CloudEngine 6800&5800 V100R003C00 Product Description 01

CloudEngine 6800&5800 Series Switches
V100R003C00
Product Description
Issue 01
Date 2013-12-31
HUAWEI TECHNOLOGIES CO., LTD.

Copyright Huawei Technologies Co., Ltd. 2013. All rights reserved.
No part of this document may be reproduced or transmitted in any form or by any means without prior written
consent of Huawei Technologies Co., Ltd.

Trademarks and Permissions
and other Huawei trademarks are trademarks of Huawei Technologies Co., Ltd.
All other trademarks and trade names mentioned in this document are the property of their respective holders.

Notice
The purchased products, services and features are stipulated by the contract made between Huawei and the
customer. All or part of the products, services and features described in this document may not be within the
purchase scope or the usage scope. Unless otherwise specified in the contract, all statements, information,
and recommendations in this document are provided "AS IS" without warranties, guarantees or representations
of any kind, either express or implied.
The information in this document is subject to change without notice. Every effort has been made in the
preparation of this document to ensure accuracy of the contents, but all statements, information, and
recommendations in this document do not constitute a warranty of any kind, express or implied.

Huawei Technologies Co., Ltd.
Address: Huawei Industrial Base
Bantian, Longgang
Shenzhen 518129
People's Republic of China
Website: http://enterprise.huawei.com
Issue 01 (2013-12-31) Huawei Proprietary and Confidential
Copyright Huawei Technologies Co., Ltd.
i
About This Document
Intended Audience
This document describes the positioning, characteristics, usage scenarios, functions, system
architecture, operations and maintenance, and specifications of CloudEngine (CE) switches.
This document helps you understand the characteristics and features of CE switches.
This document is intended for:
l Network planning engineers
l Hardware installation engineers
l Commissioning engineers
l Data configuration engineers
l Onsite maintenance engineers
l Network monitoring engineers
l System maintenance engineers
Symbol Conventions
The symbols that may be found in this document are defined as follows.
Symbol Description
Indicates an imminently hazardous situation
which, if not avoided, will result in death or
serious injury.
Indicates a potentially hazardous situation
which, if not avoided, could result in death or
serious injury.
which, if not avoided, may result in minor or
moderate injury.
Product Description About This Document
ii
Symbol Description
which, if not avoided, could result in
equipment damage, data loss, performance
deterioration, or unanticipated results.
NOTICE is used to address practices not
related to personal injury.
NOTE
Calls attention to important information, best
practices and tips.
NOTE is used to address information not
related to personal injury, equipment damage,
and environment deterioration.

Change History
Changes between document issues are cumulative. The latest document issue contains all the
changes made in earlier issues.
Issue 01 (2013-12-31)
Initial commercial release.
Product Description About This Document
iii
Contents
About This Document.....................................................................................................................ii
1 Product Positioning and Characteristics...................................................................................1
1.1 Product Positioning.........................................................................................................................................................2
1.2 Product Characteristics...................................................................................................................................................2
1.2.1 High Performance and High Port Density...................................................................................................................2
1.2.2 Front-to-Rear/Rear-to-Front Ventilation Channels.....................................................................................................3
1.2.3 Simplified Operations and Maintenance.....................................................................................................................4
1.2.4 Easy-to-Deploy, Easy-to-Maintain High-Performance Stacking................................................................................4
1.2.5 Abundant Data Center Service Features......................................................................................................................5
2 Typical Applications.....................................................................................................................7
2.1 ToR Application.............................................................................................................................................................8
2.2 EoR Application.............................................................................................................................................................9
3 Structures of CE6800&5800 Series Switches..........................................................................10
4 Product Features...........................................................................................................................14
4.1 Feature List...................................................................................................................................................................15
4.2 Ethernet Features..........................................................................................................................................................21
4.2.1 Link Aggregation.......................................................................................................................................................21
4.2.2 Interface-based Flow Control....................................................................................................................................22
4.2.3 Traffic Suppression....................................................................................................................................................22
4.2.4 VLAN........................................................................................................................................................................22
4.2.5 QinQ..........................................................................................................................................................................23
4.2.6 VLAN Mapping.........................................................................................................................................................23
4.3 STP/RSTP/MSTP.........................................................................................................................................................24
4.3.1 STP and RSTP...........................................................................................................................................................24
4.3.2 MSTP.........................................................................................................................................................................24
4.3.3 MSTP Protection.......................................................................................................................................................24
4.4 Port Security.................................................................................................................................................................25
4.5 Link Detection..............................................................................................................................................................25
4.6 IP Features....................................................................................................................................................................26
4.6.1 IPv4/IPv6 Dual-Stack................................................................................................................................................26
4.6.2 IPv4............................................................................................................................................................................26
Product Description Contents
iv
4.6.3 IPv6............................................................................................................................................................................26
4.6.4 IPv6 Transition Technologies....................................................................................................................................27
4.7 Routing.........................................................................................................................................................................27
4.8 Multicast.......................................................................................................................................................................27
4.8.1 Layer 2 Multicast.......................................................................................................................................................27
4.8.2 Layer 3 Multicast.......................................................................................................................................................28
4.9 VPN Features................................................................................................................................................................29
4.10 QoS.............................................................................................................................................................................29
4.10.1 Traffic Classification...............................................................................................................................................29
4.10.2 Access Control and Re-Marking.............................................................................................................................30
4.10.3 Traffic Policing........................................................................................................................................................30
4.10.4 Congestion Management.........................................................................................................................................31
4.10.5 Congestion Avoidance.............................................................................................................................................31
4.10.6 Rate-limiting on an Interface...................................................................................................................................31
4.10.7 Two-Rate-Three-Color............................................................................................................................................31
4.11 Security.......................................................................................................................................................................32
4.11.1 Device Security........................................................................................................................................................32
4.11.2 Service Security.......................................................................................................................................................33
4.12 MAC-Forced Forwarding...........................................................................................................................................34
4.13 DHCP..........................................................................................................................................................................35
4.14 Network Management................................................................................................................................................36
4.14.1 LLDP.......................................................................................................................................................................36
4.14.2 NQA.........................................................................................................................................................................36
4.14.3 NetStream................................................................................................................................................................37
4.14.4 sFlow.......................................................................................................................................................................38
4.15 Smart Link and Multi-Instance...................................................................................................................................38
4.16 Stacking......................................................................................................................................................................39
4.17 Data Center Features..................................................................................................................................................39
4.17.1 TRILL......................................................................................................................................................................39
4.17.2 DCB.........................................................................................................................................................................40
4.17.3 FCoE........................................................................................................................................................................40
4.17.4 VM Detection..........................................................................................................................................................40
4.17.5 Forwarding Based on the VEPA.............................................................................................................................41
4.17.6 NLB Cluster Association.........................................................................................................................................42
5 Operation Maintenance and Network Management ...........................................................43
5.1 Maintenance and Management.....................................................................................................................................44
5.1.1 Configuration Modes.................................................................................................................................................44
5.1.2 Monitoring and Maintenance.....................................................................................................................................44
5.1.3 Diagnosis and Debugging..........................................................................................................................................45
5.1.4 Software Upgrade and In-Service Patching...............................................................................................................46
5.1.5 Hardware Fault Handling..........................................................................................................................................46
v
5.2 eSight............................................................................................................................................................................46
6 System Parameters.......................................................................................................................48
6.1 Specifications................................................................................................................................................................49
vi
1 Product Positioning and Characteristics
About This Chapter
1.1 Product Positioning
1.2 Product Characteristics
Product Description 1 Product Positioning and Characteristics
1
1.1 Product Positioning
NOTICE
The CloudEngine 6800&5800 (CE6800&5800) series data center switches are class A products
and may cause radio interference. Customers should take appropriate preventative measures.
Huawei CE6800&5800 series switches are next-generation data center switches designed for
high-performance data centers. The CE6800 series switches support 10GE access, while the
CE5800 series switches support GE access.
The CE6800&5800 series switches function as access switches. As shown in Figure 1-1, servers
connect to CE6800&5800 switches through 10GE/GE uplinks; and CE6800&5800 switches
connect to core switches CE12800 through 10GE/40GE uplinks.
Figure 1-1 CE6800&5800 switches on the network
CE12800
CE6800/5800

