cisco live! :: cisco asr 9000 architecture :: brkarc-2003 | las vegas 2017
TRANSCRIPT
Cisco ASR 9000 System Architecture
Yongzhong Peng
Technical Marketing Engineer, SP Routing Infrastructure
BRKARC-2003
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Cisco Spark
Questions? Use Cisco Spark to chat with the speaker after the session
1. Find this session in the Cisco Live Mobile App
2. Click “Join the Discussion”
3. Install Spark or go directly to the space
4. Enter messages/questions in the space
How
cs.co/ciscolivebot#BRKARC-2003Cisco Spark spaces will be available until July 3, 2017.
• ASR9000 Products Introduction
• ASR9000 Distributed Control Plane
• ASR9000 Data Packet Processing
• ASR9000 QoS Architecture
• IOS-XR & IOS-XR 64 Bit
• Conclusion
Agenda
ASR 9000 Products Introduction
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Compact & Powerful
Access/AggregationFlexible Service Edge
High Density Service Edge
and Core
• Small footprint with full IOS-XR for distributed environments
• Optimized for ESE and MSE with high M-D scale for medium to large sites
• Scalable, ultra high density service routers for large, high-growth sites
One Platform, One OS, One Family
Fixed 2RU
240 Gbps
2 LC/6RU
16 Tbps
8 LC/21RU
14 Tbps
10 LC/30RU
80 Tbps
20 LC/44RU
160 Tbps
4 LC/10RU
7 Tbps
ASR 9904
ASR 9001
ASR 9006
ASR 9010
ASR 9912
ASR 9922
MSE E-MSE Peering P/PE Mobility Broadband
nV Satellites ASR 9000v, NCS5000
CE
8 LC/21RU
64 Tbps
ASR 9910
4 LC/14RU
32 Tbps
ASR 9906ASR 9901
Cisco ASR 9000 System Comprehensive Portfolio
BRKARC-2003 6
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Form Factor, Ports & BW
• Combination of 1G, 1/10G and 100G ports
• 2 RU box with 2 Tomahawk NPU(Ironman is 2 RU with 2 Typhoons)
• Depth of ~23 inches,(Ironman is 18”)
• 456G Duplex BW(Ironman is 120G Duplex)
• Fixed ports available; no MPAs
• 42 ports on the faceplate : 16X1G, 24X1/10G, 2X100G(QSFP28)
• 1G ports : LAN & MACSEC10G ports: LAN & MACSEC100G port : LAN & MACSEC
Ports / Port Density
• Redundant Power & Fan-trays
• Front to back Airflow
• NEBS, EMC Compliant
• All ports/power cabling on front plate; fan trays on backside
• Typical Power Consumed : 1200W
Mechanicals & Commons
• 64 bit XR only
• PAYG mode for 120G,240G,360G and 456G
• Full feature parity with Tomahawk feature-set
SW & Licensing
ASR9901 Overview
BRKARC-2003 7
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 8BRKARC-2003
ASR 9901 Port mapping
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Cisco ASR 9904 Overview
Feature Description
Total Capacity 16T
Capacity per Slot 4T
Slots 4 slots - 2 Line Cards and 2 RSPs
Rack size 6RU
Power 1 Power Trays, 4 Power Modules
2.1 KW DC / 3.0 KW AC supplies
Fan Side to Side Airflow, Front-to-Back Optional
1 Fan Tray, FRU
RSPs Integrated Fabric, 1+1 Redundancy
Line cards Tomahawk
Typhoon
SIP700, VSM
Front-to-back air flow with air flow baffles, 10RU
BRKARC-2003 9
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Feature Description
Capacity 32T
Capacity /Slot 4T
Slots 6 slots - 4 Line Cards and 2 RSPs
Size 14RU
Architecture Mid-plane
Power
1 Power Tray
4.4 KW DC supplies
6.0 KW AC supplies
FanFront-to-Back Airflow
2 Fan Trays, FRU
RSPs Integrated Fabric, 1+1 Redundancy
Fabric Cards
5 Fabric cards in the back for additional capacity
230G per FC at FCS
6+1 Redundancy – 2 RSP’s integrated fabric + 5 FCs
Line cardsTomahawk
Not supported LCs – SIP700, VSM and Typhoon
IOS-XR 64 Bit
Front View Rear View
Introducing ASR9906
BRKARC-2003 10
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Cisco ASR 9910 Overview
Feature Description
Capacity 64 Tbps; mid-plane architecture
Capacity /Slot 4 Tbps
Slots 10 slots - 8 Line Cards and 2 RSPs
Rack size 21RU
Power 2 Power Trays (4.4 KW DC supplies; 6.0 KW AC supplies)
Fan Front-to-Back Airflow. 2 Fan Trays, FRU
RSPs Integrated Fabric, 1+1 Redundancy
Fabric Cards
5 Fabric cards in the rear for additional capacity
230G per FC
Up to 6+1 Redundancy using RSP’s integrated fabric
Line cards
Fixed high density 10/100GE Typhoon and Tomahawk LCs
Modular
MOD80, MOD160, MOD200, MOD400
MPAs: 20xGE, 4x10G, 8x10G, 2x40G, 1x100G, 20x10G, 2x100G
Low Density Fixed
40x1GE, 4x10G+16x1GE
Service Card: Virtualized Service Module (VSM)
5 Fabric Cards
2 Fan Trays
2 RSPs
8 LC Slots
2 Power Trays
BRKARC-2003 11
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
ASR 9000 Linecards Evolution
3rd Gen
Tomahawk
Class
800G
Tomahawk
28nm
240 Gbps
Tigershark
28nm
200 Gbps
SM15
