Question

How to reduce/load balance Bandwidth utilization on a Single Link between two Core Switch.


Userlevel 3
Hi, Please advice on How to reduce/load balance Bandwidth utilization on a Single Link between two Core Switch?

13 replies

Hello Yan Linn Aung,

1 If you are running VRRP, you can load share the master and backup between the cores.

For example

For odd vlan tag you can have core1 as master and for even vlan tag you can have core2 as master

2 You can group the ports between the core
Userlevel 3
Ravisekar, Rajamanickam wrote:

Hello Yan Linn Aung,

1 If you are running VRRP, you can load share the master and backup between the cores.

For example

For odd vlan tag you can have core1 as master and for even vlan tag you can have core2 as master

2 You can group the ports between the core

Thanks Ravisekar,

Please kindly provide the sample configuration commands. Thanks!
Userlevel 3
Ravisekar, Rajamanickam wrote:

Hello Yan Linn Aung,

1 If you are running VRRP, you can load share the master and backup between the cores.

For example

For odd vlan tag you can have core1 as master and for even vlan tag you can have core2 as master

2 You can group the ports between the core

Thanks Ravisekar,

Please kindly provide the sample configuration commands. Thanks!
Userlevel 3
Ravisekar, Rajamanickam wrote:

Hello Yan Linn Aung,

1 If you are running VRRP, you can load share the master and backup between the cores.

For example

For odd vlan tag you can have core1 as master and for even vlan tag you can have core2 as master

2 You can group the ports between the core

Hi Ravisekar,

Please kindly provide the sample configuration commands. Thanks!
Ravisekar, Rajamanickam wrote:

Hello Yan Linn Aung,

1 If you are running VRRP, you can load share the master and backup between the cores.

For example

For odd vlan tag you can have core1 as master and for even vlan tag you can have core2 as master

2 You can group the ports between the core

Hello Yan Linn,

Sample config for vrrp. In the configuration I am creating two vlans , vlan9 and vlan10 and then going to run VRRP between them

Core1
create vlan vlan9 tag 9
conf vlan vlan9 ipaddress 192.168.9.11/24
conf vlan vlan9 add ports XXX
create vlan vlan10 tag 10
conf vlan vlan10 ipaddress 192.168.10.11/24
conf vlan vlan10 add ports XXX
create vrrp vlan vlan10 vrid 1
create vrrp vlan vlan9 vrid 1
configure vrrp vlan vlan10 vrid 1 add 192.168.10.10
configure vrrp vlan vlan9 vrid 1 add 192.168.9.10
enable vrrp vlan vlan10 vrid 1
enable vrrp vlan vlan9 vrid 1

Core 2
create vlan vlan9 tag 9
conf vlan vlan9 ipaddress 192.168.9.10/24
conf vlan vlan9 add ports XXX
create vlan vlan10 tag 10
conf vlan vlan10 ipaddress 192.168.10.10/24
conf vlan vlan10 add ports XXX
create vrrp vlan vlan10 vrid 1
configure vrrp vlan vlan10 vrid 1 priority 255
create vrrp vlan vlan9 vrid 1
configure vrrp vlan vlan9 vrid 1 priority 255
configure vrrp vlan vlan10 vrid 1 add 192.168.10.10
configure vrrp vlan vlan9 vrid 1 add 192.168.9.10
enable vrrp vlan vlan10 vrid 1
enable vrrp vlan vlan9 vrid 1

From this configuration core 2 act as default gateway for both vlans since priority in core 2 is set higher than core 1. If you want to load balance for vlan 9 you can configure higher priority in core1 and leave the priority to the default value(100). Below is the output from show commands

core1
sh vrrp
Virtual Master
VLAN Name VRID Pri IP Address State MAC Address TP/TR/TV/P/T
vlan10(En) 0001 100 192.168.10.10 BKUP 00:00:5e:00:01:01 0 0 0 Y 1
vlan9(En) 0001 100 192.168.9.10 BKUP 00:00:5e:00:01:01 0 0 0 Y 1

En-Enabled, Ds-Disabled, Pri-Priority, T-Advert Timer, P-Preempt
TP-Tracked Pings, TR-Tracked Routes, TV-Tracked VLANs

Core2

sh vrrp
Virtual Master
VLAN Name VRID Pri IP Address State MAC Address TP/TR/TV/P/T
vlan10(En) 0001 255 192.168.10.10 MSTR 00:00:5e:00:01:01 0 0 0 Y 1
vlan9(En) 0001 255 192.168.9.10 MSTR 00:00:5e:00:01:01 0 0 0 Y 1
Ravisekar, Rajamanickam wrote:

Hello Yan Linn Aung,

1 If you are running VRRP, you can load share the master and backup between the cores.

For example

For odd vlan tag you can have core1 as master and for even vlan tag you can have core2 as master

2 You can group the ports between the core

Static grouping configuration is simple.. If you have 4 ports from core1 are connected to core2. Lets take example as ports 40 to 44. (40 is the master port of the group that means after enabling sharing , any changes on port 40 will change the config on that group ports41-44)

enable sharing 40 grouping 40-44

Additionally if you want to set algorithm for loadsharing you can add few more in the above command

enable sharing 40 grouping 40-44 algorithm address-based/port based

These all for static sharing , if you want to enable LACP mode you can do as below

enable sharing 40 grouping 40-44 LACP
Userlevel 3
Ravisekar, Rajamanickam wrote:

Hello Yan Linn Aung,

1 If you are running VRRP, you can load share the master and backup between the cores.

For example

For odd vlan tag you can have core1 as master and for even vlan tag you can have core2 as master

2 You can group the ports between the core

Thanks Ravisekar,

Even after i enable Lacp .. port is not sharing. Utilization is high on Link 1 only. I want to use load balance on both links. Please kindly see the below output and advice thanks.

