Skip to main content

Nine things to consider for your Windows Server 2003 migration

It’s been a long time coming, but on 14 July this year the support for Windows Server 2003 and Windows Server 2003 R2 will end. (opens in new tab)

This may well prove to be a challenge for some IT professionals, but it certainly doesn’t have to be a problem. Here, Rory Higgins, product manager at Eaton, outlines nine points to help IT professionals prepare for life beyond Windows Server 2003 and even to discover that this unavoidable change is actually an opportunity.

“If it ain’t broke don’t fix it” is excellent advice most of the time, but not when unsupported operating systems (OSs) are being considered. Of course, you could decide to stick with Windows Server 2003 (opens in new tab) when support ends and, if you do, your systems will continue to work – for a while anyway. You will, however, be increasingly exposed to security risks as time goes on. In fact, to put it bluntly, in reality moving to a supported OS isn’t optional – it’s a business essential.

But what do you need to consider to make sure that your migration isn’t just successful, but also that it delivers significant and lasting benefits? Let’s take a look.

  1. Think power

If you’re installing new servers as part of your upgrade, as many organisations will be, and possibly adding a few more to cope with future expansion, your power requirements will undoubtedly increase.

Like many others, if you decide to virtualise and consolidate workloads, it is worth thinking about what will happen to your virtual servers and valuable data when the power goes off. Even if you decide to move some workloads to the cloud rather than installing new servers, you’ll probably need to upgrade your network switch infrastructure and bandwidth.

It will be essential to ensure that adequate – and dependable – power is available to support these critical network components. Not factoring in power as a part of your IT application upgrade may lead to increased risk of compromising your overall business continuity.

  1. Consider an integrated approach

No one wants to work with 20 different tools and dashboards, and today there is no need to, as it’s possible to integrate all layers of the IT applications and monitor everything from a single pane of glass. Power management is a part of such an integrated approach: the best power management software readily integrates with leading virtualisation environments like VMware vCenter, Citrix XenCenter and Microsoft SCVMM.

Using this software makes it possible to implement comprehensive monitoring and management of power devices from your virtualisation dashboard together with server, storage and network devices, all from that single pane of glass. If you decide to replace your IT application with converged infrastructure solution, ensure your power management solutions are also validated for this.

  1. Recognise the importance of business continuity strategies

Modern power management solutions - uninterruptible power supplies (UPSs) and rack power distribution units (PDUs), not only integrate with virtualisation platforms for monitoring and management purposes – they are also key for implementing business continuity policies on power and environmental events.

Typically you’ll want to keep your critical applications up for as long as possible. You can do this by prioritising application and shedding non-critical loads. This will extend battery runtime and the extra runtime gained could be the difference between a minor power event and a major extended downtime issue.

Power and environmental alerts can also trigger live migration of virtual machines to a zone or a backup site that isn’t affected by an adverse power event. They can, in addition, initiate the replication of critical data to the backup site and the controlled shutdown of devices in cases of prolonged power outage.

  1. Maximise efficiency, minimise costs

Make sure that you maximise the benefits you get from your OS migration exercise. As well as the benefits that power system integration can bring, consider also the improved energy savings your new power infrastructure could deliver.

Remember though that older UPSs are less efficient than their modern counterparts, so it’s perfectly possible that new units will pay for themselves in a relatively short time because of the energy savings they deliver. The latest UPSs are exceptionally energy efficient – the outstanding protection of a double conversion system can now be combined with an efficiency of around 98 per cent.

Take into account that this doesn’t just mean reduced energy bills for the UPS system, you’ll also need less cooling as new generation rack PDUs can now work continuously at up to 60ºC without derating thus you’ll also see your HVAC energy bills shrink.

  1. Plan for the future and pay as you grow

Think ahead about your likely future needs, but don’t be tempted into expensive and unnecessary over provision.

Instead, look for modular UPSs that meet your needs today, and can be expanded easily later, removing the need to over invest now as an attempt to future-proof requirements. This will allow you to implement a pay-as-you-grow approach.

  1. Check and optimise power consumption of your IT hardware

Remember that a UPS to protect the power supply to your IT applications is not the only thing you need for an optimum power solution. You should also think about how you distribute power to your IT devices and how you measure and control their power consumption in an intelligent way.

Use rack mountable PDUs, which not only distribute power, but also measure power consumption down to socket level, if required. The best types have meters that can measure energy usage to IEC ±1 per cent billing-grade accuracy. This means users can quickly determine exactly where energy is being used, ensuring that rogue hardware that is consuming more energy than it should is quickly identified.

Accurate metering also simplifies load balancing and reveals locations where there is spare power capacity. Furthermore, billing-grade metering means that the energy data provided by the rack PDU can be used to apportion costs between company departments in enterprise installations and between clients in data centres.

  1. Check the capacity of your utility supply

Installing additional UPSs to meet the increased power requirements of your upgraded infrastructure is a logical step, but it’s also important to be sure that your utility supply system can support the extra load.

If there is any doubt about this, check with your facilities or building management team and, if necessary, make provisions for increasing the supply capacity.

  1. Replace or retain the UPSs?

As previously mentioned, there are many good reasons for replacing the existing UPSs as part of your upgrade as ageing power protection solutions are not always able to meet the requirements of modern IT applications.

However, if you do consider retaining old UPSs, check their age and the battery replacement date. If your existing UPSs are soon going to need new batteries, there’s an even stronger case for fitting new units as the batteries represent a significant proportion of the total cost of a UPS.

  1. Ensure complete peace of mind throughout the life cycle of your solution

Choose power devices whose flexible, modular design and compact size makes them easy to install and use; this will free up considerable valuable space for your IT applications.

And, finally, don’t forget warranty and support. You’ll want your systems to deliver peak performance throughout their lives, so look for a supplier that can offer support services that match your budget and business needs, and is prepared to back its products with a full and fair warranty.

Forced upgrades are never welcome but, in this case, migration from the venerable and increasingly outdated Windows Server 2003 (opens in new tab) environment to a more modern and fully supported OS is much more of an opportunity than a threat.

Follow the guidelines discussed in this article and work with an experienced vendor that can fully support your migration exercise and you’ll emerge with systems that are far more flexible, energy efficient and reliable, which will serve you faithfully for many years to come.