Edge Compute is the Future in Government, if Agencies Can Securely Get There
These days, government agencies are trying to modernize their infrastructure and improve upon the digital delivery of key services. At the same time, most agencies are employing a remote workforce, and billions of new internet of things (IoT) sensors and portable devices also need access to core networks. In this new environment, pushing as much processing power out to where it’s needed will be the key to managing all of that new data and traffic.
This is Where Edge Computing Can Save the Day in Government
Edge computing is defined as bringing computation and data storage closer to the devices where the data is being collected, which is often at the very edge of a network. This means rather than having a server thousands of miles away from the users accepting and processing their data, agencies instead develop the means to add processing power at the network edge. After processed, only a small fraction of the raw data might get sent back to the main servers, saving time, bandwidth, money and processing power.
Getting to the edge can be extremely rewarding, but it can be tricky to deploy. Three experts on edge computing gave their advice on a recent FedInsider webinar about the best ways that agencies can move forward.
Featured Experts:
Gerald Caron
Director of Enterprise Network Management,
Bureau of Information Resources,
U.S Department of State
Frank Indiviglio
Deputy Director, High Performance Computing & Communications, NOAA
Dave Young
Senior Vice President, Public Sector & Global Hyperscaler Business, Lumen
1. Modernize the Current Infrastructure
As with any major technology rollout, there is going to be a lot of work that needs to be done in preparing systems. Edge computing is no exception to this rule. For edge computing to work best, agencies should start by modernizing their current infrastructure. This often means moving as many applications to the cloud as possible, since that will act as a conduit to support edge computing in the future.
“Our own modernization continues as we move to the cloud,” said Gerald Caron, Director of Enterprise Network Management at the Bureau of Information Resources with the Department of State. “We are moving most applications to the cloud, which is another way to take advantage of edge computing.”
According to Caron, by moving applications and systems to the cloud, agencies should have fewer issues setting up the needed infrastructure to employ edge computing. Agencies should pay particular attention to legacy systems and aging infrastructure that could become a hinderance or bottleneck to edge computing programs.
2. Optimize Edge Devices to Work in a Fully Edge Computing Environment
Most agencies process a lot of data, but that is taken to the extreme at the National Oceanic and Atmospheric Administration (NOAA), which is responsible for everything from weather forecasting to fisheries management. The agency deploys millions of sensors all around the world, from the driest deserts to the bottom of the ocean floor. Just putting together a typical weather forecast requires that NOAA process over six billion readings from its sensor network.
“The weather data is collected across the world through our satellite radars and other sensors,” said Frank Indiviglio, Deputy Director of High Performance Computing and Communications for NOAA. “We’re looking to modernize those edge assets to be smarter for forecasters by getting data from those assets to them in a much more rapid fashion.”
In the past, most sensors collected data and then sent everything back to central locations for NOAA to process. But Indiviglio says the agency is improving operations by upgrading and optimizing edge computing devices to process much of that data locally, and maybe only send a key fraction of it back for further processing. Once NOAA has fully optimized its edge computing devices, it will mean faster and more accurate forecasts and better models for predictions.
3. There Is No Need to Reinvent the Wheel for Edge Computing at Agencies
Edge computing is a fairly new concept for many government agencies, but companies and organizations in the private sector have been experimenting and deploying the technology for at least a few years. As such, there are many good case studies as well as contractors who have had a lot of success in deploying edge computing. This includes adding cutting edge technologies like artificial intelligence and machine learning to enhance edge networks. And because the government is already very experienced with using some of those key technologies, they can easily build on that success to improve their edge computing initiatives.
“Sometimes, it’s easier to see in the commercial world, but programs like artificial intelligence, machine learning and robotics are all things that our government uses and incorporates,” said Dave Young, Senior Vice President of Public Sector and Global Hyperscaler Business at Lumen. “That’s going to be the foundation of edge as applications continue to evolve around these areas.”
The government’s success with innovative technologies like AI will help drive edge computing initiatives, Young said. And they can quickly employ them without having to start from scratch.
4. Manage Data Movement When Edge Computing Is Needed in Non-Ideal Locations
Edge computing, by definition, is centered around getting data processing closer to where it’s needed the most. So what happens when the data is needed in non-ideal places where bandwidth might be extremely limited? That is where agencies will have to practice smart data management to ensure that information is processed as efficiently as possible, with only the least amount of data possible being transmitted back to a central location.
“Data movement has been a hurdle and continues to be a hurdle for us,” Indiviglio said. “We tend to produce a lot of data and collect a lot of data, and then move that data back to our computing infrastructure.”
Indiviglio said that NOAA was dealing with that problem by adding processing power and storage capabilities to many of its remote sensors to support edge computing. The key is to balance those upgrades so that the sensor or device has just enough processing power and storage to do what it needs without putting a drain on sparse local resources like the availability of a reliable power supply. The goal is to have the sensors process data at the edge, and then send back their results over whatever limited bandwidth channels are available. That will in turn optimize NOAA’s edge computing network for both today and far into the future.
Getting Closer to The Edge
Edge computing is a highly complex topic where a lot is happening very quickly. As discussed in this webinar, federal agencies with real challenges are already using edge solutions to transform their operations, and are realizing measurable benefits because of that. But in many cases, this is just the tip of the iceberg.
To learn more about what key factors are motivating agency leaders to shift workloads to the edge, and to get tips about how to best implement your own edge computing programs, download the IDC Infobrief on this topic, or visit the Lumen public sector edge computing page.
