4 through Layer 7 features for management and operations. All Layer 3 interfaces are configured for dual stack, and the traffic is dual stack for all VLANs.

In the following table, the Verified Limit column lists the verified scaling capabilities with all listed features enabled at the same time. The scale numbers listed here exceed those used by most customers in their topologies. These numbers are not the maximum verified values if each feature is viewed in isolation.

Table 12. Layer 2/Layer 3 Aggregation Topology (Max-Host Routing Mode)

Feature

9508 Verified Limit (Max-Host Routing Mode)

Fully loaded chassis

1 N9K-X9636PQ, 1 N9K-X9564TX, 2 N9K-X9564PX, 1 N9K-X9432PQ, 1 N9K-X9536PQ

Physical interfaces enabled

276

Multicast S,G routes

653

Multicast *,G routes

500

IPv4 unicast routes (LPM)

5000

IPv6 unicast routes (LPM)

850

IPv4 ARP

65,000

IPv6 ND

40,000

MAC addresses

90,000

VLANs

490

vPCs*

200

OSPFv2 neighbors

20

OSPFv3 neighbors

4

BGP (IPv4) neighbors

65

BGP (IPv6) neighbors

65

SVIs

490

STP logical ports

2800 (RPVST)

HSRP VLANs (IPv4/IPv6)

490

Virtual ports

700

Port channel links

8

* The number of VLANs per vPC supported should be within the MST or RPVST virtual port count specified in this table, depending on the topology.

Layer 2/Layer 3 Aggregation Topology (Default Routing Mode)

This Layer 2/Layer 3 aggregation topology consists of Cisco Nexus 9516 switches as virtual port channel (vPC) aggregation pairs. These aggregation nodes are fully loaded with N9K-X9564TX, N9K-X9564PX, and N9K-X9536PQ line cards. The chassis is fully loaded with five line cards configured for breakout mode. The Cisco Nexus 9396PX and 93128TX switches are used as top-of-rack units with Cisco Nexus 3000 Series switches to achieve the desired vPC scale. The Cisco Nexus 9516 nodes are running in default routing mode. The Cisco Nexus 3164Q switch is also used as a core Layer 3 node that connects to a pair of vPC aggregation nodes.

The focus of the topology is to test IPv4 ARP, IPv6 neighbor discovery (ND), Layer 2 scalability, IPv4 and IPv6 LPM routing, Layer 2 and Layer 3 multicast routing for IPv4, and Layer 4 through Layer 7 features for management and operations. All Layer 3 interfaces are configured for dual stack, and the traffic is dual stack for all VLANs.

In the following table, the Verified Limit column lists the verified scaling capabilities with all listed features enabled at the same time. These numbers are not the maximum verified values if each feature is viewed in isolation.

Table 13. Layer 2/Layer 3 Aggregation Topology (Default Routing Mode)

Feature

9516 Series Verified Limit (Default Routing Mode)

9300 Series Verified Limit (Default Routing Mode)

Chassis configuration

5 N9K-X9432PQ line cards

4 N9K-X9464PX line cards

3 N9K-X9536PQ line cards

3 N9K-X9464TX line cards

1 N9K-X9564TX line card

9372

Physical ports

1335

50

vPCs

303

24

SVIs

450

450

VRFs

100

100

IPv4 ARP

40,000

40,000

IPv6 ND

10,000

10,000

STP logical ports

10,000

6000

BGP neighbors (IPv4 + IPv6)

502 + 502

502 + 502

IPv4 LPM routes

50,000

6000

IPv6 LPM routes

10,000

1000

BFD (IPv4 + IPv6)

300

102

IGP OSPFv2 neighbors

502

502

IGP OSPFv3 neighbors

502

502

HSRP (IPv4 + IPv6)

450 + 450

450 + 450

IGMP groups

2000

2000

Multicast *,G routes

2000

2000

Multicast S,G routes

8000

6000

Tracking objects

450

450

VLANs

500

500

PIM neighbors

502

502

MAC addresses

60,000

60,000

Network address translation (NAT)

Not applicable

756

sFlow

256

32

FEX System Topology

The FEX 9500 multi-dimensional scale topology consists of Cisco Nexus 9508 switches as virtual port channel (vPC) pairs. Each switch has multiple X9564PX line cards. Each switch has 32 FEX uplinks connected to them. The FEX 9300 multi-dimensional scale topology consists of two Cisco Nexus 9396PX switches used in vPC mode along with 16 FEX uplinks connected to each switch. Multiple FEXs of type Nexus 2248TP-E, 2232PP, 2248PQ, and 2348UPQ are used.

The switches are used at the Layer 2 and Layer 3 boundary and are also configured as VXLAN VTEPs. The FEX host ports are operating as Layer 2 ports. The switches are configured as gateways with the use of SVI interfaces.

In the following table, the Verified Limit column lists the verified scaling capabilities with all listed features enabled at the same time. The scale numbers listed here exceed those used by most customers in their topologies. These numbers are not the maximum verified values if each feature is viewed in isolation.