28nm
1.20 Tbps
X86
6 Core
2 Ghz
2nd Gen
Typhoon
Class
360G
Typhoon
55nm
60 Gbps
Skytrain
65nm
60 Gbps
Sacramento
65nm
220 Gbps
PowerPC
Quad Core
1.5 Ghz
Trident
90nm
15 Gbps
Octopus
130nm
60 Gbps
Santa Cruz
130nm
90 Gbps
PowerPC
Dual Core
1.2 Ghz
1st Gen
Trident
Class
120G
BRKARC-2003 12
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
ASR 9K Ethernet Line Card Overview
A9K-MOD80 A9K-MOD160
MPAs20x1GE2x10GE4x10GE1x40GE2x40GE
-TR, -SE
MPAs1x100GE2x100GE20x10GE
+ Typhoon MPAs
MOD400/MOD200A9K-4x100GE A9K-12x100GE
A9K-36x10GE A9K-24x10GE
A9K-40GE
A9K-4T16GE
2nd LC Typhoon
NPU: 60Gbps,
~45Mpps
3rd LC Tomahawk
NPU: 240Gbps,
~150Mpps
A9K-8x100GE A9K-400G-DWDM24/48x10/1G LC
BRKARC-2003 13
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Tomahawk Forwarding Engine and LC
• High Density/Scale
• 240 Gbps throughput/ 150 MPPS+ for “line
rate” 100GE ports
• Increased scale for multi-dimensional
profiles
• 2M MAC addresses(6M Capacity)
• 10M v4 or 5M v6 routes
• 1M queues/NP
• Flexible Interface Support
• Supports dense 100GE/40GE/10GE
• “Mix and match” interface types on the same
LC
• CPAK Advantage
• Pluggable CPAK based on customer
demand (1x100GE, 2x40GE, 10x10GE)
• In-line Security
• MACSec support in-line on LC at line rate
100G (800G/LC)
• “Green” Design
• Ability to power down components for power
savings
• High Availability Customized Silicon – Hitless
FPD Upgrades
• Hardware Offload of CPU Intensive Protocols
• Elimination of loopback for most feature paths
BRKARC-2003 14
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Tomahawk LC: 8x100GE Architecture
XB
AR
Se
pa
rate
d S
witc
h F
ab
ric
LC CPU
TomahawkNP
FIAPHY
TomahawkNP
FIAPHY
TomahawkNP
FIAPHY
TomahawkNP
TigersharkFIA
PHY
CPAK:
100G, 40G, 10G
L2/L3/L4 lookups, all VPN types, all
feature processing, mcast
replication, QoS/Queuing, ACL, etc …
Slice Based Architecture
Macsec Suite B+, G.709, OTN,
Clocking
VoQ buffering, Fabric credits,
mcast hashing, scheduler for fabric
and egress port
240G
240G
240G
240G
240G
240G
240G
240G
FPOE, Auto-Spread,
DWRR, RBH, replication
Per Slice Power Management: (100-200W Power Savings)
PE1(admin-config)# hw-module power saving location ? slice [0-3]
Up to
14x115G
BRKARC-2003 15
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Line Card Components
NPUFIA Switch
Fabric ASIC
TM
PHY
CPU
P1
P2
P3
BEP1
P2
P3
BE
Runs distributed control plane protocols for increased scale
BFD, CFM, ARP
Receive FIB table from RP and program hardware forwarding table
Provides data connection to switch fabric
Manage VoQ, Superframe and loadbalancingdata traffic across switch fabric
Mcast replication table for replication toward NPs
Main forwarding engine L2 and L3 lookups
Multicast replication toward Optics
User level QoS and Security features
Dedicated queue ASIC – TM (traffic
manager) per NPU for QoS functions
User Configurable Queue on TM.
Default Port Queue Always Created.
BRKARC-2003 16
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Network Processor Architecture Details
• TCAM: VLAN tag, QoS and ACL classification
• Stats memory: interface statistics, forwarding statistics etc
• Frame memory: buffer, Queues
• Lookup Memory: forwarding tables, FIB, MAC, ADJ
• TR/SE
• Different TCAM/frame/stats memory size for different per-LC QoS, ACL, logical interface scale
• Same lookup memory for same system wide scale mixing different variation of LCs doesn’t impact system wide scale
-
STATS MEMORY
FRAME MEMORYLOOKUP
MEMORY TCAM
FIB MAC
NPU Complex
Forwarding chip (multi core)
TR and SE has different
memory sizeTR and SE has same
memory size
-TR: transport optimized, -SE: Service edge optimized
BRKARC-2003 17
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Tomahawk Line Card Architectures
4x100G/40x10G Port LC
MOD400 LC
Switch
Fabric
CPU
Up to
14x115GSwitch
Fabric
FIA
FIA
FIA
FIANP
NP
NP
NP
PHY
PHY
PHY
PHY
CPAK
CPAK
CPAK
CPAK
CPU
Switch
Fabric
FIA
FIANP
NP
SerDes
XBAR
Bay
Bay
CPU
P
H
Y
P
H
Y
MOD200 LC
Up to
14x115G
Switch
FabricFIANPSerDes
XBAR
Bay
Bay
CPU
P
H
Y
P
H
Y
Up to
14x115G
Up to
14x115G
FIAPHY NP
FIAPHY NP
FIAPHY NP
FIAPHY NP
FIAPHY NP
FIAPHY NP
12x100G Port LC
BRKARC-2003 18
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
• Supports 1G, 10G HW configurations available on Typhoon
• Feature and scale parity on PAR with Typhoon LC with minimal exceptions
• No OTN/MACSec• Support for all SFP+/SFP currently
available on Typhoon• 64bit XR capable
Ultra-Dense dual rate 1G/10G fixed line card.
