Benjamin Franklin’s old saying, “An ounce of prevention is worth a pound of cure,” is probably not thrown around the managed IT services industry too often. If you think about it, though, it is really well suited, as data center redundancy, when provisioned as a fail-safe, can save the day in times of unexpected demand or equipment failure.
Like insurance, airbags, or a parachute, you never truly appreciate network redundancy until you need it. If you don’t have adequate redundant hardware and software available when traffic spikes or a DDoS attack occurs, the cost of deploying it for the next time will seem like a bargain.
Here are four ways you’ll be able to meet your service level agreement promises with a redundant infrastructure.
Power and Telecommunications
Just like with plumbing, you can have great endpoint hardware, and awesome pressure in terms of content going through it. Yet if the “pipes” transporting it doesn’t have the capacity or quality to sustain the flow of, in this case, data and power, even the best servers can’t bail you out. Where available, invest in multiple energy suppliers and telco service providers.
Often, hosting or colocation companies can pass the cost of utility redundancy on to end-user customers, by either offering it as a value added service to customers or an optional add-on to offset the cost.
If one wireline provider’s power grid fails, not having the option to fail over means a miss on your SLAs, loss of productivity for your stakeholders, and damage to your reputation. Keep power and data flowing to and from your hardware and software.
N+1 Redundancy for Hardware and Software
Having a redundant environment available in your data center means going all in. Leveraging resources like:
- Virtual servers backing up physical hardware, or vice versa
- Data redundancy within database environments
- Multiple security appliances at the perimeter and endpoints
- Physical server clustering or load balancing configuration
Matching each data center technology asset with a replacement, which can be brought online immediately in case of anticipated or actual failure or high latency, goes a long way toward preventing your services from failing, or degrading the quality of your services.
Regional Redundancy
What if a natural disaster, local grid failure, or other regional outage occurs that’s beyond your control?
Having a redundant environment available across the country or in another nearby state can maintain your uptime, profitability in the short term, and customer satisfaction levels.
Contracting with a strategic, non-competitive data center provider, or contracting racks or modules in other data centers can help you realize the value of a redundant environment in another geographic location, without all the costs associated with running multiple data centers yourself.
Cooling and Physical Security
If a seasonal- or sudden spike in traffic occurs, and your servers’ cooling systems fail, all your efforts to establish redundancy could be for naught. Provisioning a secondary, fail-over cooling precaution is wise to ensure your network doesn’t suffer damage from peaks and valleys in extreme heat.
Precautions like retinal scanners, along with swipe card security or multiple security protocols (similar to multi-factor authentication), are wise steps to keep the “bad guys” out of your facility.
There are many strategies for provisioning data center redundancy in your server rooms and their related infrastructure. These tips will mitigate downtime and prevent the long-term ramifications of outages and poor performance.
Have you had events in your data center that made you thankful for implementing redundancy measures? Tell us about them in the comments section.
If you're in the data center, mission-critical, or cloud services industries, or you sell to the data center industry, don't miss our weekly update newsletter -- Data Center Sales & Marketing Institute (DCSMI) Update Newsletter. Get notified about new reports, events, podcasts, and blog posts.
Learn more about Data Center Facilities Companies and Go-to-Market Strategy (GTM) for Growth.
Submit a comment