Table 14. FEX System Topology

Feature

9500 Platform Verified Limit

9300 Platform Verified Limit

Fabric Extenders

32

16

Up interfaces

1100

560

Port channels

426

256

vPC members

390

360

VLANs

744

416

PVLAN VLANs

56

56

Secondary VLANs per primary VLAN

25

25

MAC addresses

45,000

25,000

HSRP

365

365

ARP

12,000

10,000

Neighbor discovery (ND)

5000

5000

Multicast (*,G)

4000

4000

Multicast (S,G)

4000

4000

Multicast System Topology

Two Cisco Nexus 9508 switches are configured as vPC peers in one domain, and two Cisco Nexus 9372PX switches are configured as vPC peers in the other domain. The chassis are fully loaded with N9K-X9432PQ, N9K-X9464PX, N9K-X9536PQ, N9K-X9564PX, N9K-X9564TX, and N9K-X9636PQ line cards. eBGP routing is used to connect these two PIM domains. OSPF is used as IGP in one domain, and EIGRP is configured in the other domain. This setup is configured with multiple rendezvous points (RPs) to serve different multicast group ranges. BSR is used to advertise RP information in both of these PIM domains. PIM anycast is used in one domain, and MSDP anycast is used in the other domain for redundancy and load balancing. Static RP configuration is also used for a range of multicast groups.

The Cisco Nexus 9516 and Cisco Nexus 7000 Series switches are used as Layer 3 core routers in one domain. The Cisco Nexus 3164Q switches are used as Layer 3 core routers in the other domain. This topology also includes the Cisco Nexus 9396PX, Cisco Nexus 9372PX, and Cisco Nexus 3016/3064T switches in the access layer.

In addition to including Layer 2/Layer 3 IPv4 multicast routing, this topology also covers IPv4 and IPv6 host and LPM routing and Layer 2 unicast forwarding. All interfaces are configured for dual stack.

In the following table, the Verified Limit column lists the verified scaling capabilities with all listed features enabled at the same time. These numbers are not the maximum verified values if each feature is viewed in isolation.

Table 15. Multicast System Topology

Feature

9500 Series Verified Limit

9300 Series Verified Limit

Chassis configuration

N9K-X9636PQ, N9K-X9536PQ, N9K-X9564PX, N9K-X9564TX, N9K-X9432PQ, N9K-X9464PX, N9K-X9432PQ, C3164PQ

C9372PX, C9396PX, C3164PQ

Multicast S,G routes

17,500

5000

Multicast *,G routes

2500 (IGMP)

12500 (snooping)

500 (IGMP)

2500 (snooping)

Sources

2000, 200, 100, 40, 10, 3, 2, 1

2000, 200, 100, 40, 10, 3, 2, 1

Replications

40

20

ECMPs

16

8

SVIs

200

200

HSRP/VRRP

200 HSRP

100 VRRP

MAC addresses

40,000

10,000

ARP

20,000

4000

Unicast LPM IPv4 routes

20,000

4000

Unicast LPM IPv6 routes

10,000

1000

IPv4 ARP

18,000

4000

IPv6 ND

4000

2000

MSDP peers (fully mesh)

4

4

Anycast RPs (MSDP and PIM anycast) 10

2 MSDP

2 PIM anycast

IPv4 multicast routes with PIM bidirectional groups

2500

2500

10 This multicast system topology consists of two multicast PIM domains. The Multicast Source Discovery Protocol (MSDP) is used to exchange multicast source information between these two domains.

VXLAN BGP/eVPN iBGP Centric Topology Using Multicast

This VXLAN BGP/eVPN iBGP centric topology consists of Cisco Nexus 9300 and 9500 Series switches acting as VXLAN vPC tunnel endpoints (VTEPs) and VXLAN non-vPC VTEPs. VXLAN VTEPs establish iBGP sessions to a Cisco Nexus 9508 switch (route reflector) acting as a spine node. VXLAN-distributed anycast gateway SVIs are configured for dual stack, and the traffic is dual stack.

The focus of this topology is to test VXLAN overlay network scale and underlay Layer 2 switching and other routing, multicast, and Layer 4 through Layer 7 features for management and operations. Underlay PIM neighbors and IS-IS adjacency were tested with the default timer and Bidirectional Forwarding Detection (BFD) enabled on all links.

In the following table, the Verified Limit column lists the verified scaling capabilities with all listed features enabled at the same time. These numbers are not the maximum verified values if each feature is viewed in isolation.

Table 16. VXLAN BGP/eVPN iBGP Centric Topology Using Multicast

Feature

9500 Series Verified Limit

9300 Series Verified Limit

VXLAN VTEPs

128

128

VXLAN Layer 2 VNIs

1000

1000

VXLAN Layer 3 VNIs/VRFs

500

500

VXLAN multicast groups

25

25

VXLAN overlay MAC addresses

60,000

60,000

VXLAN overlay IPv4 host routes

60,000

60,000

VXLAN overlay IPv6 host routes

4000

4000

VXLAN IPv4 LPM routes

10000

10000

VXLAN IPv6 LPM routes

2000

2000

VXLAN VLAN logical port VP count

5200

5200

VLANs on VTEP node

1700 (total VLANs)

1500 (VXLAN VLANs)

200 (non-VXLAN VLANs)

1700 (total VLANs)

1500 (VXLAN VLANs)

200 (non-VXLAN VLANs)

MST instances

40

40

STP logical ports

3500

3500

vPC port channels

50

20

Underlay IS-IS neighbors

64

32

Underlay PIM neighbors

200

200

Underlay HSRP groups for regular VLANs

200

200

Underlay vPC SVIs

200

200

Underlay multicast S,G routes

4000 (ASM)

1000 (SSM)

4000 (ASM)

1000 (SSM)