When it's time to replace a server it can be tempting to reduce capital costs by buying from the grey market that has continued to expand over recent years. With economic pressure on IT managers to reduce their capital costs where possible, this can be highly tempting.

Some IT managers are using a different strategy, shifting servers approaching their replacement anniversary to other less demanding duties. Units for mail or print that require relatively low levels of processing power can be shifted into these roles, leaving capital budgets to purchase new servers for mission critical or business development purposes.

Deciding when a server needs to change its role within your company is the key question IT managers often ask themselves. Using clustering techniques, throttling to reduce power consumption and of course virtualisation can extend its life, but at some point the risk of data loss or complete failure becomes too great.

Generally, a server more than five years-old will have 21% more downtime than one that is three years-old.

In a white paper on the subject, market intelligence firm IDC observed: "In many cases, extending the life of servers too long can lead to an increase in operational expenses that could pay for investments in new technology. Before simply delaying new hardware purchases, organisations need to assess the impact on both capital budgets and operational budgets."

The decision when to switch the role of a server or to purchase new hardware should form the core component of a server roadmap. The overall needs of a business should be mapped and the current server deployments noted.

The future growth of the company will then give an indication where the server network will have to change, and which servers can fulfil which roles.

Grey is good?

IT managers should think carefully before investing in grey servers, as they could be a false economy. Budgets are tight across every business sector, but it is important to fully understand the risks that the gray server market can potentially bring.

In a report on the grey IT market, KPMG concluded: "For end customers, faced with the sometimes daunting task of discerning which entities are authorized to sell genuine goods and which are doing so through correct channels, a decision based solely on price may be the best bet, but that bet could go awry fast."

A better approach is to look at a business's server landscape and assess where each server is in its lifecycle. This will provide a clear understanding of the potential failure rates.

With this information in hand, it is possible to increase servers' levels of virtualisation increased, or modify their roles based on the forward planning of server need. It will then become clear where new server capacity is needed and make it possible to minimise the risk of server failure.