Networking

Implement regular monitoring and improve Cisco router performance

David Davis provides advice on what you can do to properly monitor your routers and suggests some tweaks that you can apply to enhance router performance.

Do you know how your routers are performing? What could you do to improve performance? I'll try to answer those questions by giving you some tips on monitoring and improving your router performance.

What is meant by router performance?

What network admins actually mean when they talk about router performance could vary from person to person, as there are several different aspects involved. Here are some of the issues related to how well your router performs:

  • Resource issues, such as the performance of the CPU and RAM
  • Router IOS configuration changes
  • Bandwidth management: Quality of Service (QoS)
  • Layer 1 network issues: Bad circuits or cables
  • Errors and failure of the router hardware

Let's cover each of these areas of router performance to learn how to monitor and potentially improve router performance.

Router resource issues

Just like any PC or server, routers have a CPU, RAM, and backplanes where the data flows between these component parts. Thus, at some point, every router's resources can be pushed to their maximum.

To help you get more details about your router's resource usage, there are some GUI performance tools available such as SDM, PRTG, or MRTG. Besides monitoring router CPU and RAM, these packages can also monitor interface utilization.

From the command line, here are a couple of very useful monitoring commands:

Router# show processes cpu
CPU utilization for five seconds: 5%/4%; one minute: 5%; five minutes: 5%
 PID Runtime(ms)   Invoked      uSecs   5Sec   1Min   5Min TTY Process
   1           0        13          0  0.00%  0.00%  0.00%   0 Chunk Manager
   2          28    675969          0  0.00%  0.00%  0.00%   0 Load Meter

And:

Router# show processes cpu history

Here is what its output looks like:

Router output
Click image to view router output.

You can also use show memory to get into detail on how the router's memory is being utilized.

What can you do to the router to resolve these resource issues? Well, the easiest answers are always either to send and receive less data, control what is sent with another bandwidth management solution, or to replace the router hardware with a newer, faster, router (for more information, see my article, "Case study: How a Cisco router upgrade affects network performance"). However, those aren't always possible and, as network admins, it is our job to get the most out of the hardware we have.

Router IOS configuration changes

Outside of controlling bandwidth utilization with your router (which we will talk about next), there are a few things you can do in your IOS to improve relative performance for the end users.

  • Change your method of switching used by your router - routers can actually process packets in different ways. For more information on this topic, see:
  • Tweak router buffers: While the default router buffer configuration is often optimal, there may be reasons you want to tweak the router buffers and this may lead to performance improvements. See Cisco's Modifying the System Buffer Size (Note: I don't recommend doing this without support from Cisco.)
  • Disable unneeded services: Perhaps someone enabled debug mode on the router or configured NetFlow or other resource-intensive applications. Another option may be that your routing table has grown over the years, and you need to do some route summarization, use stub networking, or just turn off default routing on an older, underpowered router.

Bandwidth management: Quality of Service (QoS)

To resolve performance issues or improve performance, you may need to implement some form of bandwidth and/or traffic management. This is commonly called Quality of Service (QoS) but there are many different types of QoS.

My recent article, "Control unwanted traffic on your Cisco router with CAR" covers one type of QoS. If you look at the Cisco IOS Quality of Service Solutions Configuration Guide, you will see that Cisco breaks these down into ten different categories, each with their own subcategories. If you are interested in using your Cisco IOS router for a QoS solution, I suggest you spend some time researching the different types of QoS available before selecting and implementing one.

One quick way to see what the utilization is on your LAN or WAN circuit is to use the show interfaces command and look for the TX/RX Load as well as the five-minute input/output rate. Here are some examples of the show interfaces output that I am referring to:
reliability 255/255, txload 1/255, rxload 1/255
5 minute input rate 0 bits/sec, 0 packets/sec
5 minute output rate 0 bits/sec, 0 packets/sec

I have personally used these to determine what is maxing out a router's circuit and to see in what direction that traffic is flowing.

Layer 1 network issues: Bad circuits or cables

Many times, the reason that users are complaining about performance is that there is a Layer 1 (Physical) network issue. For example, there is an issue with an Ethernet LAN cable or a T1 WAN connection. Errors that cause slow performance are especially common with WAN connections that span long distances.

The best way to check to see if your LAN or WAN connections are causing the slow performance is to use the show interfaces command. For more information on that, please see my article, "Getting to know the show interfaces command."

