Uptime 101 – 20 Lazy Ways To Keep Your IT Working

When your IT stops working, everyone notices. Whatever the cause, your IT team gets blamed and there's a mad scramble to fix the problem.

But it doesn't have to be this way. Many problems are avoidable, if the right technology is in place.

Here are 20 ways you can minimise IT downtime and keep everything running smoothly.

Keep your email up and running

• Switch to Office 365 or Google Apps for Work. One option is to let Microsoft or Google manage your mail server. While the SLAs on their hosted email services only guarantee 99.9% uptime, the actual performance levels are significantly better than that. For example, Microsoft disclosed that Office 365's quarterly uptime figures are typically 99.97% or better. That's good enough for most businesses.
• Stick your mail server in its own virtual machine. This stops other server applications from crashing your mail server, by reducing software compatibility issues and stopping other apps from hogging the resources your mail server needs to function.
• Pick a backup solution that's email-friendly.Select a backup solution that lets you restore individual email messages and folders, rather than just the whole email database.
• Subscribe to an email continuity service. These accept email intended for your domain, queuing it up for when your mail server is back online. In the interim, they may provide web-based access to recent emails, enabling staff to work even when your mail server is down.

Safeguard your website

• Use CloudFlare to reduce the load on your web servers. Designed to accelerate and secure websites, this freemium service provides a reverse-proxy that acts as an intermediary between your web site and its visitors.
• Get a VPS to boost uptime. Virtual private servers offer better uptime than shared web hosting.
• Use 'Hosted Virtualisation'. This is the next step up from a VPS. It offers far greater control over the operating system environment and access to more resources. Some hosted virtualisation services incorporate live migration of workloads from one server to another, allowing your website to stay online even if there's a problem with an underlying server or data centre.
• Put a firewall between your site and the Internet. These stop some unwanted traffic from reaching your servers.
• Subscribe to a Denial of Service mitigation service. Firewalls are good, but they can only handle a limited volume of traffic. Network-level DoS mitigation services can cope with far more. They use fancy algorithms to filter out unwanted requests, helping protect your servers and firewall from being overwhelmed.

Boost your office's connectivity

• Install a backup connection. It doesn't have to be as good as your main one. It just has to be good enough to tide you over until the connection problem is fixed. Network monitoring tools can help you determine how much bandwidth you really need.
• Set up 'autofailover'. This will ensure that you automatically switch over to a backup connection, in the event there's a problem with your primary connection, without your IT team having to manually intervene.
• Choose different carriers for your primary and backup connections. That way, you reduce your dependence on any one physical network.
• Make sure your connection is monitored & supported 24x7. The sooner problems are spotted, the sooner they can be fixed.

Keep your internal servers running

• Partition your server applications. Keep your major applications apart from each other, in separate virtual machines. That way, a crash by one application won't drag down everything else on the same server.
• Give your servers the right hosting environment. One with reliable power, powerful air conditioning and strong physical security. Colocation is the simplest option here – rather than spend a fortune building the proper hosting environment, you just pay a hosting company (or an ISP) to host your server in a high-spec data centre.
• Regularly backup your internal servers. So if data is lost or corrupted, you can roll back to an earlier working version. If you've virtualised your servers, it may be worth taking snapshots of your virtual machines. These allow you to restore the whole working environment, not just the files.
• Consider opting for Continuous Data Protection. Daily backups are great, but even when they work perfectly, you stand to lose up to a day's worth of emails and changes. If that's unacceptable, consider upgrading to a backup solution that backs up your data continuously.

Safeguard your office's phone system

• Buy phones that support 'Power over Ethernet'. That way, you can protect all your office phones against power cuts by buying a relatively small number of PoE switches and plugging them into an Uninterruptable Power Supply (UPS).
• Plug on-site PBXs into a UPS. If your business uses an on-site PBX, make sure there are at least two connections to the phone network. These could be SIP trunks run over two separate data connections, or a combination of a data circuit and an ISDN circuit.
• Make sure any cloud-hosted PBX service you use is resilient. If you opt for a cloud-based PBX service, check that the service hosts your PBX in at least two data centres, a fair distance apart, so your PBX isn't vulnerable to regional flooding or local power cuts.
• Set up 'Disaster Recovery' call diverts. Some phone companies and phone systems let you set user-specific 'disaster recovery' numbers. These ensure that if your connection to the phone system is lost, inbound calls are automatically diverted to alternative number(s). For example, you may be able to set calls to be diverted to your employees' mobiles.

It's Not Just About Downtime Prevention

For many businesses, these steps will form part of a broader business continuity plan.

However, the steps aren't just about improving uptime and business continuity. Many have additional benefits.

For example, using server virtualisation to partition your server applications from each other doesn't just improve uptime. It can help you consolidate your IT systems on to fewer physical servers, simplifying hardware maintenance, cutting hosting costs and improving hardware utilisation.

By taking some of the 20 steps above, you can help keep your IT working and your business running smoothly.