1.2 Product Characteristics
CE6800&5800 switches use industry-leading hardware and software platforms. Their optimized
system architecture and features provide higher performance, larger capacity, and richer services
for data centers. In addition, these switches are easy to install and maintain.
1.2.1 High Performance and High Port Density
CE6800&5800 switches use cutting-edge hardware platforms in the industry. By using a 1 U (1
U = 44.45 mm) box, CE6800&5800 switches provide high port densities and line-rate forwarding
capabilities. Next-generation, high-performance servers in super high density arrangements can
easily connect to CE6800&5800 switches.
2
l CE5810-24T4S-EI:Provides twenty-four 10/100/1000BASE-T Ethernet ports, four 10G
SFP+ Ethernet optical ports.
l CE5810-48T4S-EI:Provides forty-eight 10/100/1000BASE-T Ethernet ports, four 10G
SFP+ Ethernet optical ports.
l CE5850-48T4S2Q-EI: Provides forty-eight 10/100/1000BASE-T Ethernet ports, four 10G
SFP+ Ethernet optical ports, and two 40G QSFP+ Ethernet optical ports.
l CE5850-48T4S2Q-HI: provides forty-eight 10/100/1000BASE-T Ethernet ports, four 10G
SFP+ Ethernet optical ports, and two 40G QSFP+ Ethernet optical ports. The switch also
supports a large buffer to handle burst traffic.
l CE6850-48S4Q-EI: Provides forty-eight 10G SFP+ Ethernet optical ports and four 40G
QSFP+ Ethernet optical ports
l CE6850-48T4Q-EI: Provides forty-eight 10G BASE-T Ethernet ports and four 40G QSFP
+ Ethernet optical ports
CE6800&5800 switches provide high-performance 40GE ports, which can connect to high-
density 40GE line processing units (LPUs) on CE12800 switches to construct full-40G data
center networks.
1.2.2 Front-to-Rear/Rear-to-Front Ventilation Channels
CE6800&5800 switches use front-to-rear/rear-to-front ventilation channels. This design isolates
cold air from hot air channels, improves heat dissipation efficiency, and lowers power
consumption, without the need to reconstruct racks in the data center equipment room.
Figure 1-2 and Figure 1-3 show the front-to-rear/rear-to-front ventilation channels on
CE6800&5800 switches. The airflow direction in the ventilation channels can be changed by
configuring fan modules and power modules.
Figure 1-2 Front-to-rear ventilation channels
1 2 3 4 5 6 7 8 9 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 48 1 2
1 2
3 4
Cold air
Hot air
PAC-150WA ~100-240V;50/60Hz,2.5A
STATUS
PAC-150WA ~100-240V;50/60Hz,2.5A
STATUS STATUS
FAN-40EA-F
STATUS
FAN-40EA-F
CONSOLE
ETH
SYS
MST
ACT
L/A
ID
PWR1 FAN1 FAN2 PWR2
CE5850-48T4S2Q-EI
3
Figure 1-3 Rear-to-front ventilation channels
1 2 3 4 5 6 7 8 9 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 48 1 2
1 2
3 4
SYS
MST
STAT
SPEED
STACK
MODE/ID
CE5850-48T4S2Q-EI
Cold air
Hot air
STATUS
FAN-40EA-B
STATUS
FAN-40EA-B
CONSOLE
ETH
SYS
MST
ACT
L/A
ID
PWR1 FAN1 FAN2 PWR2
CE5850-48T4S2Q-EI
PAC-150WA ~100-240V;50/60Hz,2.5A
STATUS
PAC-150WA ~100-240V;50/60Hz,2.5A
STATUS
1.2.3 Simplified Operations and Maintenance
CE6800&5800 switches' architecture separates the data plane from the management plane.
l The management ports, fan modules, and power modules are at the front side of the switch
for easy maintenance.
l The data ports are at the rear side of the switch to facilitate cabling and maintenance.
CE6800&5800 switches optimize indicators in the following aspects to facilitate easy
maintenance of data center networks with high device densities:
l Redundant system indicator
CE6800&5800 switches have system indicators on both the front side (with management
ports) and rear side (with data ports). These system indicators show the system status and
stack status, helping administrators easily monitor the system status.
l Easy-to-read port indicator
Innovative 40G port indicators clearly show the running status of 40GE ports that have
been converted into four 10GE ports.
l Easy-to-maintain stack indicator
The stack indicator shows the role and ID of the switch in a stack system, helping
administrators maintain the stack system.
l Innovative positioning indicator
CE6800&5800 switches have a positioning indicator that allows administrators to remotely
position a switch quickly. Administrators can turn on switches' positioning indicators
through the network management system (NMS) or console so that they can quickly find
the switches that require maintenance. Positioning indicators are blue, making them easy
to find.
1.2.4 Easy-to-Deploy, Easy-to-Maintain High-Performance
Stacking
4
A maximum of 16 CE6800&5800 switches can be added to a stack system. The stack system
has the following advantages:
l High performance
In the stack, a maximum of 768 GE or 10GE interfaces are allowed.
l High bandwidth
Stacked CE5800 switches support 80 Gbit/s stack bandwidth, and stacked CE6800 switches
support 160 Gbit/s stack bandwidth, making a stacking bandwidth bottleneck unlikely to
occur.
l Fast recovery
The ring stack topology allows for system recovery within 200 ms.
l Easy to deploy and maintain
The pre-deployment and offline configuration functions allow users to plan and pre-
configure the system and add devices on demand. This feature offers a Pay As You
Grow solution.
Users can specify device IDs in a stack system to easily identify, locate, and maintain
devices.
Indicators clearly identify the role and status of the device in a stack system. With these
indicators, users can complete basic maintenance tasks on a stack system without a PC.
l Simple upgrade
A stack system supports quick software upgrades and automatic software upgrades,
simplifying the upgrade process and reducing workload.
1.2.5 Abundant Data Center Service Features
CE6800&5800 switches have a wide range of data center service features, including the
following:
l Fiber Channel over Ethernet (FCoE) and Data Center Bridging (DCB)
FCoE, Data Center Bridging Exchange (DCBX) in 802.1Qaz, Priority-based Flow
Control (PFC) in 802.1Qbb, and Enhanced Transmission Selection (ETS) in 802.1Qaz
allow fiber channels (FCs) to run on a converged lossless Enhanced Ethernet, thereby
lowering networking costs.
FCoE and DCB can seamlessly interconnect with the existing FC infrastructure,
protecting investments in the FC storage area network (FC SAN).
NOTE
Only CE6800 supports FCoE and DCB.
l Virtualization and virtual machine (VM) access
Server virtualization improves data center efficiency.
VM detection enables switches to automatically migrate network policies during VM
migrations, so network sources can be allocated on demand. With the technologies that
enable large Layer 2 networking, VMs can migrate freely across the entire data center
network.
l Transparent Interconnection of Lots of Links (TRILL)
5
TRILL is an Internet Engineering Task Force (IETF) standard that allows for super
large, flexible networking.
TRILL implements multi-path load balancing to balance traffic among multiple paths
in response to service requirements.
TRILL can quickly detect network changes and complete network convergence within
a short time.
6
2 Typical Applications
About This Chapter
2.1 ToR Application
2.2 EoR Application
Product Description 2 Typical Applications
7
2.1 ToR Application
Top of rack (ToR) is a cabling mode in a server cabinet. Switches deployed in ToR mode are
called ToR switches. The ToR mode applies to data center networks with high server densities.
As shown in Figure 2-1, ToR switches are deployed at the top of server cabinets. Two ToR
switches in two adjacent server cabinets form a stack system, and servers are dual-homed to the
two ToR switches. The access ports on the two ToR switches constitute a link aggregation group
(LAG).
Figure 2-1 ToR application
Aggregation Switch
RACK
Switch
Server
Server
Server
Server
Server
Server
Server
Server
Switch
Server
Server
Server
Server
Server
Server
Server
Server
RACK

ToR networking has the following advantages:
l The stack system can eliminate bandwidth bottlenecks. In the stack system, ToR switches
are stacked using 10GE/40GE ports, and all stack cables work in Active state, greatly
improving stack bandwidths.
l The access reliability of the stack system is high. Master and backup ports on servers are
connected to two ToR switches simultaneously, and the access ports on the two ToR
switches work in LAG mode. Therefore, the Spanning Tree Protocol (STP) is not required,
and a switchover is completed within 100 ms once a fault has occurred.
l ToR switches support 40GE uplink ports that can be used together with high-density 40GE
LPUs on CE12800 switches to construct high-performance 40GE data center networks.
8
2.2 EoR Application
End of row (EoR) is a cabling mode in a server cabinet. Switches deployed in EoR mode are
called EoR switches.
The EoR mode applies to data center networks with low densities of servers. At the end of each
row of server cabinets, there are two network cabinets where access switches are installed.
Servers in cabinets along the row share these access switches. As shown in Figure 2-2, access
switches in the network cabinets form a stack system and provide high access port densities.
Figure 2-2 EoR application
40G
Switch
Switch
Switch
Switch
Switch
Switch
Switch
Switch
Switch
Switch
Switch
Switch
Switch
Switch
Switch
Switch