* I try L2 static, L3_L4 and LACP. all the same ports are not load balance. *

enable sharing 1 grouping 1,4

enable sharing 1 grouping 1,4 L3_L4

enable sharing 1 grouping 1,4 algorithm address-based L3_L4 lacp


CoreSW # sh port 1,4 utilization bandwithPort Link Link Rx Peak Rx Tx Peak Tx
State Speed % bandwidth % bandwidth % bandwidth % bandwidth
================================================================================
Link_1> A 1000 82.06 94.94 48.23 55.72
Link_2> A 1000 0.25 0.30 1.34 1.57
================================================================================
> indicates Port Display Name truncated past 8 characters
Link State: A-Active, R-Ready, NP-Port Not Present, L-Loopback

CoreSW #sh sharing
Load Sharing Monitor
Config Current Agg Ld Share Ld Share Agg Link Link Up
Master Master Control Algorithm Group Mbr State Transitions
==============================================================================
1 1 LACP L3_L4 1 Y A 0
L3_L4 4 Y A 0
==============================================================================
Link State: A-Active, D-Disabled, R-Ready, NP-Port not present, L-Loopback
Load Sharing Algorithm: (L2) Layer 2 address based
(L3_L4) Layer 3 address and Layer 4 port based
Number of load sharing trunks: 1

Please kindly advice. Thanks.
Userlevel 4
write an acl to drop unnecessary multicast trafic sent from host.
Are you using Mlag,STP ,RSTP,ESRP?
Any of the above protocols there are better options.
is this a triangular toplogy.
Userlevel 3
PARTHIBAN CHINNAYA wrote:

write an acl to drop unnecessary multicast trafic sent from host.
Are you using Mlag,STP ,RSTP,ESRP?
Any of the above protocols there are better options.
is this a triangular toplogy.

What will be the impact if we use the existing link on the access switch and link up it in between the 2 Core switches?

How can we reduce network bandwidth usage between Core to Core Link and Access to Core links?

Bandwidth is being saturated with many multicast traffic.

Please see the attached Sample Network Diagram. Thanks.

Userlevel 3
Thanks Ravisekar and PARTHIBAN,

I am newbie in Extreme Network, I will study and update the current setup of this network. Client is using Extreme for their CCTV Multicast Network. Thank for your assistance.

BRgds,
Yan Linn
Userlevel 4
In show sharing you must see all the ports in lag.
could you check the configuration.
on both the switches.
enable sharing 1 grouping 1,2 algorithm address-based l2
Userlevel 3
PARTHIBAN CHINNAYA wrote:

In show sharing you must see all the ports in lag.
could you check the configuration.
on both the switches.
enable sharing 1 grouping 1,2 algorithm address-based l2

Even after i enable Lacp .. port is not sharing. Utilization is high on Link 1 only. I want to use load balance on both links. Please kindly see the below output and advice thanks.

* I try L2 static, L3_L4 and LACP. all the same ports are not load balance. *

enable sharing 1 grouping 1,4

enable sharing 1 grouping 1,4 L3_L4

enable sharing 1 grouping 1,4 algorithm address-based L3_L4 lacp


CoreSW # sh port 1,4 utilization bandwithPort Link Link Rx Peak Rx Tx Peak Tx
State Speed % bandwidth % bandwidth % bandwidth % bandwidth
================================================================================
Link_1> A 1000 82.06 94.94 48.23 55.72
Link_2> A 1000 0.25 0.30 1.34 1.57
================================================================================
> indicates Port Display Name truncated past 8 characters
Link State: A-Active, R-Ready, NP-Port Not Present, L-Loopback

CoreSW #sh sharing
Load Sharing Monitor
Config Current Agg Ld Share Ld Share Agg Link Link Up
Master Master Control Algorithm Group Mbr State Transitions
==============================================================================
1 1 LACP L3_L4 1 Y A 0
L3_L4 4 Y A 0
==============================================================================
Link State: A-Active, D-Disabled, R-Ready, NP-Port not present, L-Loopback
Load Sharing Algorithm: (L2) Layer 2 address based
(L3_L4) Layer 3 address and Layer 4 port based
Number of load sharing trunks: 1

Please kindly advice. Thanks.
Userlevel 3
Even after i enable Lacp .. port is not sharing. Utilization is high on Link 1 only. I want to use load balance on both links. Please kindly see the below output and advice thanks.

* I try L2 static, L3_L4 and LACP. all the same ports are not load balance. *

enable sharing 1 grouping 1,4

enable sharing 1 grouping 1,4 L3_L4

enable sharing 1 grouping 1,4 algorithm address-based L3_L4 lacp


CoreSW # sh port 1,4 utilization bandwithPort Link Link Rx Peak Rx Tx Peak Tx
State Speed % bandwidth % bandwidth % bandwidth % bandwidth
================================================================================
Link_1> A 1000 82.06 94.94 48.23 55.72
Link_2> A 1000 0.25 0.30 1.34 1.57
================================================================================
> indicates Port Display Name truncated past 8 characters
Link State: A-Active, R-Ready, NP-Port Not Present, L-Loopback

CoreSW #sh sharing
Load Sharing Monitor
Config Current Agg Ld Share Ld Share Agg Link Link Up
Master Master Control Algorithm Group Mbr State Transitions
==============================================================================
1 1 LACP L3_L4 1 Y A 0
L3_L4 4 Y A 0
==============================================================================
Link State: A-Active, D-Disabled, R-Ready, NP-Port not present, L-Loopback
Load Sharing Algorithm: (L2) Layer 2 address based
(L3_L4) Layer 3 address and Layer 4 port based
Number of load sharing trunks: 1

Please kindly advice. Thanks.

Reply