Errors and failure of the router hardware

While the show interfaces command might find issues with your connections, those errors could also be caused by your router hardware. For example, you could have a bad HWIC T1 card that is causing slow performance and causing errors to increment in the show interfaces output.

If this is a WAN circuit, many times, your carrier will assist you in testing and troubleshooting that circuit. Unfortunately, many carriers make you first prove that your hardware is not failing by replacing the router or the HWIC card.

Be proactive

Optimal router performance is critical. Never do you want to have your CIO come and tell you that a network connection is slow. You should be proactively monitoring those connections, and you should be able to quickly and easily check the issue causing that slow performance.

Want to learn more about router and switch management? Automatically sign up for our free Cisco Routers and Switches newsletter, delivered each Friday!

12 comments
samisas2
samisas2

its good i want one solution i have the two internet connection one is airtel & vsnl i have 1900 series of switch i want the configuration . switch to connect two connection. if any one connection is don another connection is up attomatically plz send the configuration thanz emailaddress:-samisas2@yahoo.co.in

gk
gk

Hi David, I like all of your articles. I think Overall processes are covered however it would help a lot if you could also add automated process/tools/notification (to email or cell/pager). Its hard to perform these tasks manually on a regular basis in medium to large network environments.

a2makarov
a2makarov

The article itself is very informative and can definitely help you in troubleshooting, BUT.. It's named "Implement regular monitoring and improve Cisco router performance" and I haven't found anything disclosing the first part, i.e. regular monitoring. Is it supposed to do manual monitoring by logging into each router on the network and lookinh through logs, interfaces , etc. What I really was hoping to read is a story about different tools used for regular monitoring, how they migth help to reveal network problem, how then can prevent potential problem, etc.

zeeshan.saleem01
zeeshan.saleem01

Good one..... Can you give us the detail commands document In which every command function is given is detail Thanks

Chris910
Chris910

I would like to see a couple of comprehensive articles about: -Cisco devices in a mixed network such as a Cisco router used as a gateway and Dell or HP layer three Switches for routing the VLAN's. -Vlan Trunking/Tagging between Cisco, Dell, HP. -Etherchannel/loadbalancing between models. -Routing protocols in a mixed network. -Common Cisco Commands and their equivelant on Dell or HP. -Help me be more confident when my new customer opens up the server closet and all of the the switching gear is Dell/hp/other.

zloeber
zloeber

I like the coverage of this article, it is quite comprehensive in an easy to digest and well thought out format. I might add one extra section near the Qos area on snmp monitoring over a period of time (with perhaps a pre-canned cacti vmware virtual appliance) to detect traffic trends. Also, for the sh int command a quick glance at dropped in or out packets and CRC errors (after clear counters) can quickly determine point-to-point line issues. Finally, if you are running a router with many routes you can try to setup route summaries and reduce multiple routing protocols from running on the machine to help increase performance. FYI, I love these tips of the trade type articles, it sure as heck beats reading through the (often dry) cisco documentation!

ddavis
ddavis

What do you all think? Was this article helpful? What other topics would you like to see?

raycamara
raycamara

There are a lot of us out here that either haven't had much experience with Routers, and switches, or little or no training, and I think some basic How to Questions and answers on configuring the Cisco Routers and security issue are in need here. See this link for questions, and ansers. Apparently we will need to know more as time progresses, and your forrum is much appreciated, and needed. http://www.techinterviews.com/?p=319

ldimeglio
ldimeglio

Suggesting that people be proactive is great. However, I think you miss something in this piece and that is that network monitoring can't occur in a vaccuum. Network monitoring gives you one piece of the picture, but without monitoring of the servers, and the apps that run on them, you still might get a call that the "network is slow" and will have no visibility to combat that. Having a tool that gives you a single view in to all layers of the stack allows you to prevent slowdowns and failures no matter the cause rather than just concentrating on the network layer.

CrimsonPaw
CrimsonPaw

You did a good job as these things are often overlooked or not checked at all. One item I would have touched on is logging either internally or, more preferably, to a syslog server. In my opinion this is something that can dramatically help with troubleshooting issues. Keep it up, I truly enjoy getting these knowledge bites in my inbox every Friday!

Anthony.V
Anthony.V

Great article! Very informative.

portcullischain
portcullischain

I've linked it as well as your piece on securing Cisco routers over at my site.

Editor's Picks