Available in 24-port and 48-port versions
(Oversubscribed)
Powered by the latest generation Tomahawk NPU
& ASICS
Price and power optimized LC
• Ideal low cost replacement for both fixed and modular Typhoon LC
• PAYG consumption model available
Dual-rate dense 10G/1G Tomahawk LCA9K-48x10G-1G – Enables easy migration from Typhoon
BRKARC-2003 19
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
• Supports all 1G and 10G. 1G and 10G are configurable follow certain rule
• 48-port has a two NPU slices with shared 80Mb TCAM
• PHY does not support OTN Framing or MACSec
Ports0-7 PHY0
FIA SM15Ports8-15 PHY1
Ports16-23 PHY2
NPU 0
SFPs0-7
SFPs8-15
SFPs16-23
IFE
0IF
E1
Ports24-31 PHY3
FIA SM15Ports32-39 PHY4
Ports40-47 PHY5
NPU 1
SFPs24-31
SFPs32-39
SFPs40-47
IFE
0IF
E1
• 240G Raw Bandwidth/ 200G
Data Traffic Bandwidth
• Potentially Oversubscribed
8 Ports
8 Ports
4 Ports
4 Ports
A9K-48x10G-1G Architecture
BRKARC-2003 20
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
• Supports all 1G and 10G. 1G and 10G are configurable follow certain rule
• 24-port has a single NPU slice with 40Mb TCAM
• Phy does not support OTN Framing or MACSec
Ports0-7 PHY0
FIA SM15Ports8-15 PHY1
Ports16-23 PHY2
NPU 0
SFPs0-7
SFPs8-15
SFPs16-23
IFE
0IF
E1
• 240G Raw Bandwidth/ 200G
Data Traffic Bandwidth
• Potentially Oversubscribed
8 Ports
8 Ports
4 Ports
4 Ports
A9K-24x10G-1G Architecture
BRKARC-2003 21
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
A9K-48x10G-1G Port Organization• A9K-48X10GE-1G-SE/TR has two NPU slices, A9K-24X10GE-1G-SE/TR has one
• Each NPU slice consists of two Interface Engines (IFE)
• Interface Ports are organized in port-groups
• Each port-group consists of 4 interface ports
• Each IFE consists of three port-groups
A9K-48X10GE-1G-SE/TR
00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
TOMAHAWK NPU
IFE
TOMAHAWK NPU
IFE IFE IFE
12 ports per IFE 12 ports per IFE 12 ports per IFE 12 ports per IFE
port-group port-group port-group port-group port-group port-group port-group port-group port-group port-group port-group port-group
BRKARC-2003 22
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
A9K-48x10-1G & A9K-48x10-1G Port Speed Rules
• Interface Port Speed is configurable: 1G / 10G
• Per default, all ports are 10G
• The 4 Rules:
1. All ports in a port-group (group of four ports) have to have the same speed.
2. If the first port-group on an IFE (group of three port-groups) is configured as 1G, then all ports in that same IFE must be 1G.
3. If the first port-group of an IFE is 10G, then the other two port-groups in that same IFE can be any combination of 1G or 10G.
4. When configuring port speeds, all ports of a given linecard need to be configured together in one single CLI command.
• Deviations from these rules will results in a CLI rejection
• Changing port speed on an interface does not impact traffic on other interfaces
BRKARC-2003 23
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Possible Port Speed Combination per IFE
P0 P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11
Configuration 1 1G 1G 1G 1G 1G 1G 1G 1G 1G 1G 1G 1G
Configuration 2 10G 10G 10G 10G 1G 1G 1G 1G 1G 1G 1G 1G
Configuration 3 10G 10G 10G 10G 10G 10G 10G 10G 1G 1G 1G 1G
Configuration 4 10G 10G 10G 10G 1G 1G 1G 1G 10G 10G 10G 10G
Configuration 5 10G 10G 10G 10G 10G 10G 10G 10G 10G 10G 10G 10G
Interface Engine (IFE)
• 5 possible port speed combinations for each IFE
• IFE are independent of each other for port speed configuration
BRKARC-2003 24
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Port Speed Configuration Example
hw-module location <rack>/<slot>/CPU0 port-mode RxS,RxS[,RxS…]
R = Run-length (number of consecutive ports with same speed)
S = Speed (1 or 10, in Gbps)
EXAMPLE:
hw-module location 0/4/CPU0 port-mode 8x10,4x1,4x10,8x1,20x10,4x1
00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
BRKARC-2003 25
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
• MPA’s Supported:2x100GE CPAK. 1x100GE CPAK. 20x10GE SFP+All Typhoon MPA
• Flexibility to use CPAK 10G/40G/100G optics on 100G MPAs
• 2 Flavors for FlexibilityMOD400 has 2 Tomahawk ASICSMOD200 has 1 Tomahawk ASIC
MPA #1(2x100G)
MPA #2(20x10G)
Mo
d4
00
LC
2 x 100GCPAKs
20 x 10G SFP+
MOD200Support Matrix
Comb 1 Comb 2 Comb 3
EP0 2x100G-MPA 20x10G-MPA
1x100G or anyTyphoon MPAs
EP1 None None 1x100G or any Typhoon MPAs
MOD400Support Matrix
Comb 1 Comb 2 Combo 3 Comb 4
EP0 2x100G-MPA 20x10G-MPA 2x100G-MPA 1x100G or anyTyphoon MPAs
EP1 2x100G-MPA 20x10G-MPA 20x10G-MPA 2x100G-MPA OR20x10G-MPA
26BRKARC-2003
Tomahawk Modular LCs – Mod400 & Mod200
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
CPAK Interface Flexibility
Configurable 100GE Interconnect
Options for 10GE interfaces:
10GE Interconnect Options
40GE Interconnect Options
hw-module 0/x/cpu0 port z breakout 10xTenGigE
hw-module 0/x/cpu0 port z breakout 2xFortyGigE
DUPLEX SC TO
LC /SC/ST SM
