The Big Picture: Capturing the Cloud
February 25, 2011
While many surveying and mapping professionals are trying to understand what the cloud is and how to best use it to their advantage, entrepreneurs at Aerial Services Inc. are creating their own. Their vision? To become an integral part of the new paradigm in geospatial data sourcing.
It’s the kind of scenario that keeps many business owners awake at night--a new technology is introduced that threatens to pull the rug out from under everything you’ve spent the last few decades carefully building. For Mike Tully, CP, GISP, now president and CEO of Aerial Services Inc. in Cedar Falls, Iowa, the technology was digital aerial cameras. The year was 2000, and the new owners of Aerial Services, led by Tully, were not willing to sit idly by and watch the 33-year-old geospatial services firm dwindle into oblivion as large corporations acquired the latest technology and gobbled up market share.
“We could see that the business model for the industry was going to change,” Tully says. “Large firms with multimillion-dollar airborne platforms and specialized expertise in acquiring LiDAR or SAR [synthetic aperture radar] imagery would start collecting much larger areas of data. Although we were determined to keep our technology current, we knew it would be strategic to develop an alternative source of revenue.”
As he considered the coming changes, Tully realized that the drive to acquire ever-larger datasets would also create a new market for firms that could interpret and add value to that data. At the same time, he began carefully eyeing developments in Web 2.0 and the move toward interactive Web applications. The launch of Amazon Web Services (AWS) in 2002 sparked Tully’s imagination. He could envision a time when computing would become a utility, and companies would buy computing power, storage and bandwidth from a “utility company” rather than depending on their own internal resources--not unlike the move to electric power 100 years ago.
Under Tully’s direction, Aerial Services began planning for this new paradigm. Meanwhile, the company continued investing in its aerial acquisition technology. In November 2008, the firm became the first U.S. geospatial firm to purchase Leica Geosystems’ third-generation state-of-the-art line sensor technology, the ADS80 Airborne Digital Sensor, which was capable of orthophotos and 3D feature mapping. The new system allowed Aerial Services to compile an increasingly large library of geospatial data.
Then, in November 2010, after extensive beta testing, Aerial Services made the leap to the cloud with the introduction of SpatialCloud.com. “Because we’re a geospatial company, we understand and work with heavy geospatial data--imagery and LiDAR, etc.--every day,” Tully says. “We know it takes massive server farms and storage arrays to handle the data. So we wanted to create a company that would help free other companies and other users from having to manage that heavy geospatial data.”
The new company’s goal was simple: Provide a platform that data providers and application developers could use to host, distribute and resell heavy geospatial data. The pay-as-you-go service would assist developers looking for enhanced, unrestricted image layers to use in their own applications as well as image content providers looking to outsource the distribution of their data. “We’re essentially fostering the creation of a very rich, deep library of geospatial data. But it’s even more than that,” explains Mark Korver, chief technology officer of SpatialCloud and a lead visionary for the firm. “It’s also a platform for building and providing services using geospatial data from a variety of sources and combining those, adding value to them, and delivering them to somebody’s desktop when they need them and in the exact format they need.”
Other companies had already begun offering similar services, but Tully believed SpatialCloud had found a niche. After all, the firm was starting with its own imagery--imagery that was current, high-resolution and positionally accurate. The firm could offer flexible, inexpensive subscriptions through the AWS platform. And the move just made sense from a business standpoint. “Our traditional model is that a client hires us to acquire an area; we then sell them back that data for a large sum once, and the client owns the data,” Tully says. “But that’s changing. We are beginning to acquire large areas of imagery and retain ownership of it so that we can resell it multiple times at a much lower rate. This benefits the client by decreasing their costs and creates new sales channels for the data provider. Through cloud technology, we can expose all that data to a world of clients. And we hope to make more revenue with that model than by selling the data once to a single client.”
Tully acknowledges that there may be cases where clients insist on having complete ownership of the data. But there, too, the cloud offers advantages compared to traditional data storage models. By leveraging security systems and using authentication keys, data on the cloud can be hidden from public view but made easily accessible on-demand to the individuals who are authorized to use it. The result is more-transparent operation and faster, more-informed decision-making within organizations. Streams of the data can later be easily shared with other groups if desired.
“What I find most exciting about cloud technology is the scalability--the fact that it no longer matters how many people are trying to access the data,” Tully says. “We’re now using enterprise-class infrastructure, which is something that I could never build on my own. I could never bring a huge Internet pipe into our facility and build a massive storage array and server farm to support 1,000 or 100,000 or 1 million users. But Amazon can--they do it every day. And I can get that very inexpensively and pass it on to clients for their geospatial data needs.”
SpatialCloud provides global landsat and a seamless high-resolution (1 meter GSD) color dataset covering the United States as a tiled mapping service compliant with the Open Geospatial Consortium (OGC) Web Map Tile Service (WMTS) specification as a demonstration that the platform is capable of hosting a dataset of any size to any number of simultaneous users and achieve enterprise-class performance and scalability. According to Tully, the service opens a source of imagery for a vast part of the market that has never had imagery. “Web application developers can build businesses using that imagery,” he says. “As a result, we’ll see all kinds of new applications as Web services that are mashed up with other Web services and algorithms hosted on the cloud. And that should drive the demand for more data and higher-quality data.”
A number of hurdles still remain to widespread adoption of the cloud, not the least of which is a lack of understanding about what the cloud is and the benefits it can provide. Additionally, many corporations have a lot invested in their data “silos” and storage infrastructure and are reluctant to shift to a different type of platform. But Tully believes it’s only a matter of time before these barriers are eliminated. “Economics will be the primary driving factor,” he says. “As the data infrastructure within organizations needs to be replaced or upgraded, we’ll see a very rapid adoption of the cloud because no one will want to put another $100,000 or $1 million into their infrastructure knowing that they can get it for 15 cents per gigabyte per month from a Web service and never worry about its maintenance or replacement.”
Without that in-house infrastructure, a paradigm shift will occur. Software-as-a-service will become standard, with users only paying for what they need when they need it. Expensive desktop applications will become less important. Rich layers of geospatial data and imagery will be accessible on demand, and a host of new applications will emerge. “One of the most difficult aspects of the cloud is that it’s so hard to get your mind around all the possibilities,” Tully says.
He likes the electricity analogy. “When I walk into a room and flip the switch on the wall, I expect the light to come on; I don’t have to think about where the electricity is coming from,” he says. “We’re moving in the same direction for computing power and imagery. When I start the application, the imagery is there and I use it. I don’t care who distributes it and who owns it--I just know it’s there when I need it and it does not cost much at all.”