EoR networking has the following advantages:
l High access port densities. A maximum of 16 CE6800&5800 switches can be added to a
stack system and provide more than 768 GE/10GE access ports.
l Improved user experience. CE6800&5800 switches use the same operating system as the
CE12800 and support chassis architecture. After being stacked, CE6800&5800 switches
can work as a chassis switch and provide the same application experience as the CE12800.
l Super-high stack bandwidth. The EoR networking mode eliminates the stack system
bottleneck to the maximum extent and improves performance of the entire system.
9
3 Structures of CE6800&5800 Series Switches
Front View
NOTE
The figures in this document are for reference only.
The front views (power supply side) of the CE6800&5800 series switches are the same except
for the device model silkscreens. Figure 3-1 shows the front view of a CE5850-48T4S2Q-EI
switch as an example.
Figure 3-1 CE5850-48T4S2Q-EI front view (power supply side)
1 2 3 4 56 8 7
l 1. Power supply slot 1
l 2. Power supply slot 2
Available power modules:
l CE5800 series: 150 W AC
power module, 350 W DC
power module
l CE6850-48S4Q-EI: 350 W
AC power module, 350 W
DC power module
l CE6850-48T4Q-EI: 350 W
AC power module, 600 W
AC power module
l 3. Fan slot 1
l 4. Fan slot 2
Available fans:
l CE5810 series: FAN-40SB
series fan modules
l CE5850 series: FAN-40EA
series fan modules
l CE6800 series: FAN-40EA
series fan modules
NOTE
Earlier versions of the
CE5850-48T4S2Q-EI use the
FAN-40SA serial fan modules.
5. Console port
NOTE
This port is used for first-
time login or local device
configuration.
Product Description 3 Structures of CE6800&5800 Series Switches
10
6. ETH management port
NOTE
This port is used for local or remote
device configuration.
7. Bar code label
NOTE
This label is drawable, and you
can pull it outward to view the bar
code and MAC address of the
switch.
8. USB port
NOTE
This port is used for device
deployment, configuration
file transfer, and upgrade.
Rear View
Figure 3-2, Figure 3-3, Figure 3-4, Figure 3-5, Figure 3-6, and Figure 3-7 show rear views
(port side) of CE6800&5800 chassis.
Figure 3-2 CE5810-24T4S-EI rear view (port side)
7 4
Figure 3-3 CE5810-48T4S-EI rear view (port side)
1 4
Figure 3-4 CE5850-48T4S2Q-EI rear view (port side)
5 1 4
Figure 3-5 CE5850-48T4S2Q-HI rear view (port side)
1 2 3 4 5 6 7 8 9 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 48 1 2 3 4
SYS
MST
STAT
SPEED
STACK
MODE/ID
CE5850-48T4S2Q-HI
1 40GE
Breakout
2 3 4
1 2
1 4 5
Figure 3-6 CE6850-48S4Q-EI rear view (port side)
3 6
11
Figure 3-7 CE6850-48T4Q-EI rear view (port side)
2 6
1. Forty-eight
10/100/1000BASE-T Ethernet
electrical ports
2. Forty-eight 10GBASE-T
Ethernet electrical ports
NOTE
When a CE6850-48T4Q-EI switch
uses 350 W AC power modules and all
its ports are in use, the length of each
network cable used on the switch
cannot exceed 30 m.
3. Forty-eight 10GE
SFP+ Ethernet optical
ports
4. Four 10GE SFP+ Ethernet
optical ports
5. Two 40GE QSFP+ Ethernet
optical ports
NOTE
l A 40GE QSFP+ port of a
CE5850-48T4S2Q-EI switch
cannot be converted into four
10GE SFP+ ports.
l A 40GE QSFP+ port of a
CE5850-48T4S2Q-HI switch can
be converted into four 10GE SFP
+ ports.
6. Four 40GE QSFP+
Ethernet optical ports
NOTE
A 40GE QSFP+ port of
a CE6800 switch can be
converted into four
10GE SFP+ ports.
7. Twenty-four
10/100/1000BASE-T Ethernet
electrical ports
- -
Side View
Figure 3-8, Figure 3-9, Figure 3-10, and Figure 3-11 show side views of CE6800&5800
chassis.
Figure 3-8 CE5800 side view (left side)
1 2
1 1
2
2 2
12
Figure 3-9 CE5800 side view (right side)
1 2
1 1
2
2 2 3
Figure 3-10 CE6800 side view (left side)
1
2
1 1
2
2 2 4 4
4 4
Figure 3-11 CE6800 side view (right side)
1
2
1 1
2
2 2 4 4
4 4
3
1. Three port-side
mounting holes for
mounting brackets
2. Four power-supply-
side mounting holes
for mounting brackets
3. Ground screw 4. Four middle
mounting holes for
mounting brackets
13
4 Product Features
About This Chapter
4.1 Feature List
4.2 Ethernet Features
4.3 STP/RSTP/MSTP
4.4 Port Security
4.5 Link Detection
4.6 IP Features
4.7 Routing
4.8 Multicast
4.9 VPN Features
4.10 QoS
4.11 Security
4.12 MAC-Forced Forwarding
4.13 DHCP
4.14 Network Management
4.15 Smart Link and Multi-Instance
4.16 Stacking
4.17 Data Center Features
Product Description 4 Product Features
14
4.1 Feature List
Table 4-1 lists the features supported by CE6800&5800 switches.
Table 4-1 Features supported by CE6800&5800 switches
Feature Description
Ethernet Ethernet l Operating modes of full-duplex and auto-negotiation
l Operating rates of an Ethernet interface, including 10 Mbit/
s, 100 Mbit/s, 1000 Mbit/s, 10 Gbit/s, 40 Gbit/s, and auto-
negotiation
NOTE
l GE electrical interfaces support the operating rates of 10 Mbit/
s, 100 Mbit/s, and 1000 Mbit/s.
l 10GE electrical interfaces support the operating rates of 100
Mbit/s, 1000 Mbit/s, and 10 Gbit/s.
l 10GE optical interfaces support the operating rates of 1000
Mbit/s and 10 Gbit/s.
l 40GE optical interfaces support only the operating rate of 40
Gbit/s.
l Flow control on interfaces
NOTE
10GE/40GE optical interfaces do not support flow control.
l Jumbo frames
l Link aggregation
l Load balancing among links within a trunk
l Interface isolation and forwarding restrictions
l Broadcast storm suppression
Virtual Local
Area
Network
(VLAN)
l Multiple interface types: access, trunk, hybrid, and 802.1Q-
in-802.1Q (QinQ)
l Multiple access modes: access, trunk, hybrid, and QinQ
l VLAN assignment: port-based, MAC address-based, IP
subnet-based VLAN assignment
l VLAN aggregation
l MUX VLAN
l Transparent Transmission of Protocol Packets in a VLAN
l 1 to 1 VLAN mapping.
15
Feature Description
Media
Access
Control
(MAC)
l Automatic learning and aging of MAC addresses
l Static, dynamic, and blackhole MAC address entries
l Packet filtering based on source MAC addresses
l Interface-based MAC learning limiting
Link Layer
Discovery
Protocol
(LLDP)
Support for LLDP
Virtual
interface
configuratio
n table
(VCT)
Support for VCT
Ethernet
loop
protection
Multiple
Spanning
Tree
Protocol
(MSTP)
l Spanning Tree Protocol (STP)
l Rapid Spanning Tree Protocol (RSTP)
l MSTP
l Bridge protocol data unit (BPDU) protection, root
protection, and loop protection
l Partitioned STP
IP features Address
Resolution
Protocol
(ARP)
l Static and dynamic ARP entries
l ARP in a VLAN
l Aging of ARP entries
l ARP and Reverse Address Resolution Protocol (RARP)
l ARP proxy
l Auto-detection
IPv6 l IPv4/IPv6 dual-stack
l Neighbor Discovery (ND)
l IPv6 over IPv4 Manual Tunnel
l IPv6 over IPv4 GRE Tunnel
l 6to4 Tunnel
Dynamic
Host
Configuratio
n Protocol
(DHCP)
l DHCP server
l DHCP relay
l DHCPv6 Relay
l DHCP snooping
16
Feature Description
IP
forwardin
g
Unicast
routing
l IPv4/IPv6 static routing
l Routing Information Protocol version 1 (RIP-1), RIP-2, and
RIPng
l )Open Shortest Path First (OSPF), including OSPFv2 and
OSPFv3
l Intermediate System to Intermediate System (IS-IS)
l Border Gateway Protocol 4.0 (BGP4) and Border Gateway
Protocol for IPv6 (BGP4+)
l Routing protocol
l Policy-based routing
l Unicast Reverse Path Forwarding (URPF) check
l Virtual Routing Forwarding (VRF)
VPN l GRE Tunnel
l Virtual Routing Forwarding (VRF)
Multicast
routing
l Internet Group Management Protocol Version 1/2/3
(IGMPv1/v2/v3)
l PIM-SM (IPv4) and PIM-SM (IPv6)
l PIM-SSM (IPv4) and PIM-SSM (IPv6)
l MLDv1 and MLDv2
l MLD SSM Mapping
l Multiprotocol BGP (MBGP)
l Multicast Source Discovery Protocol (MSDP)
l Multicast routing policies
l Reverse Path Forwarding (RPF)
l Bidirectional PIM (IPv4) and Bidirectional PIM (IPv6)
17
Feature Description
Device
reliability
Bidirectional
Forwarding
Detection
(BFD)
l BFD (IPv4) and BFD (IPv6)
l Association between BFD and Eth-Trunk
l BFD for OSPF
l BFD for OSPFV3
l BFD for IS-IS
l BFD for IS-IS IPV6
l BFD for BGP
l BFD for BGP4+
l BFD for PIM(IPv4)
l BFD for PIM(IPv6)
l BFD for static routing(IPv4)
l BFD for static routing(IPv6)
l BFD for VRRP
l BFD for VRRP6
Others l Virtual Router Redundancy Protocol (VRRP) and VRRP6
l Device Link Detection Protocol (DLDP)
l Smart Link
l Ethernet in the First Mile (EFM), defined in 802.3ah
Layer 2
multicast
features
Layer 2
multicast
l IGMP snooping
l IGMP proxy
l Fast leave
l Multicast traffic control
l Multicast VLAN
QoS
features
Traffic
classification
l Traffic classification based on combination of the L2
protocol header, IP 5-tuple, outbound interface, and 802.1p
priority
l Traffic classification based on the C-VID and C-PRI of QinQ
packets
Traffic
behavior
l Access control after traffic classification
l Traffic policing based on traffic classifiers
l Re-marking based on the traffic classification result
l Class-based packet queuing
l Association between traffic classifiers and traffic behaviors
18
Feature Description
Queue
scheduling
l Priority queuing (PQ) scheduling
l Deficit round robin (DRR) scheduling
l PQ+DRR scheduling
l Weighted round robin (WRR) scheduling
l PQ+WRR scheduling
Congestion
avoidance
Weighted Random Early Detection (WRED)
Rate limiting
on outbound
interfaces
Rate limiting on outbound interfaces
Virtualizat
ion
Many-to-one
virtualizatio
n
l Intelligent Stack (iStack)
l Stack split and merge
l Dual-active detection
l Version and configuration synchronization
Data
center
features
Transparent
Interconnecti
on of Lots of
Links
(TRILL)
TRILL features
IGMP over TRILL
NOTE
CE5810 does not support TRILL.
Data Center
Bridging
(DCB)
l Data Center Bridging Exchange Protocol (DCBX)
l Priority-based Flow Control (PFC)
l Enhanced Transmission Selection (ETS)
NOTE
Only CE6800 supports DCB.
Fiber
channel over
Ethernet
(FCoE)
FIP Snooping Bridge (FSB)
NOTE
Only CE6800 supports FCoE.
Virtual
awareness &
server
association
l Virtual awareness
l Automatic policy deployment
l Automatic policy migration
l Cluster of NLB servers
Association between virtual IP addresses of NLBs and
multicast MAC addresses
Association between one multicast MAC address and
multiple outbound interfaces
19
Feature Description
Forwarding
mode based
on the
Virtual
Ethernet Port
Aggregator
(VEPA)
Forwarding mode based on the VEPA
Configura
tion and
maintenan
ce
Terminal
service
l Command line configuration
l Error messages and online help in English and Chinese
l Login through console and Telnet terminals
l Send function and data communications between terminal
users
File system l Directory and file management
l File upload and download using File Transfer Protocol (FTP)
and Trivial File Transfer Protocol (TFTP)
Debugging
and
maintenance
l Unified management of logs, alarms, and debugging
information
l Electronic labels
l User operation logs
l Detailed debugging information for network fault diagnosis
l Network test tools such as tracert and ping commands
l Port mirroring and traffic mirroring (remote mirroring
supported)
Version
upgrade
l Device software loading and in-service software loading
l In-service upgrade using the basic input/output system
(BIOS) menu
l In-service patching
l ZTP
20
Feature Description
Security
and
managem
ent
System
security
l Hierarchical command-line protection based on user levels,
preventing unauthorized users from using commands to
access CE6800&5800 switches
l Secure Shell (SSH)
l RADIUS (IPv4) and RADIUS (IPv6) authentication for
login users
l HWTACACS (IPv4) and HWTACACS (IPv6)
authentication for login users
l Access control list (ACL) filtering
l Dynamic ARP inspection (DAI)
l DHCP packet filtering (appending the Option 82 field)
l Defense against control packet attacks
l Defense against attacks of source address spoofing, LAND,
SYN flood (TCP SYN), smurf, ping flood (ICMP echo),
teardrop, and ping of death
l Logs about attacking MAC addresses
Network
management
l ICMP-based ping and tracert
l Simple Network Management Protocol Version 1/2c/3
(SNMPv1/v2c/v3)
l Standard management information base (MIB)
l Remote network monitoring (RMON)
l NetStream, with output statistics packets in the V5, V8, or
V9 format
l sFlow
l Network quality analysis