CPAK-100G-LR4
LGX
Panel
MPO24 TO 10X
DUPLEX
LC/SC/ST MM
CPAK-100G-SR10
CPAK-10X10G-LRMPO24 TO 10X
DUPLEX LC
/SC/ST SM
LGX
Panel
LGX
Panel
CPAK-2X40G-LR4
LC TO DUPLEX
LC/SC/ST SM
LGX
Panel
Interface HunGigE 0/x/y/z
Breakout Interface Convention:
Rack/Slot/Bay/Port (phy)/Breakout#
Interface TenGigE 0/x/y/z/0
Interface TenGigE 0/x/y/z/1
…
Interface TenGigE 0/x/y/z/9
Interface FortyGigE 0/x/y/z/0
Interface FortyGigE 0/x/y/z/1
BRKARC-2003 27
ASR 9000 Distributed Control Plane
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
ASR 9000 System Architecture “At-a-Glance”
Fully Distributed Architecture for
High Performance and High Multi-
dimensional Control Plane Scale
Active-Active Switch Fabric
RSP
RSP/RP
FIC
CPU
BITS/DTI
FIA
CPU
Switch
Fabric
FIA
FIA
FIA
FIANP
NP
NP
NP
PHY
PHY
PHY
PHY
CPAK
CPAK
CPAK
CPAK
CPAK
CPAK
CPAK
CPAK
CPU
CPU
Switch
Fabric
FIA
FIA NP
NP
SerDes
XBAR
Bay
Bay
CPU
Switch Fabric
Line Card
Data forwarding is fully
distributed across NPs
Control plane split
among RSP/RP
and LC CPU
Network Processor
Network Processor
BRKARC-2003 29
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Switch
Fabric
FIA
FIA
FIA
FIANP
NP
NP
NP
PHY
PHY
PHY
PHY
CPAK
CPAK
CPAK
CPAK
CPAK
CPAK
CPAK
CPAK
CPU
ASR9000 Fully Distributed Control Plane
Switch Fabric
PuntFPGA FIA
Switch Fabric
RP
LPTS
LC CPU: ARP, ICMP, BFD, NetFlow,
OAM/CFM, L2 Protocols etc
RP CPU: Routing, MPLS, IGMP, PIM,
HSRP/VRRP, etc
LPTS (local packet transport service):
control plane policing
Control packet
Punt Switch
CPU
CPU
NP Offloading: BFD
LC
BRKARC-2003 30
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
LC NP
31BRKARC-2003
L3 Control Plane Architecture
LC CPU
RSP/RP CPU
LDP RSVP-TE BGP
ISIS
OSPF
EIGRP
Static
HW FIB Adjacency
ARP/NDP
LSD RIB
AIB
SW FIB
AIB: Adjacency Information Base
RIB: Routing Information Base
FIB: Forwarding Information Base
LSD: Label Switch Database
RSP/RP
LC
ISIS/OSPF BGP
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Distributed ARP Processing
LC
CPU
PHY NP FIALPTS
Punt Switch
ARP
spio netio
Tsec Driver
SPP
Line card
FIA
Sw
itch
Fab
ric
RP
Sw
itch
Fab
ric
ARP
RP CPUIncomplete ADJ Packets
Incoming ARP Packets
Outgoing ARP Packets
BRKARC-2003 32
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
NP
Switch Fabric
3x10GE
SFP +
3x10GE
SFP +
3x10GE
SFP +
3x10GE
SFP +
3x10GE
SFP +
3x10GE
SFP +
3x10GE
SFP +
3x10GE
SFP +
FIA
FIA
FIA
FIA
Sw
itch
Fab
ric A
SIC
CPU
PuntFPGA FIA
CPU
Switch Fabric
RP
LC1
3x10GE
SFP +
3x10GE
SFP +
NP
NP
3x10GE
SFP +
3x10GE
SFP +
NP
NP
3x10GE
SFP +
3x10GE
SFP +
NP
NP
3x10GE
SFP +NP
NPFIA
FIA
FIA
FIA
Sw
itch
Fab
ric A
SIC
CPULC2
1 2
2
NP learn MAC address in hardware (around
4M pps)
NP flood MAC notification (data plane)
message to all other NPs in the system to sync
up the MAC address system-wide. MAC
notification and MAC sync are all done in
hardware
1
2
NP
NP
NP
NP
NP
NP
NP
Hardware based MAC learning: ~4Mpps/NP
33BRKARC-2003
MAC Learning and Sync
Data packet
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Distributed BFD Architecture
OSPF ISIS BGP
BFD Session Tables on RP
BFD Events
BFD session info
RP
BFD Session Tables on LC
BFD EventsBFD Session info
LC1-CPU
BFD Hellos
BFD Session Tables on LC
BFD Hellos
BFD Session info
……
LCn-CPU
BRKARC-2003 34
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
HW-offloaded BFD
OSPF ISIS BGP
BFD Session Tables on RP
BFD Events
BFD session info
RP
BFD EventsBFD Session info BFD Session info
……
BFDOnNP
LCn-CPU
BFD Hellos
BFDOnNP
BFDOnNP
BFDOnNP
BFDOnNP
LC1-CPU
BFD Hellos
BFDOnNP
BFDOnNP
BFDOnNP
hw-module bfd-hw-offload enable location 0/0/CPU0
BRKARC-2003 35
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Distributed Netflow Architecture
NPU
LC-CPU1M Record Cache
NPU NPU NPU
RPNetflow show/clear
Command
Netflow
Configure
NPU: • Traffic Filtering
• Traffic Sampling
• Extract flow
header information
• LPTS Policer
• Aggregate 200Kpps netflow
policer rate
• Evenly Divided among
netflow-enabled NPs
netIO
EXPORT
• Manage flow table
• Export flow
information
LC
BRKARC-2003 36
ASR 9000 Data Packet Processing
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
CPAK 0
NP FIA
FIA
FIA
FIA
CPU
CPAK 1
PHY
CPAK 2
NP CPAK 3
PHY
CPAK 4
NP CPAK 5
PHY
CPAK 6
NP CPAK 7
PHY
…
Up to
14x120
G
Switch Fabric (SM15)
CPAK 0
NP FIA
FIA
FIA
FIA
CPU
CPAK 1
PHY
CPAK 2
NP CPAK 3
PHY
CPAK 4
NP CPAK 5
PHY
CPAK 6
NP CPAK 7
PHY …
Up to
14x120G
Switch Fabric (SM15)
Switch Fabric
Switch Fabric
Distributed Two-Stage Packet Processing
• Ingress lookup yields packet egress port and applies ingress features
• Egress lookup performs packet-rewrite and applies egress features
Uniform packet flow for simplicity and predictable performance
12
1
2
BRKARC-2003 38
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
ASR 9000 Life of a Packet – Tomahawk LC
• MACSEC Decryption• G.709/OTN/WAN-
PHY/LAN-PHY• Line Clocking
PHY
FIA
CPUCPU
Switch Fabric
CPAK 0
NP FIACPAK 1
PHY
TM
12
3
4
5
