How to Transform Infrastructure at the Edge

By Jyeh Gan, Director of Product Management, Marketing & Strategy, Extreme Scale Infrastructure (ESI), Dell EMC

Jyeh Gan, Director of Product Management, Marketing & Strategy, Extreme Scale Infrastructure (ESI), Dell EMC

The transformation produced by the Internet of Things (IoT) can only be compared to the global proliferation of the Internet and the rapid consumerization of smartphones. There is no arguing that IoT is fast growing, with its ecosystem projected to reach close to 30 Bn connected autonomous devices by 2020. The ability to connect multiple devices together, gather and analyze the raw data produced by these devices and transform the data into usable information makes IoT the next logical step forward in building an intelligent world.

In order to unleash the true power of IoT, data must be processed, stored and analyzed closer to the end user. To accomplish this, organizations need to bring this power to the edge of the network–or in other words, they need to leverage edge computing solutions as the brain behind IoT.

Edge computing—in its simplest form—puts compute and storage closer to the user at the “edge” of the network. With edge solutions, it is no longer necessary to have huge amounts of data carried upstream into the network to a mega-data center that is located far away and then bring it back to the user. By having local solutions that can respond faster, edge computing ultimately drives a better quality experience with faster response times for end users.

“Edge computing—in its simplest form—puts compute and storage closer to the user at the 'edge' of the network”

What’s the best way to move forward with edge solutions? Some try to do it themselves, exploring a variety of technologies and hoping they can be combined into a viable solution. But to enable edge computing in a variety of locations quickly and efficiently, edge solutions such as a micro Modular Data Center (MDC) should be considered.

Modular Data Centers

MDCs have been around for some time now. They are probably best known for large scale-out environments, where cloud providers purchase huge volumes of racks for their centralized data centers and have these delivered in a way that reduces their time to market—with IT, power and cooling housed inside a modular solution.

However, with the advent of IoT, there’s a need to purchase smaller modules—‘micro MDCs’—and quickly place these anywhere data capacity is needed, while still achieving the same time-to-market goals.

As a data center infrastructure leader who is thinking deeply about edge computing from both the IT and facility perspective, Dell EMC outlined the top six features customers should look for when considering a micro MDC solution.

The Top Six Considerations:

1) Ensure solution flexibility regardless of the environment: Because not all customer environments are the same, look for solutions that provide the flexibility required. The one-size- fits-all approach doesn’t always work well here. For example, will the solution be placed indoors or outdoors; will it leverage outside air or mechanical cooling technologies; will the micro MDCs be configured with all IT equipment or a mixture of IT and power and cooling? Look for a provider who can meet specific needs instead of forcing a solution that isn’t optimized for the environment.

2) Build an open and agile infrastructure: When it comes to IT, find solutions that offer the scalability, agility and manageability needed to distribute more processing power closer to the edge. While the IoT era opens new opportunities, it can also create headaches for IT departments as workload requirements vary greatly. Having the underlying infrastructure needed to automatically scale to different ratios of resources becomes even more critical in order to achieve service agility. One way to achieve this is by choosing infrastructure that supports Redfish, a new open management standard that improves security and allows for management of heterogeneous vendor systems

3) Empower analytics at the edge: It is one thing to process data at the edge and another to analyze data at the edge. Micro MDCs should have the capacity to perform analytics locally, close to the devices and sensors generating data to help deliver business-driving insights faster.

4) Accelerate time to value: Not only should micro MDCs be built to specific requirements with the option to be configured with compute, storage, networking, power and cooling and even IoT gateways, but they should also be delivered as a pre-integrated solution. This will ensure rapid installation and quick time to value.

5) Manage from a single source: Managing edge computing solutions as they become more widespread is no easy task. In order to do this successfully, operators need solutions that give them the ability to administer and manage multiple MDCs from all over the world as well as the associated IT and data center sensors from a single point of control.

6) Deploy with confidence: Finally, look for a solution that can be deployed around the globe with peace of mind. This means working with a vendor who can build, ship and support a micro MDC in nearly any geographical location that is required. Having flexible investment options is another item to consider in order to get started immediately while ensuring the right financial fit.

Experts at BI Intelligence estimate that 5.6 Bn IoT devices will use edge computing for data collection and processing in 2020. Now is the time to explore how to make the most of the data in today’s highly digital world.