Cisco Catalyst Jumbo MTU Support and config

I am setting up a 3750 switch and want to enable jumbo frames by setting the system mtu to 9000 bytes. Since this isn't the default, I am wondering what the disadvantages are to doing this. Anyone have any thoughts on this?

Never needed to do such a thing on a Cisco switch - I'd be interested in knowing why you need such a large MTU; iScsi ?

You might encounter bottlenecks in your network where there is severe fragmentation of the packets. Obviously where the MTU falls back on the regular size i.e. 1500, that's where you will experience degradation in network speed. The translation speed of your network at nodes where fragmentation is taking place, will be very slow.

Arani, There is plenty of logic to what your saying. If this though is used for a special application/service (not talking about Windows server or workstation) which is able to send such large packets and the switch is configured to work with them, isn't it true that they won't suffer from any fragmentation?


Here is a portion of the information I discovered while researching on the topic:

Catalyst 3750/3560 Series

Catalyst 3750/3560 Series switches support an MTU of 1998 bytes for all 10/100 interfaces. All Gigabit Ethernet interfaces support jumbo frames up to 9000 bytes. The default MTU and jumbo frame size is 1500 bytes. You cannot change the MTU on an individual interface. You must set the MTU globally. Reset the switch afterwards for the MTU change to take effect.
Configure

Use the system mtu command to change the MTU for all 10/100 interfaces. This command only effects 10/100 interfaces.

Use the system mtu jumbo command to change the MTU for all Gigabit Ethernet interfaces. This command only effects Gigabit Ethernet Interfaces.

Note: Gigabit Ethernet ports are not affected by the system mtu command; 10/100 ports are not affected by the system mtu jumbo command. If you do not configure the system mtu jumbo command, the setting of the system mtu command applies to all Gigabit Ethernet interfaces.

p.s I'm moving this thread to the correct category.

Hi Chris,

I was working under the assumption that traffic between host and target was going through several nodes, some of which may or may not support variable MTU size. Let's hope traffic in this case is flowing through nodes and all of them support variable MTU size. Then perhaps for a stream line flow, it could be set to 1900 or any other consistant value, minimising, if not removing the issues of fragmentation. That in turn would obviously reduce transmission delays and translation issue within any specific intermediate node.