NPTM
6
7
8
1
• Ingress L2/L3 FIB lookup, ACL/QoS lookup
• Ingress PBR/ABR, ACL, uRPF• Ingress QoS: classification,
marking, policing• Packet Punting• Ingress ECMP/LAG hashing
2
Ingress side of LC
Egress side of LC
• Ingress Queuing Processing
• Bypassed in case no ingress queuing support
3
• Buffering packet from NP• Requesting fabric credit• Manage superframe and
load-balancing packet across fabric
• Manage system VoQ
4
• Re-assembling packets from superframe
• Send packet to corresponding NP
• Release buffer and fabric credit
5
• Egress L2/L3 FIB lookup, ACL/QoS lookup
• Egress PBR/ABR, ACL, uRPF• Egress QoS: classification,
marking, policing, shaping• Incomplete Adj Packet Punting• Egress ECMP/LAG hashing
6
• MACSEC Encryption• G.709/OTN/WAN-
PHY/LAN-PHY• Line Clocking
8
• Egress Queuing Processing
7
BRKARC-2003 39
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
ASR 9000 Switch Fabric High-Level Architecture3-Stage Non-Blocking Fabric (Separate Unicast and Multicast Crossbars)
FIAFIA
FIARSP0
Arbiter
fabric
RSP1
Arbiter
fabric
fabricFIAFIA
FIA
fabric
Fabric frame format:
Super-frame
Fabric load balancing:
Unicast is per-packet
Multicast is per-flow
Virtual Output Queue
Arbitration
Stage 1 Stage 2 Stage 3
TomahawkTyphoon LC
Ingress LinecardEgress Linecard
Active-Active Fabric
BRKARC-2003 40
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Upto 1.2T linerate/LC capacity
7 Fabric: 6+1 redundancy
800G linerate /LC (protected) capacity – all
Typhoon, Tomahawk LCs (except
Skyhammer) interoperate at full throughput
SFC2
SM15
5-Fabric LC
1.2T with 6 Fabric
1.4T with 7 Fabric
7-Fabric LC
SM15
200G
1T with 5 Fabrics
800G w/ 4 Fabrics
200G
ASR 99xx Switch Fabric – SFC2 and Mixed Gen LCs
100G 83G100G
BRKARC-2003 41
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 42BRKARC-2003
Switch Fabric Arbitration
Fabric ASIC
and VOQ
SFC or RSP1
Arbitration
Fabric
ASIC
Fabric
ASIC
Fabric
ASIC
Fabric
ASIC
ArbitrationFabric ASIC
and VOQ SFC or RSP0
1: Fabric Request
3: Fabric Grant
2: Arbitration
4: load-balanced transmission across fabric links
5: credit return
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
• Unicast traffic sent across first available fabric link to destination (maximizes efficiency)
• Each frame (or super frame) contains sequencing information
• All destination fabric ASIC have re-sequencing logic
• Additional re-sequencing latency is measured in nanoseconds
Fabric Load Balancing – Unicast
Fabric ASIC
and VOQ
SFC or RSP1
Arbitration
Fabric
ASIC
Fabric
ASIC
Fabric
ASIC
Fabric
ASIC
ArbitrationFabric ASIC
and VOQ SFC or RSP04 3 2 1
BRKARC-2003 43
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
RSP880
RSP880
Tomahawk Line Card
FIA
Fabric
SM15
FIA
FIA
FIA
NPSFP+SFP+SFP+
SFP+SFP+
NP
NP
NP
SFP+SFP+SFP+
SFP+SFP+
SFP+SFP+SFP+
SFP+SFP+
SFP+SFP+SFP+
SFP+SFP+
Distributed Multi-Stage Multicast Replication
CPAK 0
NP FIA
FIA
FIA
FIA
CPAK 1
PHY
CPAK 2
NPCPAK 3
PHY
CPAK 4
NPCPAK 5
PHY
CPAK 6
NPCPAK 7
PHY
Up to
14x120
G
Switch Fabric(SM15)
Tomahawk Line Card
FIA
Fabric
SM15
FIA
FIA
FIA
NPSFP+SFP+SFP+
SFP+SFP+
NP
NP
NP
SFP+SFP+SFP+
SFP+SFP+
SFP+SFP+SFP+
SFP+SFP+
SFP+SFP+SFP+
SFP+SFP+
1
1. 512k MGID/FGID lookup
Flow based (RBH key calc)
multi link load balance
2
3
2 & 3. Flow based (using RBH)
multi link load balance
4
4. 22-bit FGID for slot bitmask
based unique fabric LC
replications and FPOE lookup
5
5
5. MGID lookup into FPOE
table (256k entries) for
unique FIA replications
6
6
6
6
6. MGID in FIA DI table lookup
for unique NPU replication
Flow based load balance
7
7
7. MGID lookup for NPU
to egress OIF interface
replication
7
7
BRKARC-2003 44
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
• Multicast traffic hashed based on (S,G) info to maintain flow integrity
• Very large set of multicast destinations preclude re-sequencing
• Multicast traffic is non arbitrated – sent across a different fabric plane
Fabric Load Balancing – Multicast
Fabric ASIC
and VOQ
SFC or RSP1
Arbitration
Fabric
ASIC
Fabric
ASIC
Fabric
ASIC
Fabric
ASIC
ArbitrationFabric ASIC
and VOQ SFC or RSP0A
1
A
2
B
1
A
3
B
2
C
1
Flows exit in-order
BRKARC-2003 45
ASR 9000 QoSArchitecture
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
ASR 9000 Priority-Based QoS Architecture
• Ingress (sub-)interface QoS Queues
• User Configurable Ingress QoS Policy
• 4xVOQ per VQI• Up to 8K VOQs per TSK FIA
(vs 4k per SKT FIA)
4x Egress Destination Qs per VQI, aggregated at egress port rate
• Dedicated Traffic Manager(TM) for Traffic Queuing
• User Configurable QoS Policy on Ingress/Egress NP
• End-to-End priority propagation Guarantee bandwidth, low latency for high priority traffic
• Unicast VOQ and back pressure
User-configuration with Ingress MQC
Egress (sub-)interface QoS Queues
User-configuration with Egress MQC
Implicit ConfigurationNot User-controllable
Ingress side of LC Egress side of LC
NP0 PHY
NP2 PHY
NP3 PHY
NP1 PHYFIA
CPUCPU
