All About Jumbo Frames

Many networks use 1500-byte MTU size, but the MTU size can be reduced by encapsulation, tunneling or other overlay network protocols. These situations reduce the end-to-end effective MTU size which reduces throughput and network efficiency and sometimes causes application problems. Many network devices now support larger sizes of Ethernet frames and the use of Jumbo Frames is becoming more common. This article covers how to determine if your network is capable of using Jumbo Frames and if you should enable this feature.

 In order to understand the jumbo frame we need to understand the following terminology:

MTU - Maximum transmission unit - this is the largest physical packet size measured in bytes that a network can transmit. Any packet larger than this MTU is divided or fragmented into smaller packets before transmission. The standard MTU on an Ethernet network is 1500 bytes plus the size of the L2 header and frame check sequence which is an additional 18 bytes. So standard MTU size for Ethernet is 1518.

Jumbo frame - Jumbo Frames are network-layer PDUs that have a size much larger than the typical 1500 byte Ethernet MTU size. These jumbo frames are sometimes also called "Giants". Also, the term "Baby Giants" refers to frames that are just slightly larger than the 1500 byte Ethernet MTU size (e.g. 1998 bytes). In some situations, jumbo frames can be used to allow for much larger frame sizes if the networking hardware is capable of this configuration. Most modern routers and switches are capable of jumbo frames and much of the networking hardware within the data centers is capable of this.

A frame that is larger than the standard MTU size of 1518 bytes. This definition is vendor dependant and is not part of any IEEE standard. For Cisco, 1518 or 1522 in the case of Q in Q is the standard MTU. Anything larger is considered a jumbo

frame.

When encapsulation from Layer 2 to Layer 3 occurs, if you have a frame that says, 1600 bytes in size, in order for it to be encapsulated to layer 3, if there is an MTU setting of 1518, this frame will be fragmented into two parts, and each part will be placed in a separate packet. So one frame can be fragmented into two (or more) packets when encapsulated.

A network can be configured to accommodate jumbo frames, that is on layer 2, frames of sizes up to 9216 bytes for Gigabit Ethernet and faster interfaces can be used. 10/100Mbps ports are limited to 8092 bytes. MTU is configured on interfaces, such as physical interfaces of a switch or router, or on SVIs. Some platforms have a global MTU size configuration parameter example for Cisco 9K, 3850 switches we can not configure the jumbo frame per-interface basis you have to configure that globally. On other platforms i.e 4500, 6500, 6800, etc, you can enable it per port. If all interfaces on all devices within a flat layer 2 networks are configured to accommodate jumbo frames, this could be very useful especially when configuring the core or backbone of a network. Large bulks of data can be sent in this manner with much less overhead.

When using jumbo frames, it is important to be sure that the design you create is sound. If there is routing involved, that is if these large frames are going up to layer 3, then the fragmentation of these frames may occur and this can easily slow down a network or even drop frames that cannot be fragmented.

Configuration of Jumbo frames on different Cisco switch model:

On 9300

Switch(config)#system mtu ?

 <1500-9198> MTU size in bytes

 Switch(config)#system mtu 9198

Global Ethernet MTU is set to 9198 bytes.

Note: this is the Ethernet payload size, not the total

Ethernet frame size, which includes the Ethernet

header/trailer and possibly other tags, such as ISL or

802.1q tags.

Switch#show system mtu

Global Ethernet MTU is 9198 bytes.

Now the valid question would be how does it affect the interfaces that don’t support jumbo, since we have enabled it in a global mode?

The system MTU can be seen like a tunnel diameter, if you increase the MTU, the switch will be able to receive and transmit frames from end devices with MTU above the default value, and it still will be able to forward/receive normal MTU traffic. Therefore, it is not expected impact in interfaces was are not receiving frames with higher MTU.

On 3750

Switch(config)#system mtu ?

  <1500-1998>  MTU size in bytes

  jumbo        Set Jumbo MTU value for GigabitEthernet or TenGigabitEthernet

               interfaces

  routing      Set the Routing MTU for the system

 

Switch(config)#system mtu jumbo ?

  <1500-9198>  Jumbo MTU size in bytes

 

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 are 1500 bytes. Users cannot change the MTU on an individual interface.

On 4500

switch_4500(config)#system mtu ?
  <1500-1552>  MTU size in bytes


switch_4500(config)#interface fastEthernet1

switch_4500(config-if)#mtu ?

 <64-9198> MTU size in bytes 

 

Note:

 The key concept to keep in mind is that all the network devices along the communication path must support jumbo frames. Jumbo frames need to be configured to work on the ingress and egress interface of each device along the end-to-end transmission path. Furthermore, all devices in the topology must also agree on the maximum jumbo frame size. If there are devices along the transmission path that have varying frame sizes, then you can also end up with fragmentation problems. Also, if a device along the path does not support jumbo frames and it receives one, it will drop it.

Recommendations

Problems with MTU size reduction due to tunnels, IPsec encryption, and overlay protocols can be problematic. If you are using encapsulation technologies then you should consider increasing the MTU size. It would be nice if we were able to increase the MTU size in the core of the network or WAN to avoid the fragmentation and PMTUD issues.