4.2 Ethernet Features
4.2.1 Link Aggregation
Link aggregation binds multiple physical interfaces on one or more devices into a logical
interface (such as an Eth-Trunk). This logical interface is called a load balancing group or a link
aggregation group (LAG).
After link aggregation, traffic on a logical interface is load balanced among the physical
interfaces bound to the logical interface. When a physical interface fails, the system switches
traffic on this interface to the other physical interfaces so that services are not interrupted. When
the physical interface recovers, the system re-distributes traffic among physical interfaces to
ensure balanced traffic distribution.
21
CE6800&5800 switches support link aggregation and implement load balancing based on any
of the following information:
l Source MAC address
l Destination MAC address
l Source and destination MAC addresses
l Source IP address
l Destination IP address
l Source and destination IP addresses
l Transport-layer source port
l Transport-layer destination port
l Transport-layer source and destination ports
l User-defined load balancing modes for Layer 2 packets, IPv4 packets, and MPLS packets
Link aggregation technology increases transmission bandwidth and improves link reliability
efficiently and cost-effectively, without the need to upgrade hardware.
4.2.2 Interface-based Flow Control
Flow control based on interfaces is a method for congestion management. CE6800&5800
switches implement flow control on interfaces using a hardware backpressure mechanism. When
an interface works in full-duplex mode, CE6800&5800 switches implement flow control on the
interface in accordance with the IEEE 802.3x standard.
When congestion occurs, CE6800&5800 switches send consecutive Pause frames to the
upstream device, requesting the upstream device to stop sending data for a specified period of
time. When the upstream device receives the Pause frames, it reduces the volume of traffic sent
from its outbound interface. Interface-based flow control takes effect on all traffic types.
4.2.3 Traffic Suppression
Traffic suppression limits the number of unknown unicast packets, multicast packets, and
broadcast packets to within a proper range to ensure network efficiency.
On the CE6800&5800 series switches, you can configure traffic limits for unknown unicast
packets, multicast packets, and broadcast packets. When the rate of these packets on an interface
or a VLAN exceed the limits, the switches drop excess packets to control the traffic rate within
a proper range, ensuring normal operations of network services.
The CE6800&5800 series switches can also control the percentages of unknown unicast packets,
multicast packets, and broadcast packets on an interface.
4.2.4 VLAN
A local area network (LAN) can be divided into several logical LANs. Each logical LAN is a
broadcast domain, called a virtual LAN or VLAN. To put it simply, devices on a LAN are
logically grouped into different LAN segments, regardless of their physical locations. VLANs
isolate broadcast domains on a LAN.
22
VLAN Aggregation
To implement communication between VLANs on CE6800&5800 switches, users can configure
VLANIF interfaces and assign an IP address to each VLANIF interface. This implementation,
however, wastes IP addresses when there are many VLANs. To conserve IP addresses, VLAN
aggregation is introduced.
VLAN aggregation means that multiple VLANs are aggregated into a super-VLAN. The VLANs
that form the super-VLAN are called sub-VLANs. A super-VLAN cannot contain physical
interfaces, but a VLANIF interface can be created for the super-VLAN and be configured with
an IP address. A sub-VLAN can contain physical interfaces, but a VLANIF interface cannot be
created for the sub-VLAN. Interfaces in all the sub-VLANs use the IP address of the VLANIF
interface created for the super-VLAN. Each sub-VLAN functions as an independent broadcast
domain, which conserves the IP addresses previously used for common VLANs.
MUX VLAN
MUX VLAN isolates Layer 2 traffic between ports on a VLAN. On an enterprise network for
example, MUX VLAN can isolate client ports but enable them to communicate with server ports.
This function involves a principal VLAN and several subordinate VLANs. Subordinate VLANs
are classified into subordinate group VLANs and subordinate separate VLANs. Ports on
subordinate VLANs can communicate with ports on the principal VLAN. Ports on a subordinate
group VLAN can communicate with each other but cannot communicate with ports on other
subordinate group VLANs. Ports on a subordinate separate VLAN cannot communicate with
each other.
4.2.5 QinQ
The 802.1Q-in-802.1Q (QinQ) protocol is a Layer 2 tunnel protocol based on the IEEE 802.1Q
technology. A frame transmitted on the public network has double 802.1Q tags (one for the
public network and the other for the private network).
In most cases, telecom carriers define VLANs on the public network, and users define VLANs
on their own private networks. Therefore, different private networks may use the same VLAN
IDs. Using the QinQ function, CE6800&5800 switches add public VLAN tags to packets coming
from private networks. Then the private VLAN tag of these packets becomes the inner VLAN
tag. In this way, packets from private networks are transmitted transparently on the public
network, keeping private networks separate from the public network.
4.2.6 VLAN Mapping
When new branches are planned on the same network segment as old branches, and VLAN IDs
in new branches are inconsistent with those in old branches, use VLAN mapping so that VLAN
deployment in old branches does not need to be changed. VLAN mapping allows users to flexibly
plan VLAN IDs, prevents VLAN ID conflicts, and implement communication between new and
old branches.
Implementation Modes
The device supports VLAN-based and MQC-based VLAN mapping. There are three VLAN-
based VLAN mapping modes:
23
l 1 to 1 VLAN mapping
When an interface configured with VLAN mapping receives a single-tagged packet, the
interface maps the VLAN tag in the packet to a new VLAN tag.
When an interface configured with VLAN mapping receives a double-tagged packet, the
interface maps the outer tag of the packet to a specified tag and transparently transmits the
inner tag as the data.
When an interface configured with VLAN mapping receives a double-tagged packet, the
interface maps the inner and outer VLAN tags in the packet to new inner and outer VLAN
tags.
MQC-based VLAN mapping uses a traffic classifier to classify packets based on VLAN IDs,
associates the traffic classifier with a traffic behavior defining VLAN mapping so that the device
can re-mark the VLAN ID in packets matching the traffic classifier. MQC-based VLAN mapping
implements differentiated services.
4.3 STP/RSTP/MSTP
4.3.1 STP and RSTP
The Spanning Tree Protocol (STP) and the Rapid Spanning Tree Protocol (RSTP) are link-layer
management protocols that are primarily applied to LANs to prevent loops. STP blocks
redundant links and trims a network into a loop-free tree topology. RSTP enhances STP,
providing fast transition of interface status to speed up network convergence.
STP and RSTP prevent broadcast storms caused by loops and provide backup links for data
forwarding.
4.3.2 MSTP
The Multiple Spanning Tree Protocol (MSTP) was developed based on STP and RSTP. MSTP
divides a switching network into multiple regions. Based on VLAN tags, each region has several
spanning trees that are independent of each other. As a result, the entire network is trimmed into
a loop-free tree topology, to prevent broadcast storms.
MSTP associates VLANs with spanning trees so that packets of different VLANs are transmitted
along different spanning trees. This speeds up network convergence and implements load
balancing.
Unlike STP and RSTP, MSTP provides multiple redundant data forwarding links to implement
load balancing among VLANs.
4.3.3 MSTP Protection
24
BPDU Protection
MSTP-enabled CE6800&5800 switches provide bridge protocol data unit (BPDU) protection.
When BPDU protection is enabled, CE6800&5800 switches shut down the edge port that
receives a BPDU, instead of turning the edge port into a non-edge port. In this case, the spanning
tree is not recalculated, and network flapping is prevented.
Root Protection
MSTP-enabled CE6800&5800 switches provide root protection. CE6800&5800 switches
protect the role of the root device by retaining the role of the designated port.
When the designated port with root protection enabled receives a BPDU of a higher priority than
the local port priority, the port does not change to a non-designated port. Instead, it enters the
Listening state and stops forwarding packets. After the designated port has not received BPDUs
of a higher priority for a long time, it returns to the forwarding state. This prevents network
flapping.
Loop Protection
CE6800&5800 switches with loop protection enabled block the root port if the root port fails to
receive any BPDUs from an upstream device. If the blocked port receives BPDUs again, it
becomes the root port and enters the Forwarding state. If the blocked port does not receive
BPDUs, it remains blocked and cannot forward packets. In this way, loops are prevented on the
network.
4.4 Port Security
Port security is a security mechanism that controls network access. This mechanism checks
whether the source MAC addresses of data frames received on an interface are valid. When
detecting packets with invalid source MAC addresses, the mechanism takes appropriate action
to protect the interface.
After port security is enabled, CE6800&5800 switches consider the following types of MAC
addresses valid:
l Static MAC addresses that are manually configured
l Dynamic or static MAC addresses in the DHCP snooping table
l Dynamic MAC addresses that are learned before the number of learned MAC addresses
reaches the upper limit
When an interface receives frames with invalid source MAC addresses, CE6800&5800 switches
discard the frames or generate an alarm.
4.5 Link Detection
CE6800&5800 switches support link detection. This link detection feature provides two means
to detect link faults on LANs: loopback detection and virtual cable test (VCT).
25
l Loopback detection is used to detect whether loops exist on a LAN. CE6800&5800
switches send specific packets to detect loops over the entire LAN.
l VCT is used to estimate network cable length and locate failure points on the cable.
CE6800&5800 switches simulate radar to detect cable faults and locate the failure points
along a single link.
4.6 IP Features
NOTE
If you need IPv6 features on CE12800 switches, buy licenses from Huawei.
4.6.1 IPv4/IPv6 Dual-Stack
IPv4/IPv6 dual-stack features good interoperability and easy implementation. Figure 4-1 shows
the IPv4/IPv6 dual-stack structure.
Figure 4-1 IPv4/IPv6 dual-stack structure
IPv4/IPv6 Application
TCP UDP
Link Layer
I P v 4 I P v 6

