Geospatial standards enable integration of geospatial services into IT infrastructures and solutions, reducing costs, adding flexibility and speed and making way for innovation in the process
Cloud computing brings together server farms (data centres), Software-as-aservice (SaaS), platform-as-a-service (PaaS), virtualisation, big data, mobile devices, apps and analytics. The use of international standards is the “glue” that allows the integration of all of these technologies. Virtually, every cloud transaction uses TCP/IP and almost every transaction uses the Web standard http. Many other open standard encodings and interfaces as well as proprietary encodings and interfaces provide the means by which services in the cloud can be requested and delivered.
Service-oriented architectures (SOA) evolved from client-server computing in the 1970s. Client-server applications utilise a network architecture in which the components can be geographically “distributed”. There are providers of services called servers and service requesters called clients. Typically, clients and servers run on separate computer systems that communicate over a network (as in cloud computing), but both the client and server may reside in the same system. A client initiates a session with a server, invoking a software programme, and the server runs the programme and returns a result.
With the advent of the Web, clients such as Mosaic and Netscape began offering access to ordinary users — via open standards — to applications (Web services) running on servers. Users don”t need to know or care where the servers are located. Web services are accessible through a simple http request. As long as users have Internet access, they have access to these applications. Open standards such as HTML, http, and URIs enabled this leap into the cloud.
As the US National Institute of Standards and Technology points out, cloud computing provides “convenient, on-demand access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction”.
The key driver for cloud computing is economies of scale. The modern day, very large data centres offer an advantage of at least six times over small data centres in terms of economies of scale. Cloud computing is not so much a technology as it is a business model designed to provide customers with the maximum amount of storage, transaction processing and analysis at minimal cost, while also relieving customers of expensive and time-consuming tasks related to purchasing and maintaining computing resources. And the bargain includes new capabilities and trends, such as a proliferation of mobile devices and apps that have access to virtually unlimited storage and processing. In addition to financial savings, cloud computing”s efficiencies translate into reduced carbon footprints and natural resource use.
Why put geospatial processing in the cloud
Two key aspects of cloud computing that are especially relevant to geospatial cloud computing are resource pooling and elasticity. In cloud computing, resources are pooled to serve multiple consumers with resources that are dynamically assigned and reassigned according to consumer demand. A consumer might, for example, work in a municipal planning office in a suburb of a large city. The city might have a GIS and many layers of data “in the cloud” and the planner has access to these through a Web browser. The city might host the GIS modules and data layers in its own data centre or it might host them with a company that owns and maintains data centres. The planning office doesn”t need to own the software and no one in the planning office needs to know the location or ownership of the servers providing the resources. Perhaps the city makes its services available to all of its suburbs, and the suburbs pool their (much reduced) GIS budgets to help pay for the shared resource. They reduce their costs and they are freed from maintenance and upgrade responsibilities. Together, they can afford to acquire new capabilities, perhaps apps that run on an unlimited number of mobile devices.
Elasticity means that the capabilities can be rapidly and elastically provisioned. For example, what if all the planning department workers in a region suddenly sat at their desks or picked up their notebook computers to perform the same task on the same data layer. Assuming the data centre has thousands of computers available and is perhaps able to spillover demand to a partner data center, the planning department workers will experience no slowness in response.
Geo-apps and the geo-cloud
Cloud plays an important role in innovation, because the time to market for ideas is much faster when new companies do not need to invest time and money in providing basic computing infrastructure. A profusion of Webbased map browsers, GPS-enabled applications, in-car navigation services, high resolution earth imaging systems and mobile smart phone location applications have tremendously expanded the geospatial market and the average person”s awareness of “maps in computers”. According to a recent survey of more than 400 mobile developers by IT industry market research firm Evans Data, 54 per cent of developers working on applications for mobile devices are including location-based and mapping services in their apps. This activity, along with other upward trends in the geospatial world, produces tremendous volumes of geospatial data, much of which resides economically in the cloud.
