Manage Learn to apply best practices and optimize your operations.

Network management - Is it time for an upgrade?

Upgrading a customer's network is an important part of any VAR's job. However, the task can be fraught with compatibility dangers and utilization risks that make knowing when to upgrade a network difficult to determine. Here we have a list of symptoms and warning signs to look for and be aware of.

There are several different drivers for network upgrades, and one upgrade has a nasty habit of driving several others. As you're analyzing your customers' networks, use this checklist of drivers and symptoms that indicate when to seriously consider an upgrade.

When you're inspecting network hardware, check the following capacity metrics:

  • CPU utilization
    This is normally very low for network hardware, but enabling certain software-based features on network equipment can quickly make this a bottleneck for performance. If CPU use is high, look for a hardware-based solution, such as offloading IPsec encryption, placing the VoIP codec onto dedicated hardware or upgrading the processor.
  • Memory utilization
    This can be complex to determine if the memory is shared between system memory, packet memory and variables like how many peers or routes are tracked. Your options here are to either tweak the allocation of your existing memory, or add more memory.
  • Interface utilization
    The trick with this is that it can vary widely because data traffic is notoriously bursty. You'll have to measure traffic for a period of time and compare it to any performance issues users may be reporting.
  • Number of ports in use
    Sometimes you're adding users and run out of ports on your switches. However, careless labeling of wires can lead to a situation where some switch ports are cabled, but not actually in use. It may be helpful to view the statistics of ports that are in a "down" state to determine when they were last used.

When you're already upgrading another component, like applying a security patch or new code, check other components for the following:

  • Hardware compatibility list
    The new code may be larger, requiring more flash memory or RAM.
  • Release notes
    Occasionally, you'll find something scary in the notes. For example, notes may report a known bug, or explain that the company is dropping support for some ancient technology -- like DECNet, Appletalk or IPX -- that's critical to an existing business application.
  • Interfaces to other equipment
    You may need to get new fiber cabling if, for example, your old hardware used GBICs and your new hardware uses SFPs.

You should also consider the following non-technical upgrade drivers:

  • End of software support/end of life
    These are the dates the hardware or software vendor publishes that indicate when the product will no longer be eligible for support. If you want options like security patches and helpdesk support, you'll need to upgrade before these dates pass.
  • End of lease
    If you lease your hardware, there may be a financial incentive to upgrade.
  • Vendor trade-in specials
    You may be able to take advantage of a trade-in program if your timing is right.

When you're adding an application or new feature to the network, you should plan to check the items listed above to make sure you won't cause a future capacity or compatibility issue. You should also do some research on the vendor's Website and on sites like sister site, to see what upgrades are typically recommended or widely used.

About the author
Tom Lancaster, CCIE# 8829 CNX# 1105, is a consultant with 15 years of experience in the networking industry. He is co-author of several books on networking, most recently,
CCSP: Secure PIXand Secure VPN Study Guide, published by Sybex.

Dig Deeper on Campus area networks and services

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.