4.6.2 IPv4
The CE6800&5800 supportthe following IPv4 features:
l TCP/IP protocol suite, including ICMP, IP, TCP, UDP, socket (TCP/UDP/Raw IP), and
ARP
l Static DNS, which the DNS server address manually specified
l FTP server/client and TFTP client
l DHCP relay, DHCP server, and DHCP snooping
l Ping, tracert, and NQA.
l Bidirectional Forwarding Detection (BFD) features, including BFD for OSPF, BFD for
ISIS, BFD for BGP, and BFD for PIM
4.6.3 IPv6
The CE6800&5800 supportthe following IPv6 features:
26
l IPv6 Neighbor Discovery (ND)
l Path MTU Discovery (PMTU)
l TCP6, ping IPv6, tracert IPv6, socket IPv6, UDP6, and Raw IP6
4.6.4 IPv6 Transition Technologies
Most IPv6 networks are upgraded from IPv4 networks. When an IPv4 network is being upgraded
to an IPv6 network, some issues occur, for example, IPv6 islands. IPv6 transition technologies
address these issues. IPv6 transition technologies include the dual stack technology and
tunneling technology.
CE series switches support the following IPv6 transition technologies:
l Dual stack technology
IPv4/IPv6 dual stack
l Tunneling technology
IPv6 over IPv4 Manual Tunnel
IPv6 over IPv4 GRE Tunnel
6to4 Tunnel
4.7 Routing
The CE12800 supports the following routing features:
l Static routes that are manually configured by the administrator to simplify network
configurations and improve network performance
l IPv4 routing protocols, including RIPv1/v2, OSPF, IS-IS (IPv4), and BGP
l IPv6 routing protocols, including RIPng, OSPFv3, IS-IS (IPv6), and BGP4+
l Virtual routing forwarding (VRF) multi-instance and IP address overlapping
l Optimal route selection using routing policies
4.8 Multicast
4.8.1 Layer 2 Multicast
The CE6800&CE5800 series switches support Layer 2 multicast (IP multicast at the data link
layer). Layer 2 multicast implements on-demand forwarding of multicast data within a broadcast
domain. This feature conserves network bandwidth and improves security of data transmission.
The CE6800&CE5800 series switches support the following Layer 2 multicast functions:
l IGMP snooping: This function is deployed on a switch between hosts and a multicast router.
The switch not only supports static multicast forwarding entries, but also generates dynamic
Layer 2 multicast forwarding entries with multicast groups, VLANs, and outbound
interfaces by listening to IGMP messages exchanged between the hosts and multicast
router. When the CE6800&CE5800 series switches receive multicast cast data packets,
27
they forward the packets to the receivers in the corresponding VLAN according to the Layer
2 forwarding table.
l IGMP snooping proxy: This function is deployed on a switch between hosts and a multicast
router. The switch functions as a proxy server between the hosts and multicast server. It
terminates the IGMP messages sent from the hosts to the multicast router and substitutes
for the hosts to respond to the IGMP Query messages sent from the multicast router. In
addition, the switch processes IGMP messages exchanged between the hosts and multicast
router to create Layer 2 multicast forwarding entries, and forward multicast data according
to the Layer 2 multicast forwarding entries.
l Fast leave: When a group member leaves a group, the host of the member sends an IGMP
Leave message. When a Layer 2 multicast switch (CE6800&CE5800) receives the IGMP
Leave message, it resets the aging timer for corresponding multicast forwarding entry and
deletes the connected member port from the entry when the aging timer expires. If each
port in a VLAN connects to only one host, you can configure the fast leave function on the
switch. With this function enabled, the switch deletes the corresponding multicast
forwarding entry on a member port immediately after receiving an IGMP Leave message
from the member port. This function conserves network bandwidth and system resources
and realizes fast service switching.
l Multicast traffic control: When the CE6800&CE5800 series switches receive unknown
multicast packets that do not have forwarding entries in the multicast forwarding table, they
drop the packets or broadcasts them in the VLAN to which the inbound interface belongs.
They can also control the multicast traffic rate by limiting the percentage of inbound
multicast traffic on an Ethernet interface.
l Multicast VLAN: A multicast VLAN aggregates and forwards multicast traffic. If hosts in
user VLANs request multicast data, bind the user VLANs to the multicast VLAN. Then
multicast data is replicated from the multicast VLAN to the user VLANs, realizing inter-
VLAN multicast replication. The CE6800&CE5800 series switches can replicate multicast
data from one multicast VLAN to a maximum of 128 user VLANs. Multicast flows are
forwarded to the multicast VLAN first, replicated at the distribution points according to
the multicast forwarding table, and then distributed to the associated user VLANs. Multicast
VLAN technology enables the CE6800&CE5800 series switches to aggregate multicast
flows in one or more multicast VLANs, saving bandwidth on the network. As unicast and
multicast data flows are transmitted in different VLANs, multicast VLAN technology
facilitates management and control of multicast flows and reduces wastes of bandwidth.
4.8.2 Layer 3 Multicast
The CE6800&CE5800 series switches support Layer 3 multicast (IP multicast at the network
layer). Layer 3 multicast enables multicast data to be forwarded over an IP network and allows
user hosts to join the multicast network.
The CE6800&CE5800 series switches support the following Layer 3 multicast functions:
l IPv4 multicast protocols: including the Internet Group Management Protocol (IGMP),
Protocol Independent Multicast Any-Source Multicast (PIM-ASM), Protocol Independent
Source-Specific Multicast (PIM-SSM), Bidirectional PIM (BIDIR-PIM), and Multicast
Source Discovery Protocol (MSDP).
l IPv6 multicast protocols: including PIM-ASM (IPv6), PIM-SSM (IPv6), BIDIR-PIM
(IPv6), and Multicast Listener Discovery (MLD).
28
l Line-speed forwarding of IP multicast traffic.
l ASM and SSM models.
l Anycast rendezvous point (RP): Multiple RPs can exist in a domain and establish MSDP
peer relationships. A multicast source can register with the nearest RP, and a receiver can
join the rendezvous point tree (RPT) of the nearest RP. In this manner, load balancing is
implemented among the RPs. When an RP fails, sources and receivers registered with this
RP choose another nearest RP. This RP redundancy mechanism enhances network
reliability.
l Multicast static routes
l Route filtering: The multicast routing module can filter multicast routes it receives and
advertises using routing policies. It can also use routing policies to filter and forward IP
multicast packets.
l Reverse path forwarding (RPF) check.
4.9 VPN Features
The CE6800&5800 supports the following virtual private network (VPN) features:
l MPLS/BGP IPv4 L3VPN
l MPLS/BGP IPv6 L3VPN
l VPLS and VPLS over GRE
l GRE Tunnel
4.10 QoS
CE6800&5800 switches provide a class-based QoS mechanism and support 802.1p priorities
for minimizing end-to-end delay and jitter and optimizing bandwidth.
CE6800&5800 switches classify traffic based on specific rules and take actions on traffic to
better support value-added services such as next generation network (NGN) services, IPTV, and
broadband access. The actions include priority re-marking, traffic policing, congestion
management, congestion avoidance, and rate limiting on an interface.
4.10.1 Traffic Classification
Traffic classification assesses packet header information against a set of rules to identify packets
of a certain type. For example, the 802.1p priority of packets sent by the operating support system
(OSS) and NMS is set to 7; the 802.1p priority of voice over IP (VoIP) packets is set to 6; the
802.1p priority of broadcast TV (BTV) packets and video on demand (VoD) packets is set to 5
or 4; the 802.1p priority of packets sent by virtual private network (VPN) users is set to 3, 2, or
1 according to the level of VPN users; and the 802.1p priority of Internet access service packets
is set to 0. Packets are classified based on their 802.1p priorities.
CE6800&5800 switches use a hardware classifier to ensure line-rate transmission of service data
on interfaces.
Users can define rules to classify packets and specify the relationships between rules.
29
l and: Packets match a traffic classifier only when the packets match all the rules.
l or: Packets match a traffic classifier as long as the packets match one of the rules.
Table 4-2 describes the traffic classification rules.
Table 4-2 Traffic classification rules
Layer Traffic Classification Rule
Layer 2 l VLAN ID in the outer tag of a VLAN packet
l VLAN ID in the inner tag of a VLAN packet
l 802.1p priority in the outer tag of a VLAN packet
l 802.1p priority in the inner tag of a VLAN packet
l Source MAC address
l Destination MAC address
Layer 3 l DSCP priority in IP packets
l IP precedence in IP packets
l IP protocol type
Layer 4 l TCP SYN flag in TCP packets
Others l Inbound interface
l Outbound interface
l ACL

4.10.2 Access Control and Re-Marking
After traffic classification, CE6800&5800 switches perform access control on packets by
permitting or rejecting the packets. When packets are permitted, CE6800&5800 switches re-
mark the following information in the packets:
l 802.1p priority (PRI field in the VLAN tag)
l DSCP field
l Precedence field of IP packets
4.10.3 Traffic Policing
CE6800&5800 switches use a token bucket algorithm to police and control incoming traffic,
implementing the committed access rate (CAR).
The traffic rate is controlled by limiting the speed at which tokens are placed in the token bucket.
When the traffic rate exceeds the upper limit, CE6800&5800 switches discard excess traffic so
that the traffic remains within an acceptable range. This function saves network resources and
protects the investments of customers.
30
4.10.4 Congestion Management
CE6800&5800 switches use queue scheduling technologies to implement congestion
management. Each outbound interface on CE6800&5800 switches has eight queues. After
packets are classified, they enter the appropriate queues based on their priorities.
CE6800&5800 switches support the following queue scheduling policies:
l PQ
l WRR
l DRR
l PQ+WRR
l PQ+DRR
4.10.5 Congestion Avoidance
To remove congestion, a switch quickly discards packets to release queue resources and does
not put packets in long-delay queues.
The CE6800&5800 switches support the Weighted Random Early Detection (WRED)
algorithm. WRED monitors packets in each queue, compares the queue length with upper and
lower drop thresholds, and when congestion occurs, processes packets in queues based on the
following rules when congestion occurs:
l When the length of a queue is smaller than the lower drop threshold, no packet is discarded.
l When the length of a queue is between the upper drop threshold and the lower drop
threshold, WRED discards packets randomly based on the maximum drop probability.
l When the length of a queue exceeds the upper drop threshold, all packets in the queue are
discarded.
4.10.6 Rate-limiting on an Interface
Rate-limiting an interface proactively adjusts the rate of traffic on the interface in order to prevent
burst traffic and lower the packet loss rate. CE6800&5800 switches use a token bucket and a
buffer to rate-limit interfaces, implementing traffic shaping. When the rate of packets exceeds
the rate limit, CE6800&5800 switches buffer excess packets and send them after the traffic rate
falls below the rate limit. In this manner, the packet transmission rate is smoothed.
4.10.7 Two-Rate-Three-Color
CE6800&5800 switches control traffic based on the traffic classification results and discard
excess packets when the rate of packets exceeds the rate limit. CE6800&5800 switches support
two-rate-three-color markers. Users can set the following parameters on CE6800&5800
switches:
l Committed information rate (CIR), which is the average rate of traffic that can pass through
an interface
31
l Committed burst size (CBS), which is the average volume of burst traffic that can pass
through an interface
l Peak information rate (PIR), which is the maximum rate of traffic that can pass through an
interface
l Peak burst size (PBS), which is the maximum volume of burst traffic that can pass through
an interface
In addition, CE6800&5800 switches can mark packets red, green, or yellow according to the
traffic volume, map behaviors (such as permit or deny) to the colors, and re-mark packets.
4.11 Security
CE6800&5800 switches ensure both device security and service security.
4.11.1 Device Security
Hierarchical Command Protection
CE6800&5800 switches authenticate users when they are logging in to CE6800&5800 switches
in Telnet mode from an Ethernet interface. Only authenticated users can configure and maintain
CE6800&5800 switches.
CE6800&5800 switches use a hierarchical protection mode for commands, and define four
command levels in ascending order: visit level, monitoring level, configuration level, and
management level. Users are also classified corresponding to the four command levels. Users
can use only the commands at the same or lower level than their own levels. This implementation
effectively controls user rights.
CE6800&5800 switches can combine command levels and user levels to extend to 16 total levels
for more finely grained user management.
Remote SSH Login
CE6800&5800 switches support Secure Shell (SSH). On an insecure network, SSH provides a
security guarantee and authentication functions for user logins and defends against multiple
attacks.
SNMP Encrypted Authentication
CE6800&5800 switches support encrypted authentication through Simple Network
Management Protocol version 3 (SNMPv3). When CE6800&5800 switches are managed by the
network management system (NMS) through SNMP, the encrypted authentication mode in the
user-based security model (USM) can be used to ensure switch security.
AAA
CE6800&5800 switches support Authentication, Authorization, and Accounting (AAA).
Together with hierarchical command protection, CE6800&5800 switches can authenticate and
authorize login users. In addition, CE6800&5800 switches can authenticate NMS users. The
AAA mechanism enables CE6800&5800 switches to prevent unauthorized access.
32
CE6800&5800 switches support multiple authentication methods such as local, Remote
Authentication Dial-In User Service (RADIUS), and Huawei Terminal Access Controller
Access Control System (HWTACACS) authentication modes.
CPU Attack Defense
CE6800&5800 switches can filter the protocol packets and management packets sent to the CPU
based on the protocol ID, port number, or combination of the port number and VLAN ID. This
filtering protects CPU channels from denial of service (DoS) attacks.
4.11.2 Service Security
VLAN Assignment
CE6800&5800 switches support division of a LAN into multiple VLANs. Devices in different
VLANs cannot communicate with each other. This function isolates broadcast domains and
improves service security.
MAC Address Learning Limit on Interfaces
Users can set the maximum number of MAC addresses that a CE6800&5800 interface can learn,
to prevent hackers from initiating source MAC address attacks from the interface. This setting
ensures that the MAC address entries on CE6800&5800 switches will not be used up.
Blackhole MAC Address Entry
CE6800&5800 switches support blackhole MAC address entries. When receiving a packet,
CE6800&5800 switches compare the source or destination MAC address of the packet with
MAC address entries. If the source or destination MAC address of the packet is a blackhole
MAC address entry, CE6800&5800 switches discard the packet.
Once detecting that packets with a specific MAC address are prone to attacks, users can set a
blackhole MAC entry to filter out packets with this MAC address. This setting defends
CE6800&5800 switches against MAC address attacks.
MAC Table Lookup
To improve interface security, CE6800&5800 switches support MAC table lookup based on
VLAN IDs and MAC addresses to improve interface security. The network administrator can
add static entries to the MAC address table. A static entry defines the mapping between a MAC
address and an interface. In this way, devices with specific MAC addresses are bound to
interfaces, which defends CE6800&5800 switches against attacks from packets with forged
MAC addresses.
Port Isolation
Port isolation prevents interfaces on the same CE6800&5800 switches from sending Layer 2
packets to each other. CE6800&5800 switches support unidirectional and bidirectional port
isolation. Port isolation ensures security of user networks and helps construct cost-effective,
intelligent community networks. Port isolation also effectively controls broadcast packets and
increases network throughput.
33
Packet Filtering
Packet filtering is used to filter out invalid or unwanted packets.
CE6800&5800 switches can filter out packets based on user-defined rules, for example, by
checking the MAC address, IP address, port number, and VLAN ID of packets. Packet filtering
does not check the session status or analyze data. The packet filtering technology enables
CE6800&5800 switches to effectively control the packets that pass through.
4.12 MAC-Forced Forwarding
The access layer provides network connections between the user-side hosts and the telecom
carrier-side access routers (ARs), including the reliable connections between the hosts and the
Internet or other IP networks.
The access layer is divided into the user network and aggregation network. The user network is
connected to an access node (AN) through a subscriber line (which is a physical line), and the
AN connects the subscriber line to the aggregation network. Therefore, the AN is the edge
between the subscriber line and the aggregation network. The aggregation network centralizes
and aggregates user traffic. Figure 4-2 shows the connections at the access layer.
Figure 4-2 Connections at the access layer
Switch C
Switch A
Switch B
EAN
EAN
EAN
Gateway
Flow through Gateway
Flow not through Gateway
Server
Server
Server
Server

