When You’re Decking the Halls, Don’t Forget to Prep the Data Center

October 11, 2019

3 Tips for Readying the IT Infrastructure of Retail Support Sites

It used to be that malls didn’t put up their trees or turn on the holiday lights until the Friday after Thanksgiving, officially kicking off the holiday season on the busiest shopping day of the year. These days, the holiday rush starts well before then with many retailers having their red and green displays up before Halloween and kicking off their “biggest sale of the year” hours well before the turkey goes in the oven.

Whether or not you’re inclined to fill plastic pumpkins with candy canes, one thing every retailer needs to be doing sooner rather than later is ensuring their digital infrastructure is ready to support all the holiday visitors that will be coming to town via online retail sites.

Last year, several big-name retailers, including Walmart and Lowes, learned this lesson the hard way when their sites crashed over the Thanksgiving weekend. Retail aggregate LovetheSale.com estimated that the technical issues impacted more than 3.5 million customers and cost retailers $9 million in lost sales. The outcries from frustrated shoppers on social media sites only made matters worse.

While last season’s glitches were likely due to heavier-than-anticipated retail site traffic, it’s a good reminder to retailers to do everything in their power to ensure their digital infrastructure can support business operations during the busy holiday shopping season. A temporarily jammed e-commerce site is one thing, but a full or partial data center outage is a different story with much more serious consequences in the future of retail.

According to the last Cost of Data Center Outages report by Ponemon Institute, the average cost of a data center outage has steadily increased to more than $740,000 with benchmarked data centers reporting an average cost of $9,000 per minute per incident. In addition to lost revenue, this cost includes the cost of detecting and correcting the root causes of the outage, lost productivity, and reputational damages. An outage is bad news any time it happens, but in the weeks leading up to the holidays, it can turn the most wonderful time of the year into a literal nightmare before Christmas.

Here’s how to get your data center ready for the holidays:

1. Inspect assets to assess their health

Given that many retailers lock down or limit access to their data centers from mid-November through January as a preventive measure to safeguard against human error or security risks, now is the time to make sure your data center is ready to handle anything that could cause downtime, from a power outage to a network failure. Start with scheduling a thorough on-site audit of your data centers. If you have a distributed network with multiple edge facilities, this can become a logistics challenge. Entrusting the task to trained service representatives can help you cover a lot of ground quickly and gain professional insights. Often this type of inspection or audit is a regular part of an existing service agreement requiring no additional expense. And it’s well-worth the effort.

Vertiv recently helped one major retailer audit the data center supporting its distribution center and found that half of the uninterruptible power supply (UPS) units didn’t work. If you have critical equipment issues that demand immediate attention, it’s better to find out now while you have time to correct the problems before business really picks up.

Along with the performance of a data center’s power and cooling systems, auditors and technicians can also check for additional issues that impact the availability and performance of critical systems. This includes heat densities in racks and rows, raised floor obstructions, and arc flash vulnerabilities in the electrical infrastructure. Learning about these issues can help data center managers understand the big picture when it comes to availability and make plans for needed improvements. 

2. Use audit results to prioritize upgrades

Based on learnings from the audit, you can create a roadmap that prioritizes your IT budget to accommodate future upgrades and support overall asset management. Given that the leading and most costly root causes of data center downtime are directly related to vulnerabilities in your power and cooling infrastructure, these are good places to concentrate any improvement efforts you need to make before the holidays.

  • Consider restoring, refreshing, or replacing UPS units

A reliable UPS coupled with regular preventive maintenance and monitoring is the key to maximizing uptime. Vertiv research shows that even a single annual preventive maintenance visit can increase the mean time between failures (MTBF) of a UPS unit by more than ten-fold. Auditing will help you ensure power continuity by determining if it’s time to replace UPS batteries or, depending on age and operating environment, possibly the units themselves.

If replacements are in order, it’s a good time to consider investing in new battery and UPS technologies that can increase the reliability of your backup power, such as lithium-ion batteries, which offer extended life and reduced maintenance. You may also want to think about investing in a power protection plan, such as the Power Assurance Package from Vertiv, which provides oversight of equipment and rapid incident response, ensuring your business-critical continuity.

  • Check for adequate cooling and proper cooling strategy

Data center downtime events related to thermal systems are not always caused by cooling equipment failure. More often than not, they can be attributed to inadequate cooling infrastructure. Optimizing your cooling strategy may involve reconfiguring assets, moving to row-based precision cooling solutions, or implementing cold-aisle containment, if you haven’t tried these methods already.

You may also want to think about investing in new technologies and products that can help you speed deployment in support of the rapid changes happening in the retail industry such as the new Vertiv™ VRC rack cooling system. It is a thermal solution that fits quickly and easily into network closest or edge environments and offers a plug-and-play solution, making cooling possible in places where space cannot accommodate more traditional cooling products.

3. Add some level of oversight

The more distributed IT networks become, the harder it is for internal IT departments to manage every piece of equipment in every location. Today, many software and hardware solutions exist that enable remote monitoring of equipment performance from a central location, which can be particularly important for retailers that limit physical access to the data center during the holidays.

Service providers like Vertiv also offer a variety of programs such as LIFE™ Services for overseeing the health and status of power equipment and cooling infrastructure. Whether you monitor yourself or use expert support, this oversight can alert you to potential problems before they cause downtime and help you optimize the efficiency of your IT staff. 

 

Get ready for a downtime-free holiday season.

Before the holidays truly kick into full gear, take some time to outage-proof your data center as best you can. With audits, preventive maintenance, and monitoring, you’ll put your business in the best position to sustain critical operations all season long and avoid the detrimental business impacts of ill-timed downtime. You’ll give your customers the gift of e-commerce sites they can depend on while helping your business enjoy a busy and profitable holiday season.

  

Feel free to subscribe to our blog to stay informed on the latest trends, technologies and news.

Language & Location