TPNA SWITCH ADVANCED FUNCTION

Quality of service ...................................................................................................................................... 2

Q1: Do you need to configure QoS in the regular network? ............................................................ 2

Q2: How QOS function of TP-Link switch works? ............................................................................. 2

Voice VLAN ............................................................................................................................................... 3

Q3: Do you still need to configure Voice VLAN? .............................................................................. 3

Q4: What is the main difference between Voice VLAN and Auto VOIP? ......................................... 4

VRRP ......................................................................................................................................................... 4

Q5: How to provide redundancy for your default gateway? ............................................................ 4

L2PT .......................................................................................................................................................... 5

Q6: How to transfer STP packets through the ISP network? ............................................................ 5

RIP&OSPF.................................................................................................................................................. 6

Q7: Do you need the dynamic routing protocol and how to choose them? .................................... 6

Q8: What is the routing metrics? ..................................................................................................... 7

OAM ......................................................................................................................................................... 8

Q9: What is OAM? ............................................................................................................................ 8

QUALITY OF SERVICE

Q1: DO YOU NEED TO CONFI GURE QOS IN THE REGULAR NETWORK?

A: If there is not VoIP application or other time-sensitive applications in your network, the QoS setting
is not very useful for you. But if you have VoIP Phone in your network, I suggest you configure Voice
VLAN or QoS function.

Our QoS consists of the following components, Classification, Marking and Queuing. We can assign
different priorities to different packets, and then different priorities map to the separate output
queue, the different queues have different priorities also, according to our setting, When there is
congestion in the network, the switch will forward high priority packet firstly. This is the primary
function of the QoS function.

So if we want to guarantee the network quality of the time-sensitive application, we need to assign
the high priority to the packet of this application and configure QoS in the switch.

Q2: HOW QOS FUNCTION OF TP-LINK SWITCH WORKS?

A: TP-Link Switch supports three different priority modes in QoS function, Port priority,802.1P priority
and DSCP priority, but a port can use only one priority to classify the ingress packets. When you
configure the port priority,802.1P priority and DSCP priority at the same time, the priority order is
802.1P priority > DSCP priority > Port priority.

In Port Priority mode, the switch prioritizes packets according to their ingress ports, if you assign a
high priority to one ingress ports, every packet from this ingress port has a high priority.

In 802.1P priority mode, the switch only prioritizes packets with VLAN tag, regardless of other fields of
packets. IEEE 802.1p is one protocol, it uses three bits in 802.1Q Tag as PRI field, so there are from 0
to 7 eight different priorities.

In DSCP priority mode, the switch prioritizes packets with ToS(Type of Service)field in the IP header of
Packets. The first six bits of TOS field is used to represent DSCP priority, so the DSCP value are from 0
to 63.

The different packets have different priority, TP-Link switch can remap different priorities for packets
also. That is mean, whatever the packet support 802.1p and DSCP or not, TP-Link switch can assign or
change the priority for every packet.

TP-Link Managed Switch supports eight output queues, and we can map different priorities to
different queues. We can configure the different weight for queues also.
When we assigned the priority and output queues for every packet, the switch can decide what's
packets can be forwarded firstly in network congestion.

VOICE VLAN

Q3: DO YOU STILL NEED TO CONFIGURE VOICE VLAN?

A: We have known the QoS function can assign the different priorities to the different packet and
transfer the packet from different queues. We can set the voice data has a high priority, and the
normal data has a low priority. If you only want to guarantee the voice traffic stably, maybe the voice
VLAN is your right choice also.

Voice VLAN can detect your VOIP Phone automatically, and add these Phones to specific VLAN,
configure the suitable priority to the voice packet also. That is mean when you set Voice VLAN. The
switch can recognize your VOIP Phone, isolate voice traffic with other data and assign a high priority
for these voice packets. The Voice VLAN combined the advantage of 802.1Q VLAN and QoS function.
Q4: WHAT IS THE MAIN DIFFERENCE BETWEEN VOICE VLAN AND AUTO VOIP ?

A: With the development of VOIP network, the Voice VLAN function cannot meet some demand. So
TP-Link design the new feature to fit the new scenario. This new function is Auto VOIP, and its effect
is the same as Voice VLAN. But they are fit for different situations. Generally, there are two different
application scenarios for the VOIP network.

Topology A: The voice traffic and data traffic are passing through the different ports of Switch A. In
this topology, the IP Phone does typically not support tagged frame.

Topology B: The voice traffic and data traffic are passing through the same port of Switch A. In this
topology, the IP Phone needs to support LLDP topology and deal with tagged packets.

If your network topology is like topology A, we suggest you configure Voice VLAN. If your network
topology is like topology B, we recommend you configure Auto VOIP.

For Voice VLAN, the switch recognizes IP Phone via the OUI information, But Auto VOIP uses the LLDP-
MED protocol to interact information.

In a word, if your IP Phone support LLDP-MED protocol and send the tagged voice data, Auto VOIP is
your best choice. If your IP Phone is old, do not support tagged frame and LLDP-MED protocol, I
suggest you use Voice VLAN.

VRRP

Q5: HOW TO PROVIDE REDUN DANCY FOR YOUR DEFAULT GATEWAY?

A: Generally, when we want to access the internet, we will transfer the data to our default gateway
firstly. If the default gateway is broken, all hosts in the LAN will not access the internet. To prevent
meet this situation, we need to provide redundancy for the default gateway.

VRRP is the abbreviation of Virtual Router Redundancy Protocol. We need to develop two or more
routers in the network firstly, then build one virtual router via VRRP. For the hosts in the LAN
network, there is only one default gateway, the virtual router. But in fact, this virtual router is consists
of the different routers or layer three switches.