34
Users have the following requirements at the access layer:
l CE6800&5800 switches must perform Layer 3 forwarding for traffic of different user hosts
in different networks. In this way, switches can filter, schedule, and charge user traffic.
l IPv4 address assignment efficiency needs to be improved to save IPv4 addresses. The
address assignment effectiveness also needs to be improved if addresses are assigned from
a large address pool but not from a small and independent network segment.
To implement user isolation at the access layer and meet the preceding requirements, the MAC-
Forced Forwarding (MFF) protocol is introduced.
MFF is a security protocol and it ensures that user hosts accessing a device with the same media
are isolated at Layer 2. When MFF is running, its security program applies to any shared access
media.
In addition to Layer 2 isolation, the AN running MFF discards any upstream broadcast packets
except for DHCP packets and ARP request packets. Especially, the AN discards DHCP Reply
packets received through the subscriber line and rate-limits the DHCP Broadcast packets.
The AN running MFF must track the IPv4 addresses allocated to the subscriber line. This is to
discard upstream traffic that uses forged source IPv4 addresses.
4.13 DHCP
DHCP Snooping
CE6800&5800 switches can be deployed between the DHCP server and client to listen DHCP
packets that are exchanged. Based on the listening result, CE6800&5800 switches create an IP
+MAC+PORT+VLAN binding table to suppress invalid packets.
In addition, CE6800&5800 switches support the Option 82 field for collecting accurate locations
of DHCP clients.
l After receiving a Request packet from a DHCP client, CE6800&5800 switches append the
Option 82 field to the Request packet. Then, CE6800&5800 switches forward the packet
to the DHCP server. The DHCP server allocates IP addresses based on the Option 82 field.
l The DHCP server appends the Option 82 field to a Reply packet and sends the reply packet
to CE6800&5800 switches. CE6800&5800 switches parse the Option 82 field, determine
the target interface, remove the Option82 field, and then forward the packet to a user.
On CE6800&5800 switches, Option 82 is implemented in two modes: Option 82 insert and
Option 82 rebuild.
The Option 82 field contains the user circuit ID that carries the user device name, inner and outer
VLAN IDs, and port number. Therefore, the Option 82 function effectively prevents attackers
from modifying DHCP packets.
DHCP Relay
The DHCP client and DHCP server broadcast DHCP packets during IP address allocation.
Therefore, DHCP applies only to scenarios when the DHCP client and DHCP server are on the
same subnet. To implement dynamic host configuration, users must configure a DHCP server
on each network segment, which increases costs.
35
The DHCP relay function is introduced to resolve this problem. By using DHCP relay, a DHCP
client in a subnet can communicate with the DHCP server in another subnet and finally obtain
an IP address. In this manner, DHCP clients on multiple network segments can share one DHCP
server. This reduces costs and facilitates centralized management.
DHCP Server
A DHCP server processes requests for address allocation, address lease extending, and address
release from DHCPv6 clients or DHCPv6 relay agents, and allocates IP addresses and other
network configuration parameters to DHCP clients.
The switch can function as a DHCP server and uses a global IP address pool to allocate IP
addresses. You can configure the following DHCP server functions on the switch:
l Configure and IP address pool, address lease, DNS server address, NetBIOS server address,
domain name suffix, and other network parameters. The DHCP server then dynamically
allocates IP addresses and other network configuration parameters to DHCP clients.
l Exclude some IP addresses in the address pool so that they will not be dynamically allocated
to clients.
l Bind IP addresses in the address pool to MAC addresses so that fixed IP addresses can be
allocated to servers or hosts used for special purpose on the network.
l Configure the DHCP server to check whether an IP address is in use by sending ping packets
before allocating the IP address to a client.
4.14 Network Management
4.14.1 LLDP
CE6800&5800 switches support the Link Layer Discovery Protocol (LLDP) that conforms to
IEEE 802.1ab. LLDP is a link layer protocol that a device uses to obtain information about
neighboring devices.
Using LLDP, the local NMS can obtain link-layer information of all devices on the local network
and details about the network topology. This expands the network management scope.
The LLDP-enabled interfaces on a CE6800&5800 periodically notify the neighboring devices
of the local interface status. When the status of an interface changes, the interface sends a status
update message to the directly connected neighboring device. The neighboring device stores the
status update message in the standard SNMP MIB. Then the NMS obtains link-layer information
of the network from the MIB to calculate the topology of the entire network.
4.14.2 NQA
With the launch of more value-added services, telecom carriers and users alike require
increasingly high QoS. Especially with the advent of voice over IP (VoIP) and video over IP
services, telecom carriers and users all tend to require Service Level Agreements (SLAs). To
ensure users with the committed bandwidth, telecom carriers need to collect statistics about
delay, jitter, and packet loss of devices. These statistics help analyze network performance in
timely fashion.
36
CE6800&5800 switches provide the network quality analysis (NQA) function to meet the
preceding requirements. NQA measures the performance of different protocols running on the
network. With NQA, telecom carriers can collect network operation indexes in real time, such
as TCP connection delay and file transfer rate. Based on these indexes, telecom carriers can
provide differentiated network services and charge differently for them. NQA is also an effective
tool for diagnosing network faults.
4.14.3 NetStream
As bandwidth on the Internet increases fast, users need to manage their network resources more
refinedly. NetStream technology can collect statistics about network traffic and usage of network
resources by sampling network traffic. This technology enables network administrators to obtain
detailed records about traffic on their data networks.
Figure 4-3 NetStream networking diagram
NDE
NetStream
NSC
NSC
NDA
NDA
NetStream traffic
traffic
NDE: NetStream Data Exporter NSC: NetStream Collector NDA: NetStream Data Analyzer

