A lot more incoming traffic: That’s what data centers face as IoT continues to infiltrate enterprises. Most of today’s data centers are designed for bandwidth requirements that involve human interaction with applications as data flows out of the data center.
IoT, on the other hand, is more about data collection instead of data distribution. In IoT-ready enterprises, more data will flow in to data centers. IoT brings more devices to our networks, along with large amounts data from the sensors embedded in these “things.” As the devices learn about their environments and users, they’ll continue to generate more data – and this data will need to be transmitted, processed and analyzed in real time if it’s going to improve business operations.
Real-time IoT data will increase data center workloads and bring new security, capacity and bandwidth and analytics challenges as we get IoT ready.
Surveillance cameras, thermostats and phones are already connected to networks in many facilities – but that’s just the beginning. In IoT-ready manufacturing facilities, production-line equipment and data will be directly connected to the corporate network instead of managed by a separate network. GPS systems connected to an organization’s vehicle fleet will ride on the corporate network, sharing data about vehicle performance and location. RFID tags that track an organization’s assets will connect to the network, allowing real-time access to each asset’s location and status.
Adding more devices means more data will be transmitted across cabling systems; this can amplify bandwidth and latency issues. Limited bandwidth will stifle IoT environments, restricting their capabilities. This may cause these organizations that aren’t IoT ready to fall behind as others move forward.
Smartphones, websites, cameras, computers and applications all generate data in different formats. Before you can truly benefit from an IoT-ready environment, it’s important to determine whether your organization is ready to handle “big data” (the data that will be provided from the devices connected to your network). Can you support and store the data? Do you have appropriate staffing levels to spend time analyzing the data so it can be used for improved decision making?
Just because data is generated by IoT devices doesn’t mean that it all needs to come through your data center. A good example: Do all temperature and humidity levels throughout the day always need to be sent through? Does data indicating that lighting levels are appropriate for the time of day and occupancy levels need to be sent to the data center? Perhaps not. You can decrease your data center requirements if you map out ahead of time which data will be relevant and helpful to your organization, and which data won’t provide a return on investment to track and monitor.
One way to deal with data center congestion is to move data closer to the edge, processing it at the edge of a network – closer to users or devices – instead of in a cloud or central data site. By doing this, you can determine which data can stay “local” and what needs to go back to the network. Instead of sending all data from global sources to one single location for processing, smaller data centers that are geographically distributed can complete initial processing faster and more cost effectively.
This data increase doesn’t just impact bandwidth; it also impacts backup storage capacity requirements. It will no longer be necessary (or even feasible) to back up every piece of information generated by IoT devices. Instead, data centers will have to be selective about the data they keep. How much storage capacity do you currently have? How much storage capacity is left? Whatever available space you have now will likely be eaten up as you connect devices to your network.
It will be tough for enterprise data centers to keep up with the demand and scale of IoT data if resources, time, personnel or budgets are lacking. In these cases, cloud and colocation providers may be an option worth considering if your enterprise data center isn’t IoT ready. They can offer infrastructure needed to support IoT, redundancy, ample storage and protection against downtime.
Belden can provide the behind-the-scenes infrastructure you need to support whatever lies ahead. Contact us so we can help your data center get IoT ready.
Throughout his career Stephane has worked in the telecommunications industry; in R&D, product management, training and marketing. Since 2014, Stephane as Director Technology and Applications is focusing on technology roadmap and ideation, networking applications and trends, and standards engagement.