Switch Fabric
3
4
1 2 34
CPAK 0
NP FIACPAK 1
PHY
1
2
P1
P2
P3
BE
P1
P2
P3
BE
P1
P2
P3
BETM
TM
P1
P2
P3
BE
BRKARC-2003 47
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Default Implicit Trust Model
internal
cos = 1L2 Bridging
802.1p = 1 *
802.1p = 1IPP=5 DSCP=44
IPP=5 DSCP=44
IPP=5 DSCP=44
internal
cos = 0L2 Bridging
UntaggedIPP=5 DSCP=44
802.1p = 0 *IPP=5 DSCP=44
IPP=5 DSCP=44
Ingress line card Egress line card
Carried in internal buffer header,
by default, internal cos is used for impositioned fields only,
For example, added vlan tag, impositioned MPLS label,
It doesn’t include VLAN tag translation or MPLS label swap
ASR 9000 would never modify
packet DSCP/IP without
a policy-map configured
L2 IF: trust outer Cos
L3 IF: trust DSCP
L3 MPLS: trust outer EXP
BRKARC-2003 48
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Define Traffic ClassConfigure Class-Map to Identify Different Classes of Traffic
Match on Different L2/L3 and System Internal Parameters
Associate Policies/Action to Each Traffic ClassConfigure Policy-Map
Policer, Shaper, Queuing, Bandwidth, Marking/Remarking
Always class-default to Match all No-Match Traffic
Attach Policy to TargetL2/L3 Physical Main Interface/Sub-interface
Bundle-Ether, BVI, PWHE Main Interface/Sub-interface
Both Ingress and Egress Directions
Class-map:
Classification
Policy-map:Action to Class
Policy Attachment
MQC – Three-Step QoS Configuration Model
BRKARC-2003 49
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
ASR 9000 QoS Classification Criteria
L2 Header Fields L3 Header Fields Internal Marking
L2
Interfaces/EF
Ps
Or L3
Interfaces
Inner/outer COS,
inner/outer vlan,
DEISource/Destination MAC address*Ethernet Access-List
Outer EXPDSCP/TOSTTL, TCP flags, Source/destination L4 portsProtocolSource/Destination IPv4address*Access-List
Flow-key/CAC
Discard-classQos-groupTraffic-class
Notes: - Support match all or match any
- Max 8 match statements per class, max 8 match entries per match statement
- Not all header fields can be used in one MQC policy-map, see details next
BRKARC-2003 50
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
•Priority Level 1•Priority Level 2•Shape, •bandwidth or bandwidth remaining•(W)RED•Police•Set (marking)
•Priority Level 1•Priority Level 2•Shape, •bandwidth or bandwidth remaining•(W)RED•Police•Set (marking)
•Shape
•Bandwidth
remaining
•1R2C policer
L4
Child policy (child of
Parent) with user-
defined classes &
class-default
L3Parent Policy on L2 (EFP) or L3 subint. class-
default or physical port /w user-
defined classes
L2Grand-Parent
Policy
L1
Port
scheduler
not
configurable
N/A
policy-map grand-parent (configures L2 layer)
class-default
shape average x
service-policy parent
policy-map parent (configures L3 layer)
class c1
priority level 1
police rate y mbps
class c2
class-default
service-policy child
policy-map child (configures L4 layer)
class-1
priority level 2
class-2
bandwidth remaining percent x
class class-default
int GigE 0/1/2/3.10 l2transport
encapsulation dot1q 10
service-policy output parent
3-Layer Hierarchical QoS (H-QoS)
L0
Port Group
scheduler
not
configurable
N/A
BRKARC-2003 51
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 52BRKARC-2003
H-QoS – Supported Classification/Policy
Policy-map hierarchy level Classification support Policy Support
Grand-parent Only class-default
• Shape Average
• Bandwidth remaining
• 1R2C policer with only drop/transimit
action(no set/mark, Tomahawk card only)
Parent
User defined fields with
restrictions based on
format/interface types.
• Priority/WRED Queue and Queue-limit on
Leaf only
• Policer/Shaper/Marking/non-Priority
Queue/Bandwidth/Bandwidth Remaining
Child
User defined fields with
restrictions based on
format/interface types.
• Priority/WRED Queue and Queue-limit on
Leaf only
• Policer/Shaper/Marking/non-Priority
Queue/Bandwidth/Bandwidth Remaining
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Default Interface Queues
P1 P2 P3 L
Level 4
Level 3
Level 2
Level 1
Queues
Schedulers
Schedulers
Schedulers
BRKARC-2003 53
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 54BRKARC-2003
MQC Hierarchy in Queuing ASIC
policy-map child
class c1
priority level 1
police rate 640 kbps
class c2
bandwidth 20 mbps
class class-default
bandwidth 1 mbps
!
policy-map parent
class class-default
shape average 35 mbps
service-policy child
!
interface GigabitEthernet0/0/0/0
service-policy output parent
P1 P2 P3 L
Port default queues
c1 c2 cd-c
MQC queues
Inactive entity
Active entity
cd-p
L1
L2
L3
L4
cd-c
cd-p
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 55BRKARC-2003
MQC Hierarchy in Queuing ASICpolicy-map child
class c1
priority level 1
police rate 640 kbps
class c2
bandwidth 20 mbps
class class-default
bandwidth 1 mbps
!
policy-map parent
class class-default
shape average 35 mbps
service-policy child
!
interface GigabitEthernet0/0/0/0.1
service-policy output parent
!