NetStream provides the following functions:
l Network management and planning
l Enterprise accounting and department billing
l ISP billing
l Data storage
l Business data collection
An IP a connectionless protocol, a service data flow on an IP network is a group of IP packets
that may be sent from any terminal to another terminal. Most data flows on an IP network are
temporary and bidirectional.
37
NetStream identifies flows of different services based on 7-tuple information consisting of the
destination IP address, source IP address, destination port number, source port number, protocol
ID, Type of Service (ToS), and inbound or outbound interface. After identifying data flows,
NetStream collects statistics for each service separately.
The NDE periodically sends the collected traffic statistics to the NSC. NSC processes the traffic
statistics and sends the statistics to the NDA. NDA analyzes the statistics, generates reports for
accounting and networking planning.
CE6800&CE5800 switches can work as an NDE and support packet-based random sampling to
sample IPv4, IPv6, and MPLS packets. The switches can create original traffic, flexible traffic,
and aggregation traffic and encapsulate NetStream packets V5, V8, or V9 format. They support
the Distributed NetStream model.
4.14.4 sFlow
Sampled Flow (sFlow) is a traffic monitoring technology that samples packets for traffic
statistics collection and analysis.
sFlow provides interface-based traffic analysis and displays traffic statistics in graphs or reports,
facilitating preventive maintenance on enterprise networks, especially for enterprises that do not
have specialized network administrators.
NetStream technology also provides traffic analysis function. As NetStream is implemented
based on traffic information, network devices must collect traffic statistics and save the collected
statistics in their buffers. Statistics are sent to the NetStream Collector (NSC) when their buffers
are full or when the traffic aging time expires. In sFlow application, network devices only need
to sample packets and do not need to save traffic statistics in their buffers. Traffic analysis is
completed by a remote collector. sFlow has the following advantages over NetStream:
l Fewer sources consumed and lower costs: Network devices do not need to save traffic
statistics in their buffers, reducing the network resources consumed and lowering costs.
l Flexible, on-demand collector deployment: Network traffic statistics collection and
analysis are completed by the collector. The traffic collection and analysis functions can
be flexibly configured on the collector according to network characteristics.
4.15 Smart Link and Multi-Instance
Dual-homing is a commonly used networking model. A dual-homing network usually runs the
Spanning Tree Protocol (STP) protocol to implement link redundancy. However, STP
convergence speed is low.
Smart Link can implement fast link switchover while providing redundancy protection. On a
dual-homed device, when the active link fails, the device switches traffic to the standby link to
ensure normal traffic forwarding.
Smart Link is specific to dual-homing networking and features fast convergence (subsecond
convergence), simple configuration, and easy user operation.
Smart Link multi-instance allows you to associate a Smart Link Group with multiple instances
bound to different VLAN ranges. Configure the standby link (using a command) to forward
traffic of some instances. Then data traffic of different VLANs is transmitted over different
paths, realizing loading balancing.
38
4.16 Stacking
Stacking enables switches located in the same place to form a reliable switch group by way of
high-speed uplink interfaces. CE6800&5800 switches implement stacking by multiplexing
10GE/40GE uplink interfaces as stack interfaces. After being stacked, CE6800&5800 switches
can be uniformly managed and maintained, which reduces maintenance costs.
Stack interfaces on CE6800&5800 switches can be bonded to improve bandwidth and to enhance
stacking reliability.
Member switches in a stack system have three roles:
l Master switch
A stack system has only one master switch. The master switch manages the entire stack
system by assigning stack IDs to member switches, collecting information about the stack
topology, and advertising information to all the member switches.
l Standby switch
When the master switch fails, the standby switch becomes the master switch and takes over
all services.
l Slave switch
In a stack system, all member switches except for the master switch are slave switches.
4.17 Data Center Features
4.17.1 TRILL
In the cloud computing era, server virtualization technology is widely used in data centers. VMs
must be able to dynamically migrate within a wide range to enhance service reliability and lower
IT costs. Moreover, collaborative computing between servers generates a large volume of east-
west traffic. Therefore, fat-tree networking is required to implement non-blocking data
forwarding. These service requirements lead to the birth of large Layer 2 networking.
Transparent Interconnection of Lots of Links (TRILL) is introduced to build large Layer 2
networks.
TRILL is an IETF standard to implement large Layer 2 networking. Data packets are
encapsulated in TRILL headers with a TTL value to prevent packet loops. RPF check performed
on multicast packets also effectively prevents broadcast storms caused by loops. Packets are
forwarded along multiple paths to improve bandwidth efficiency. TRILL supports shared links
and can seamlessly interconnect with traditional Layer 2 networks. In addition, many TRILL
parameters are generated automatically, simplifying TRILL deployment. TRILL supports
unicast and multicast services simultaneously and features easy operation and maintenance
(O&M). TRILL is applicable to building large Layer 2 networks. It has the same characteristics
as traditional Layer 2 networks: plug-and-play and flexible deployment. TRILL also addresses
many issues of traditional Layer 2 networks. For example, bandwidth efficiency is low, STP
convergence is slow, and every switch needs to learn MAC addresses from devices on the entire
network.
39
All CE6800&5800 series switches support standard TRILL.
4.17.2 DCB
Data Center Bridging (DCB) is a set of standards designed for enhancing Ethernet technologies
and can transmit service packets on the Ethernet without packet loss. The CE6800 switch
supports the following DCB features:
l Priority-based Flow Control (PFC): controls flows based on priorities. When congestion is
to occur on the queue of a priority on an interface, the switch sends a backpressure signal
to the upstream switch, requesting the upstream device to reduce the rate of packets of the
same priority in the queue according to congestion. This mechanism takes effect on each
node along the path to the source switch. Backpressure can be applied to a single priority
or to several priorities, but cannot be applied to all priorities of packets passing through the
interface.
l Enhanced Transmission Selection (ETS): classifies priorities according to bandwidth
allocation, low latency, or best effort (BE). Flows with identical or similar features are
classified into one priority group so that bandwidth is properly allocated to each group.
l Data Center Bridging Exchange (DCBX): detects the configuration of the peer device and
exchanges configurations between switches in accordance with DCB.
4.17.3 FCoE
In a traditional data center, the Ethernet and storage area network (SAN) are independent of each
other, resulting in complicated network architecture and high network maintenance and
management costs. To simplify network architecture and reduce network costs, Fiber Channel
over Ethernet (FCoE) technology has become an inevitable industry trend as this technology
can effectively converge Ethernet and SAN.
In addition, converged network adapters (CNAs) are widely applied to 10GE servers. A CNA
can function as the network interface card (NIC) on the Ethernet and the host bus adapter (HBA)
on the fiber channel (FC) network at the same time. Therefore, a CNA can directly connect to
an Ethernet switch and transmit both Ethernet packets and FC packets (FC packets are
encapsulated in Ethernet packets using FCoE). Network convergence is thereby implemented.
FIP Snooping Bridge (FSB): supports access of FCoE traffic by working with DCB features.
FSB requires fewer FC access switches, which lowers the costs and power consumption on the
entire network.
4.17.4 VM Detection
In data center server virtualization scenarios, the minimum managed object for a switch is virtual
machine (VM), but not the physical server. When VMs migrate within a data center, the switch
needs to quickly detect the migration and synchronizes service configurations to new interfaces
to ensure normal running of services. Figure 4-4 shows the virtual awareness solution.
40
Figure 4-4 Virtual awareness solution
Network
administrator
Switch
NIC
Hardware
Virtual Switch
VM VM VM
NIC
Hardware
Virtual Switch
VM VM VM
nCenter
vCenter
The CE6800&5800 series switches support the following virtual awareness features:
l Manage and control VMs, and quickly obtain VM migration information from the vCenter.
l Automatically deploy and migrate VM policies. When a VM goes online, the switch
automatically deploys a VM policy for the VM; when a VM goes offline, its VM policy is
automatically deleted; when a VM is migrated, the switch deploys the VM policy on the
new access interface.
4.17.5 Forwarding Based on the VEPA
In server virtualization scenarios, virtual machines (VMs) on the same server cannot directly
communicate with each other.
In the Virtual Edge Port Aggregator (VEPA) service model, all traffic exchanged between VMs
on the same server must be forwarded by the upstream CE6800&5800 switches. This is the
VEPA-based traffic model. VEPA-based forwarding ensures that theCE6800&5800 series
switches can correctly forward traffic to the destination VMs on the same port. Figure 4-5 shows
the VEPA-based traffic model.
Figure 4-5 VEPA-based traffic model
Hypervisor
VM1 VM2 VM3
Server
VEPA
Switch
41
4.17.6 NLB Cluster Association
The network load balancing (NLB) cluster is a server clustering technology developed by
Microsoft based on network load balancing. NLB cluster supports load balancing and
redundancy backup among servers in a cluster. Servers in an NLB cluster use the same cluster
IP address and cluster MAC address. To ensure fast switching, network devices (such as
switches) in a data center need to send data packets to all servers in an NLB cluster. Therefore,
network devices need to be associated with the NLB cluster, as shown in Figure 4-6.
Figure 4-6 NLB cluster association
Switch
Server_1 Server_2 Server_3
IF1
IF2
IF3
Server group
IP MAC
NLB cluster association supports the following features:
l Associating NLB virtual IP addresses with multicast MAC addresses
l Associating multicast MAC addresses with multiple outbound interfaces
42
5 Operation Maintenance and Network
Management
About This Chapter
5.1 Maintenance and Management
5.2 eSight
Product Description 5 Operation Maintenance and Network Management
43
5.1 Maintenance and Management
5.1.1 Configuration Modes
Configuration Methods
CE6800&5800 switches support the following configuration and management methods:
l Command line
Users can log in to the console port on CE6800&5800 switches from a console terminal
and then configure various features and parameters in the command-line interface (CLI).
l NMS configuration
Users can configure and manage CE6800&5800 switches using SNMP through a network
management system (NMS) workstation.
Login Modes
CE6800&5800 switches provide a console port for users to configure CE6800&5800 switches
locally or remotely. Users can connect a console terminal to the console port through a serial
port.
In addition, users can log in to CE6800&5800 switches' service interfaces through Telnet,
Stelnet, or SSH for configuration and management.
For user logins, CE6800&5800 switches support multiple authentication modes, including non-
authentication, local authentication, and AAA authentication.
5.1.2 Monitoring and Maintenance
Hardware Monitoring
CE6800&5800 switches provide the following hardware monitoring functions:
l Re-detect hardware faults to prevent incorrect detection caused by intermittent interference.
l Checks version mapping automatically when the CE6800&5800 system is running.
Device Management and Maintenance
CE6800&5800 switches provide the following device management and maintenance functions:
l Support online help for the command line in English and Chinese.
l Provide hierarchical commands and user rights management.
l Provide an information center to uniformly manage logs, alarms, and debugging
information and redirect information as required.
44
l Provide electronic labeling. Users can query basic information about the main control
boards, optical modules, and fan modules on the CLI, and back up the information to an
external server using FTP or TFTP.
l Display the system version, module status, ambient temperature, CPU usage, and memory
usage.
5.1.3 Diagnosis and Debugging
Ping and tracert
On traditional IP networks, CE6800&5800 switches provide the following tools to check
network connectivity:
l Ping
l Tracert
These tools test network connectivity and record the transmission paths of packets to help locate