The virtual router has its IP address, and this IP address is the default gateway of hosts in the LAN. If
there is something wrong with the master router, the backup router will instead of the master router,
but the hosts in the LAN do not know this process. Because these hosts only care about the IP address
of the virtual router.

So we can use multiple physical routers or Layer 3 switches to build one virtual router via VRRP to
provide redundancy for the default gateway.

L2PT

Q6: HOW TO TRANSFER STP PACKETS THROUGH THE ISP NETWORK?

A: Layer 2 protocol cannot transfer through the Layer 3 network. That is means Layer 2 protocol
cannot communicate with each other through the different subnets. The STP(Spanning Tree Protocol),
GVRP(GARP VLAN Registration Protocol), LACP(Link Aggregation Control Protocol) all belongs to the
Layer 2 protocol. Now, the Customer has two local networks that are connected through the ISP
network. When these two customer networks run the STP, how to transfer the BPDU (the Spanning
Tree Protocol negotiation packet) through the ISP network?
We need to configure L2PT (Layer 2 Protocol Tunneling) on the Edge switch (the switch that is
connected to the customer network and placed on the boundary of the ISP network). The L2PT
function will replace the destination MAC address of the original Layer 2 protocol to
01:00:0C:CD:CD:D0.

SW1 and SW2 have configured the STP function, and they need to interact BPDU with each other
through the ISP network. When the Edge switch received the BPDU, the Edge switch will replace the
destination MAC address of BPDU with 01:00:0C:CD:CD:D0 and then sends the PDU to the ISP
network. The ISP network identifies the PDU and directly transfer it to the other end. When Edge
Switch receives the PDU via NNI (network network interface ), it will restore the destination MAC
address to the original destination MAD address of STP.

RIP&OSPF

Q7: DO YOU NEED THE DYNAMIC ROUTING PROTOCOL AND HOW TO CHOOSE THEM?

A: We have introduced the difference between static routing and dynamic routing in the Basic Q&A.
Unlike static routing that requires manual configuration to tell the router where the destination
networks are, dynamic routing protocol leans about destination network from neighboring routers.

If there are fewer network segments for your network, and you need not always change the IP
network segment, the static routing is fit for you. But if you are designing the network for one large-
lot producer, you need to choose the dynamic routing protocol, the dynamic routing can realize that
create the routing table automatically, and redundancy IP network.

Dynamic routing protocols fall under one of three categories: distance-vector, link-state, and hybrid.
Every type of dynamic routing protocol has the advantage and disadvantages.

The RIP(Routing information protocol) V1 is a distance-vector protocol. As most of the distance-vector

protocol, RIP v1 uses local broadcasts to share routing information periodically. But the RIP V2 is a
hybrid protocol, and it is a classless protocol and supports authentication. RIP V2 uses multicast to
update routing.
The OSPF (open shortest path fast) is a link-state protocol. It uses the SFP algorithm and provides a
loop-free topology. The OSPF overcome many of the deficiencies problems that RIP had in the large
enterprise networks, now it is the most popular dynamic routing protocol in many corporates.

If you want to get more information about RIP and OSPF, you can study the course of TPNP.

Q8: WHAT IS THE ROUTING METRICS?

A: When we build one large enterprise network, typically, there is more than one path to the
destination network (we need to build the redundancy network). The router in the network uses the
same dynamic routing protocol. In this situation, a routing protocol will use a measurement called
"Metric" to determine which path is the best path.

The RIP uses "hop count" as a metric. We can see the topology as below: when PC A wants to
communicate with PC B, there are two paths we can choose. The first path is RTA->RTB, but the
bandwidth of this path is 55kbps, the second path is RTA->RTC->RTB, the bandwidth of path is 2Mbps,
but the hop count is 2. When we use RIP in this network, the data will be transfer from RTA to RTB,
because of the hop count of this path is 1.

Unlike RIP, the OSPF uses "cost" as the metric. Cost is not bandwidth, and it is the inverse of the
bandwidth of a link. The faster the speed of the connection, the lower the cost. The most preferred
path is the one with the lowest cost. We still use one topology to take the example:
There are still having two paths when PC A wants to communicate with PC B. The first path is
RTA->RTB, one hop, but the cost is 100. The second path is RTA->RTC->RTB, two hops, the cost is 29.
When we use OSPF, the best path is the second path, because of the cost of the second path is lower
than the first one.

OAM

Q9: WHAT IS OAM?

A: When Ethernet is designed, it does not include maintenance and administration features. Now, the
Ethernet network becomes the most popular network in all over the world. Some ISP network uses
the Ethernet network instead of the ATM network and so on. Just under such a background, to
increase maintenance and administration features for Ethernet becomes one important thing.

The OAM is the abbreviation of operation, administration, and maintenance. It is the Layer 2 protocol,
and we can use OAM to monitor and troubleshoot Ethernet network. OAM can monitor link
performance, link fault, and send alarms to the network administrator. This is mean the OAM can
help us to maintain and troubleshoot Ethernet network. There are two standards about the OAM.
One is the IEEE 802.3ah, and the other is the IEEE 802.1ag. TP-Link Switch only supports IEEE 802.3ah,
EFM OAM.

TP-Link switch supports Link monitoring, Remote Failure Indication, and Remote Loopback three OMA
features. The administrator can monitor link performance via Link monitoring. When there is
something wrong with the link, the switch can send notification also. If there is disruption in traffic
because of the device failure, the Remote failure indication can help the administrator to find and
handle the issue quickly. With Remote Loopback, administrators can test link performance, including
the delay, jitter, and frame loss rate during installation or troubleshooting.