It goes without saying that uptime is the lifeblood of any eCommerce operation. Every minute of downtime equals lost revenue and a prolonged outage can cause irreparable harm to a company’s reputation, especially at peak seasonal shopping and aggressive marketing periods.
That is why infrastructure is so important. The more resilient and scalable your infrastructure, and the more dependable and vigilant your provider, the less chance things will go wrong. If infrastructure works the way it should, your business will hum along with it. Sales will pour in and customers will get an enjoyable customer experience they will remember and likely recommend to others.
But getting IT infrastructure right is never easy. eCommerce traffic is notoriously variable and suffers from periodic fluctuations. For many online retailers, business is seasonal and rises substantially during certain times of the year, only to dial down shortly thereafter. Peak seasons like Black Friday, Christmas or even Valentine’s Day generate short surges in demand. And this can wreak havoc on existing systems and leave retailers short on capacity. That is not all. Spikes in traffic generated by special promotions or marketing campaigns can have a similar effect. There are even unpredictable events like a mention on a major publication that drives an expected wave of traffic.
There have certainly been real life examples of infrastructure being overwhelmed and failing to meet expectations. Just this past Christmas Netflix went down on Christmas Eve - one of tits busiest days of the year. Problems with the Amazon Web services platform caused the movie streaming service to be unavailable for hours. Several years ago, we saw Yahoo’s infrastructure collapse under the weight of Black Friday - arguably the most important day of the year for many of its online retailers. And just over a year ago, Target’s eCommerce websites suffered a series of outages. These are just a few of the many IT infrastructure troubles we have seen in the eCommerce world.
The good news is that it doesn’t have to be this way. There is no need to panic. The best approach is to work with an outsourced infrastructure provider and plan ahead well in advance.
What are some of things you can do?
1. Use cloud infrastructure. Better yet, use managed cloud infrastructure.
Cloud enables businesses to scale infrastructure resources on-demand. If there is a spike in traffic it is easy to scale-out capacity very quickly to handle the increased load. Likewise, it is easy to scale down once the demand subsides. The great thing about cloud is scaling can be done automatically. If traffic is anticipated during certain times of the week or during special events you can automate the scale-out or re-sizing of infrastructure in advance. Clouds even have alerting mechanisms that signal a need for scaling. A managed cloud provider is a great asset if you don’t have a lot of IT resources and can take over responsibility for all of this and back it up with an SLA. Cloud providers have seen numerous repeated trials and have the experience and knowledge to help scale infrastructure in an effective and affordable manner. Trusting them makes sense.
2. Be sure to have a disaster recovery solution.
Don’t be left without a Plan B. There are various different DR scenarios. You can build an active-active deployment with two “hot” sites mirrored and ready to go if one fails. Or it could be a good idea to backup infrastructure to the cloud and select a recovery time that is acceptable. Another course of action could be to build a solution that is distributed across multiple facilities with each node backed up into another facility or geography for added redundancy. Some of the scenarios can get complex and having a provider help advise, design and implement is a smart way to get things right.
3. Use an outsourced provider with real managed services capabilities.
One of the big mistakes we have seen is a failure to work with infrastructure specialists. Amazon is a great platform and Netflix clearly has an abundance of technical talent. But even with all those resources it is questionable whether there is actually a truly coordinated managed services capability there. Amazon opens up its platform for people to use and provides a great toolset. But the rest is up to you. A hosting infrastructure provider takes a different angle. It helps organizations every step of the way from planning and implementation to maintenance and monitoring. Infrastructure providers are there to do one thing – maintain uptime – and that service is available 24/7/365. Working with an infrastructure specialist eliminates guesswork, brings an experienced hand along with established and mature systems and processes that simply can’t be replicated at a reasonable cost.
4. Do you homework when it comes to providers.
There are many different kinds of service providers with various different areas of specialization. Some providers specialize in eCommerce and this kind of rare skill set could be invaluable. Other providers are great at building hybrid environments or operate high-performance global networks. Understand your requirements and find a provider that can meet them. But most importantly, find a provider that doubles as a trusted partner. Your provider should be a seamless extension of your internal IT department and a resource you can count on any time of the day.
Remember, it is never too early to start planning ahead, even 12 months ahead. All these infrastructure mishaps are avoidable if you prepare and build out the redundancy and scalability necessary to deal with the unpredictability of eCommerce traffic. There is no better way to ensure this gets done the right way than by working with a helping hand.
About the Author
Post new comment