faults.
Debugging
CE6800&5800 switches provide various debugging commands for each software feature. Each
debugging command supports multiple parameters and can be flexibly controlled. Debugging
commands can display the detailed information about process handling, packets received and
transmitted, and error checking of features.
Black Box
CE6800&5800 switches provide a black box function to record information on feature modules,
tasks, and events. To facilitate fault location, the black box records the dying gasp, process status,
and function calling track.
VCT
After a user runs the virtual cable test (VCT) command on an interface of CE6800&5800
switches, the interface sends a testing signal. According to time domain reflectometry (TDR)
theory, the interface receives the reflected signal a while after sending the testing signal. Based
on the characteristics of the reflected signal, the user can infer the cable status.
Mirroring Functions
CE6800&5800 switches support the following mirroring functions.
l Port mirroring
CE6800&5800 switches copy packets from the mirrored port to the mirroring port for
analysis and monitoring.
l Flow mirroring
CE6800&5800 switches copy all traffic on the mirrored port to the mirroring port for
analysis and monitoring.
45
Users can connect a monitoring host to a mirrored port on CE6800&5800 switches to observe
packets passing through CE6800&5800 switches in real time. The mirroring function facilitates
traffic detection, fault location, and data analysis.
The CE6800&5800 series switches also support remote mirroring and can send the packets
copied by port mirroring or traffic mirroring to a remote monitoring device through a GRE
tunnel.
5.1.4 Software Upgrade and In-Service Patching
Software Upgrade
Before a software upgrade, CE6800&5800 switches can check the integrity and applicability of
system software. Two software upgrade methods are available for CE6800&5800 switches.
l Local upgrade
When CE6800&5800 switches are powered on and starting, the software can be loaded and
upgraded using the BIOS menu.
l Remote in-service upgrade
When CE6800&5800 switches are running properly, the new software can be remotely
downloaded using FTP or TFTP. When CE6800&5800 switches restart, the new software
runs and takes effect. This implements smooth remote software upgrade.
l Zero Touch Provisioning (ZTP)
After an unconfigured device is powered on, it downloads the new system software from
the remote file server to upgrade the system.
CE6800&5800 switches support software version rollback in case of an upgrade failure.
In-Service Patching
CE6800&5800 switches support in-service patching to protect services from being affected
when a patch is installed. Device information is recorded before and after in-service patching,
so the patch can be rolled back to the previous version if necessary.
5.1.5 Hardware Fault Handling
CE6800&5800 switches support automatic and manual intervention when a hardware fault
occurs, for example, when a chip fails. The maintenance personnel can locate a hardware fault
and handle it quickly to minimize service interruption.
5.2 eSight
eSight can display the software version information, save and restore configuration files and
VRP image programs, and support in-service patching for CE6800&5800 switches through CLI.
eSight provides the following functions:
46
Resource Management
eSight provides the resource management function to manage network resources, including
devices, interfaces, and links on large and complex networks. Resource management allows
users to query and manage CE6800&5800 switches and to check and locate abnormal resources.
View Management
eSight provides a unified topology view of all devices on a network to help users easily obtain
network information. eSight also provides powerful topology management functions. Users can
browse device information in the system topology views, protocol topology views, and user-
defined views. In addition, eSight provides user-friendly interfaces for operation and
maintenance of networks and devices.
The protocol topology views include the Ethernet view, which covers the topologies of various
networking modes and network hierarchies. These views support automatic discovery of a
network topology and reflect changes in the network topology and device status in real time.
Configuration Management
eSight provides configuration management and supports management of devices, interfaces,
VLANs, Layer 2 features, software upgrades, and configuration files. eSight supports diverse
configuration modes such as end-to-end configuration, batch configuration, and wizard-based
configuration, and offers default configuration templates.
Fault Management
Fault management is essential for maintaining networks. On the graphical user interface (GUI)
provided by eSight, users can query and monitor the CE6800&5800 running status and faults in
real time, and filter, locate, acknowledge, and analyze faults. eSight can generate alarm sounds
or display alarms on the alarm panel. eSight can be connected to an alarm box for convenient
routine maintenance of networks and devices.
Performance Management
eSight can monitor device performance, collect data, and analyze the collected data. It generates
device performance data in graphs and reports. With eSight, users can query the CPU capacity,
device memory, and interfaces, and collect statistics about device load and user access. With
these statistics, users can determine the QoS of the network and evaluate and adjust network
resource configurations proactively.
Performance management is implemented based on eSight's resource management function.
Security Management
eSight provides various measures for security management. Users are authenticated uniformly
on eSight and user rights are configured based on the minimum granularity. In addition, eSight
authenticates users to ensure system security and provides detailed operation logs for users to
query and analyze user operations.
Security management includes user management, access control, user group management, and
operation management.
47
6 System Parameters
About This Chapter
6.1 Specifications
Product Description 6 System Parameters
48
6.1 Specifications
Table 6-1 lists the specifications of CE6800&5800 switches.
Table 6-1 Specifications
Item Description
Physical specifications l Dimensions (W x D x H):
CE5810-24T4S-EI: 442.0 mm x 420.0 mm x 43.6
mm
CE5810-48T4S-EI: 442.0 mm x 420.0 mm x 43.6
mm
CE5850-48T4S2Q-EI: 442.0 mm x 420.0 mm x
43.6 mm
CE5850-48T4S2Q-HI: 442.0 mm x 420.0 mm x
43.6 mm
CE6850-48S4Q-EI: 442.0 mm x 600.0 mm x
43.6 mm
CE6850-48T4Q-EI: 442.0 mm x 600.0 mm x
43.6 mm
l Weight (without optical modules, with two power
modules and two fan modules):
CE5810-24T4S-EI: 8.0 kg
CE5810-48T4S-EI: 8.2 kg
CE5850-48T4S2Q-EI: 8.85 kg
CE5850-48T4S2Q-HI: 8.8 kg
CE6850-48S4Q-EI: 11.05 kg
CE6850-48T4Q-EI: 11.35 kg
Environment
parameters
Temperature l Operating temperature: 0C to 40C (0 m to 1800 m)
NOTE
When the altitude is between 1800 m and 5000 m, the
highest operating temperature reduces 1C every time the
altitude increases 220 m.
l Storage temperature: -40C to +70C
Relative
humidity
5% RH to 95% RH, noncondensing
Altitude < 5000 m
49
Item Description
Noise (sound
pressure, 27C)
l CE5810-24T4S-EI:
Back-to-front airflow: < 43 dBA
Front-to-back airflow: < 47 dBA
l CE5810-48T4S-EI:
l CE5850-48T4S2Q-EI:
l CE5850-48T4S2Q-HI:
l CE6850-48S4Q-EI:
l CE6850-48T4Q-EI:
Power
specifications
Power source
type
AC/DC
AC power input l Rated input voltage range: 100 V AC to 240 V AC,
50/60 Hz
l Maximum input voltage range: 90 V AC to 290 V
AC, 45 Hz to 65 Hz
DC power input l Rated voltage range: -48 V DC to -60 V DC
l Maximum voltage range: -38.4 V DC to -72 V DC
Maximum input
current
l 150 W AC power: 2.5 A (100 V AC to 240 V AC)
l 350 W AC power: 5 A (100 V AC to 240 V AC)
l 350 W DC power: 11 A (-38.4 V DC to -72 V DC)
l 600 W AC power: 9 A (100 V AC to 240 V AC)
Chassis power
consumption
Maximum
power
consumption
l CE5810-24T4S-EI: 68 W
l CE5810-48T4S-EI: 92 W
l CE5850-48T4S2Q-EI: 133 W
l CE5850-48T4S2Q-HI: 131 W
l CE6850-48S4Q-EI: 272 W
l CE6850-48T4Q-EI: 380 W
50
Item Description
Typical power
consumption
l CE5810-24T4S-EI: 58 W (100% traffic load, 3 m
network cables on 24 ports, SFP+ cables on 4 ports,
double power modules)
l CE5810-48T4S-EI: 80 W (100% traffic load, 3 m
network cables on 48 ports, SFP+ cables on 4 ports,
l CE5850-48T4S2Q-EI: 103 W (100% traffic load, 3
m network cables on 48 ports, SFP+ cables on 4 ports
and QSFP+ cables on 2 ports, double power
modules)
l CE5850-48T4S2Q-HI: 109 W (100% traffic load, 3
m network cables on 48 ports, SFP+ cables on 4 ports
and QSFP+ cables on 2 ports, double power
modules)
l CE6850-48S4Q-EI: 180 W (100% traffic load, SFP
+ cables on 48 ports and QSFP+ cables on 4 ports,
l CE6850-48T4Q-EI: 305 W (100% traffic load, 3 m
network cables on 48 ports and QSFP+ cables on 4
ports, double power modules)
Surge protection Port: 1kV in common mode; power module: 6 kV in
common mode and 6 kV in differential mode
Heat
dissipation
Heat dissipation
mode
Air cooling
Airflow Front-to-back or back-to-front, which is determined by
features of fan modules and power modules.
Reliability Power module
backup
1+1 backup
Fan module
backup
The CE5850-EI and CE5810-EI support 1+1 backup of
fan modules
NOTE
A CE6800 or CE5850-HI chassis uses two FAN-40EA series
fan modules, with each fan module containing two fans. The
four fans on the CE6800 or CE5850-HI chassis work in 3+1
backup mode.
Hot swap All the power modules and fan modules support hot
swap.
51
Item Description
Technical
specifications
Processor l CE5810-24T4S-EI: 1.2 GHz, dual-core
l CE5810-48T4S-EI: 1.2 GHz, dual-core
l CE5850-48T4S2Q-EI: 1.2 GHz, quad-core
l CE5850-48T4S2Q-HI: 1.2 GHz, dual-core
l CE6850-48S4Q-EI: 1.5 GHz, quad-core
l CE6850-48T4Q-EI: 1.5 GHz, quad-core
Forwarding
capacity of the
chassis (Mpps)
l CE5810-24T4S-EI: 96
l CE5810-48T4S-EI: 132
l CE5850-48T4S2Q-EI: 252
l CE5850-48T4S2Q-HI: 252
l CE6850-48S4Q-EI: 960
l CE6850-48T4Q-EI: 960
SDRAM
Memory
2 GB
NOR Flash l CE5810-24T4S-EI: 16 MB
l CE5810-48T4S-EI: 16 MB
l CE5850-48T4S2Q-EI: 8 MB
l CE5850-48T4S2Q-HI: 16 MB
l CE6850-48S4Q-EI: 8 MB
l CE6850-48T4Q-EI: 8 MB
NAND Flash l CE5810-24T4S-EI: 512 MB
l CE5810-48T4S-EI: 512 MB
l CE5850-48T4S2Q-EI: 1 GB
l CE5850-48T4S2Q-HI: 1 GB
l CE6850-48S4Q-EI: 1 GB
l CE6850-48T4Q-EI: 1 GB
52
Item Description
Stack Service port
supporting the
stack function
l CE5810-24T4S-EI: uplink 10GE optical ports,
supporting service interface stacking
l CE5810-48T4S-EI: uplink 10GE optical ports,
supporting service interface stacking
l CE5850-48T4S2Q-EI: uplink 10GE and 40GE
optical ports
l CE5850-48T4S2Q-HI: uplink 10GE and 40GE
optical ports
l CE6850-48S4Q-EI: downlink 10GE optical ports
and uplink 40GE optical ports
l CE6850-48T4Q-EI: downlink 10GE electrical ports
(V100R002 and later versions) and uplink 40GE
optical ports
NOTE
Downlink GE electrical ports cannot be used as stack ports.
Maximum stack
bandwidth
(unidirectional)
l CE5810-24T4S-EI: 40 Gbit/s (4x10GE, a maximum
of four physical ports on a logical stack port)
l CE5810-48T4S-EI: 40 Gbit/s (4x10GE, a maximum
of two physical ports on a logical stack port)
l CE5850-48T4S2Q-EI: 80 Gbit/s (4x10GE or
2x40GE, a maximum of four physical ports on a
logical stack port)
l CE5850-48T4S2Q-HI: 80 Gbit/s (4x10GE or
2x40GE, a maximum of four physical ports on a
logical stack port)
l CE6850-48S4Q-EI: 160 Gbit/s (4x40GE or
16x10GE, a maximum of sixteen physical ports on
a logical stack port)
l CE6850-48T4Q-EI: 160 Gbit/s (4x40GE or
16x10GE, a maximum of sixteen physical ports on
a logical stack port)
NOTE
A 10GE port and a 40GE port cannot belong to the same logical
stack port.
Safety standards compliance l EN 60950-1:2006+A11:2009+A1:2010+A12:2011
l EN 60825-1:2007
l EN 60825-2:2010
l UL 60950-1:2007 2rd Edition
l CSA C22.2 No.650:2007 2rd Edition
l IEC 60950-1:2005+A1:2009
l AS/NZS 60950-1:2011
l GB4943:2011
53
Item Description
EMC standards compliance l ICES-003:2012 CLASS A
l CISPR 22:2008 CLASS A
l CISPR 24:2010
l EN 55022:2010 CLASS A
l EN 55024:2010
l ETSI EN 300 386 V1.6.1:2012
l AS/NZS CISPR 22:2009 CLASS A
l IEC 61000-3-2:2005+A1:2008+A2:2009/EN
61000-3-2:2006+A1:2009+A2:2009
l IEC 61000-3-3:2008/EN 61000-3-3:2008
l CNS 13438:2006 CLASS A
l VCCI V-4:2012 CLASS A
l VCCI V-3:2012 CLASS A
l FCC 47CFR Part15 CLASS A
l EC Council Directive 2004/108/EC
l GB9254
Safety and environmental
standards compliance
l 2002/95/EC, 2011/65/EU
l 2002/96/EC, 2012/19/EU
l EC NO.1907/2006
l ETSI EN 300 019-1-1 V2.1.4
l ETSI EN 300 019-1-2 V2.1.4
l ETSI EN 300 019-1-3 V2.3.2
l ETSI EN 300753 V1.2.1
54

CloudEngine 6800&5800 V100R003C00 Product Description 01

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

CloudEngine 6800&5800 V100R003C00 Product Description 01

Hochgeladen von

Copyright:

Verfügbare Formate

CloudEngine 6800&5800 Series Switches

Das könnte Ihnen auch gefallen

CloudEngine 6800&amp;5800 V100R003C00 Product Description 01

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

CloudEngine 6800&amp;5800 V100R003C00 Product Description 01

Hochgeladen von

Copyright:

Verfügbare Formate

CloudEngine 6800&5800 Series Switches

Das könnte Ihnen auch gefallen

CloudEngine 6800&5800 V100R003C00 Product Description 01

CloudEngine 6800&5800 V100R003C00 Product Description 01