interface GigabitEthernet0/0/0/0.2
service-policy output parent
Port default queues
c1 c2 cd-c
G0/0/0/0.1
Inactive entity
Active entity
P1 P2 P3 L c1 c2 cd-c
G0/0/0/0.2
cd-p
L1
L2
L3
L4
cd-p
cd-c
cd-p
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 56BRKARC-2003
MQC Hierarchy in Queuing ASIC
policy-map child
class c1
priority level 1
police rate 640 kbps
class c2
bandwidth 20 mbps
class class-default
bandwidth 1 mbps
!
policy-map parent
class c3
shape average 35 mbps
service-policy child
class class-default
bandwidth 1 mbps
!
policy-map grand-parent
class class-default
shape average 35 mbps
service-policy parent
!
interface GigabitEthernet0/0/0/0.1
service-policy output grand-parent
Port default queues
c1 c2 cd-c
G0/0/0/0.1
P1 P2 P3 L cd-p-vc
c3
L1
L2
L3
L4
cd-p
cd-c
cd-p
cd-gpcd-gp
IOS-XR & IOS-XR 64 Bit
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
IOSd
Kernel
Linux-BinOSH
oste
d A
pp
1
Hoste
d A
pp
2
Operational Infra
Virtualization Layer
IOS “Blob”
Cisco IOS Cisco IOS-XE Classic IOS-XR IOS-XR 64 Bit
XR Code v2
Kernel
Linux, 64bit
Distributed Infra
BG
P
OS
PF
PIM
AC
L
QoS
LP
TS
SN
MP
XM
L
NetF
low
Kernel
Linux, 64bit
System
Admin
XR Code v1
Kernel
Linux, 64bit
Distributed Infra
BG
P
OS
PF
PIM
AC
L
QoS
LP
TS
SN
MP
XM
L
NetF
low
Kernel
QNX, 32bit
Distributed Infra
BG
P
OS
PF
PIM
AC
L
QoS
LP
TS
SN
MP
XM
L
NetF
low
Control
Plane
Data
Plane
Mgmt
Plane
1990s 2000s 2003-04 Present Day
Cisco IOS – A Recap
Incremental Development, with Industry leading investment protection
BRKARC-2003 58
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 59BRKARC-2003
IOS XR Evolution: XR 64 Bit Architecture
IOS-XR
IOS XR
QNX
QNX
Linux
Linux
Routing Apps
SystemAdmin
Routing Control Plane
Admin Plane
LC-CPUs
RP
Lin
e C
ard
LC-CPU
Separate Admin Plane
64 bit Linux Kernel
Linux VM
64-bit IOS XR.
Admin Plane
Linux-based Virtualized
IOS XR 64 Bit
RP
Lin
e C
ard
IOS-XR
IOS-XR
Linux Linux
Linux Linux
Classic IOS XR
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 60BRKARC-2003
IOS XR 64 Bit Internal Connection
Routing Control Plane
Admin Plane
LC-CPUs
Admin Plane
RP
1
Lin
e C
ard
IOS-XR
IOS-XR
Guest OS Guest OS
Guest OS Guest OS
Host OS
Routing Control Plane
Admin Plane
RP
2
IOS-XR
Guest OS Guest OS
Host OS
Host OS
LC-CPUs
Admin Plane
IOS-XR
Guest OS Guest OS
Host OS
10.0.2.0/24 10.0.2.0/24
10.0.2.0/24 10.0.2.0/24
192.0.0.0/8
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 61BRKARC-2003
IOS XR 64 Bit CLI Access
Linux
Linux
Admin Plane
LC-CPUs
Admin Plane
RP
Line Card
IOS-XR
IOS-XR
Linux Linux
Linux Linux
Admin CLI
Admin Linux CLI
Host Linux CLI
XR CLI
XR Linux CLI
run
admin
run
ssh my_host
Console
Mgmt
• Level of Access Controlled by User Authority
• Linux CLI: Only root-lr/root-system users
• ”exit” to return to previous level
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 62BRKARC-2003
Operations in Different Access CLI—on XR VM
Start from IOS XR CLI
“run” to access XR Linux CLI
Linux CLI Execution
“exit” to return to previous CLI
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 63BRKARC-2003
Operations in Different Access CLI—on admin VM
Start from IOS XR CLI
“admin” to access admin CLI
“run” to access admin Linux CLI
Linux CLI Execution
Predefined hostname in /etc/hosts file.
“exit” to return to previous CLI
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 64BRKARC-2003
Operations in Different Access CLI—Host Linux CLI
Start from IOS XR CLI
“admin” to access admin CLI
“run” to access admin Linux CLI
“ssh my_host” to host Linux CLI
Linux CLI Execution
”virsh list” to show two VMs:
Admin and XR VMs
“exit” to return to previous CLI
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 65BRKARC-2003
Default Management Interface Mapping
Routing Control Plane
Admin Plane
RP
1
IOS-XR
Guest OS Guest OS RP
2 Routing Control Plane
Admin Plane
IOS-XR
Guest OS Guest OS
IOS-XR
QNX
Routing Apps
SystemAdminR
P1
Classic IOS XR
IOS-XR
QNX
Routing Apps
SystemAdminR
P2
IOS XR 64 Bit
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
How to Copy Files from/to Admin VM
BRKARC-2003 66
Admin Plane IOS-XR
Linux Linux
TFTP
x
No admin MGMT connection
Admin Plane IOS-XR
Linux Linux
TFTP
With admin MGMT connection
1
2
1. In admin linux CLI, “scp $file $act_RSP_XR:/$dir”. Replace act_RSP_XRwith the actual active RSP XR hostname
2. In XR exec CLI, use the copy command. TFTP/FTP/HTTP options supported
1
1. In admin CLI, use the copy command. Only TFTP option.
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 67BRKARC-2003
Same XR Control Plane: XR Feature Configurations Stay Unchanged
Different Boot Process
Classic XR: ROMMON. CLI based. Support TFTP/USB boot. No ZTP capability
XR 64 Bit: Open-source iPXE. Menu/Scripts process control.IPv4/IPv6 support. TFTP/FTP/HTTP/USB boot. ZTP capable. Directly boot from ISO instead of Turboboot.
Different CLI to Reload Chassis
Classic XR: “Reload location all” from ADMIN exec prompt to reload whole chassis
XR 64 Bit: “hw-module location all reload” or “reload rack 0” from ADMIN exec prompt to reload the whole chassis. “reload location all” reload the VMs only.
XR 64 Bit image upgrade: Admin VM XR VM
XR 64 Bit: Added Install commands (update/upgrade) for flexible package management
XR 64 Bit: Added “install activate issu” to support ISSU operation
Admin CLI changes: Configuration, Exec and Show commands
XR Exec and Show command CLI change. No major XR configuration CLI change
XR 64 Bit Operational Differences
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
IOS XR 64 Bit Packaging
Bootable Images
Minimum Image asr9k-mini-x64.iso-6.1.2.iso Core packages: OS, Admin, Forwarding, Modular
Services Card, Basic Routing, SNMP, Alarm Correlation
asr9k-minik9-x64.iso-6.1.2.iso Core packages + Encryp/decryp, SSL/SSH/PKI
Golden ISO Customized ISO image including mini ISO+required packages and SMUs
Optional Feature Packages
asr9k-eigrp-x64-1.0.0.0-r612.x86_64.rpm
asr9k-isis-x64-1.1.0.0-r612.x86_64.rpm
asr9k-ospf-x64-1.1.0.0-r612.x86_64.rpm
asr9k-m2m-x64-2.0.0.0-r612.x86_64.rpm
asr9k-mgbl-x64-3.0.0.0-r612.x86_64.rpm
asr9k-mpls-te-rsvp-x64-1.2.0.0-r612.x86_64.rpm
asr9k-mpls-x64-2.1.0.0-r612.x86_64.rpm
asr9k-mcast-x64-2.0.0.0-r612.x86_64.rpm
asr9k-optic-x64-1.0.0.0-r612.x86_64.rpm
asr9k-li-x64-1.1.0.0-r612.x86_64.rpm
asr9k-k9sec-x64-3.1.0.0-r612.x86_64.rpm
BRKARC-2003 68
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 69BRKARC-2003
Migrating Classic XR to IOS XR 64 Bit
FPD Upgrade
Pre-6.1.2
5.x.y
4.x.y
Post-6.1.2
6.2.x
6.3.x
Two Ways to Migrate:
• CSM Orchestrated Migration
• MOP: Manual Migration
• Follow below link for details
XR 6.1.2
or later
http://www.cisco.com/c/en/us/td/docs/routers/asr9000/migration/guide/b-migration-to-ios-xr-64-bit/b-migration-to-ios-xr-64-bit_chapter_00.html
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 70BRKARC-2003
CSM Environment For Image Management
CSM
Image Repository
CCOLocal
Server
TFTP/
SFTP/
FTP
• CSM Provides Web-GUI Interface
• Manage Software Packages for
Group of Network Devices
• Local Packages Repository or CCO
• TFTP/SFTP/FTP Needed
• Files can be pre-copied on router
• Follow the link for CSM details
http://www.cisco.com/c/en/us/td/docs/routers/asr90
00/software/smu/csmuser.html
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
IOS XR 64 Bit -- Technical Benefits
Virtualized Architecture
ISSU Ready
MPL, ZTL
Reduced Upgrading Time
High Scale
e.g. 10M IPv4/5M IPv6
Open Linux-kernel Baesd
Linux RPM Packaging
Offline Package Management
Single Bootable ISO Package
APP Hosting
Open and Secure
Multi-Level API
iPXE
Zero-Touch Provisioning
IOS XR 64 Bit
Telemetry
Both XR 64 Bit & Classic XR
BRKARC-2003 71
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 72BRKARC-2003
System Architectural Support for ISSUDifferent Version IOS XR Simultaneously Running
V1
Slice 2
LC (ZPL/MPL)RP (ZTL/MTL)
Ad
min IOS
XR V1
Linux Slice 1 Slice 3 Slice 4
V2
Ad
min IOS XR
V1
IOS XR
V2
Linux
LC-CPU
Tomahawk only
Ad
min IOS
XR V2
Linux
Active StandbyStandby Active
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Application Hosting
• Device Extension with off the shelf applications
• Multi-level API access
• Native Linux APIs
• ASIC SDK level APIs
• OS level APIs
• Complete tool chain integration
• Easy to use and integrate with customer build
environments
Hardware
Linux
Cisco OS
Linux Based
Applications
Apps Built for
Cisco OS
FWD ASIC SDK
Ap
ps A
pps
Ap
ps
Ap
ps
BRKARC-2003 73
App Hosting
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public 74BRKARC-2003
App Hosting: Modes of Hosted Applications
HOST OS/HYPERVISOR
Ad
min
Pla
ne
VM
XR VM APP Process
in XR VMAPP in Container
Linux Guest OS
XR Lib/Pkgs
App Lib/Pkgs App Lib/Pkgs
LXC/Docker
Container
RP CPU
Linux
App Hosting
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Conclusion
• ASR 9000 - Truly Carrier-Class Edge Router Provides:
• Rich Features, Flexible Service Capability
• Variety of Hardware to Meet Different Capacity Requirements
• Fully Distributed Architecture for High Performance and System Scalability
• Uniform, Open and Modularized Software Architecture for Simple Operation and Efficient Failure Protection
75BRKARC-2003
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
• Give us your feedback to be entered into a Daily Survey Drawing. A daily winner will receive a $750 gift card.
• Complete your session surveys through the Cisco Live mobile app or on www.CiscoLive.com/us.
Complete Your Online Session Evaluation
Don’t forget: Cisco Live sessions will be available for viewing on demand after the event at www.CiscoLive.com/Online.
© 2017 Cisco and/or its affiliates. All rights reserved. Cisco Public
Continue Your Education
• Demos in the Cisco campus
• Walk-in Self-Paced Labs
• Lunch & Learn
• Meet the Engineer 1:1 meetings
• Related sessions
BRKARC-2003 77
Thank you