Standards add value in the cloud environment
Open Geospatial Consortium (OGC) has beenworking to develop standards that enable geospatial content sharing and integration of geospatial services into IT infrastructures and solutions. Some of the geospatial applications and platforms currently offered as cloud-based Web services are dependent on proprietary interfaces and encodings, but most of these applications and platforms also depend on open interfaces and encodings. Openness fosters innovation, expands markets and creates new opportunities and efficiencies for both providers and users.
Standards reduce value chain anxiety
Geoprocessing services hosted at the software-as-a-service (SaaS) level on the cloud can, if they implement OGC standards, enable other services to link to and thus “bind” to them, and vice versa. Cloud lends itself to such “service chaining”. Service chaining supports flexible creation of geospatial information value chains. It has been observed that customers like the idea that cloud computing lends itself to rapid configuration and reconfiguration of value chains in a “cloud ecosystem”.
Geospatial cloud providers will want their services and data to be useful to as many customers as possible, so standard interfaces and encodings are beneficial to service providers. Both customers and providers may want to be able to mix and match services of many kinds. Providers also understand that their customers want the freedom to avoid lock-in, that is, customers want to be able to choose services and switch from one to another on short notice. Standards enable multiple cloud providers, including niche providers, to work together to deliver value-added solutions. They enable cloud providers to more easily meet the varied needs of different customers and they give customers the flexibility to do business with a new cloud provider without excessive effort or cost.
Standards help location app developers
Developers of simple location apps can avoid headaches if they encode latitude/longitude pairs using the OGC GML Point Profile. Passing a coordinate pair is simple, but if communication with other geoprocessing services is planned, it is wise to use a standard. The importance of other standards for geoprocessing service communication becomes obvious from a quick study of the multiple technologies for representing spatial information — raster, vector, triangulated irregular networks, point clouds, computer-aided design Euclidean geometry and others. There are many different proprietary interfaces, encodings and formats used in the vendors” geographic information systems (GIS), location services, earth imaging, facilities management and navigation products. However, in the recent years, the industry has made dramatic progress in implementing OGC standards and related ISO standards.
GIS or not
Prior to the wide use of OGC”s geospatial interface and encoding standards, most geospatial information was confined within GIS and earth imaging systems, their specialised spatial databases and the networks of users who used software from the same vendor. Now, there is an exploding need for spatial data and simple kinds of spatial services, such as “Get a map”, “Where am I?”, “Where is the nearest pizza shop?”, and “Is it safe to dig here?” These services are now typically provided without a GIS, and they are well served by cloud services that can immediately scale to handle a large number of transaction. Through widespread implementation of standards, geospatial data and geospatial processing have become, to a considerable degree, just another part of the world”s information infrastructure.
What is needed now
The cloud is based on a standards framework for service-oriented architectures that provides for “publish availability of service, find service, bind to (invoke) service”. What is missing here are a few other steps, such as “Get data provenance”, “Agree”, “Authenticate”, and “Encrypt/Decrypt”. The growth of the market and social utility of geospatial technology depend on stakeholders working together to develop standard ways of communicating this kind of information.
The “Publish” activity needs to be able to provide metadata describing, for a particular service or data resource, details about authentication, authorisation, confidentiality, integrity, non-repudiation, protection and privacy. Servers need to be equipped to manage these issues, and owners of data and services need the tools and understanding necessary to configure controls based on the yet-to-be-developed standards.
The OGC Geo Rights Management Domain Working Group (GeoRM DWG) has produced the Geospatial Digital Rights Management Reference Model, which provides the foundation for “agreement management” that will bring us closer to full realisation of geospatial cloud applications. The mission of the GeoRM Working Group is to coordinate, and mature the development and validation of work being done on digital rights management for the geospatial community. The OGC Security Domain Working Group and the OGC GeoXACML Standards Working Group are forums for discussing topics and developing standards related to authentication, access control and secure communication. The OGC Workflow Domain Working Group addresses geospatial workflow issues, including security and licensing issues such as data encryption, authentication, and provenance tracking.
In the OGC, users of spatial technologies define interoperability requirements like those above and users work with providers of spatial technologies to turn those interoperability requirements into encoding and interface standards and also best practice documents that provide guidance in using the standards. Member organisations from the public and private sectors each assign one or more experts to participate in the testbeds, interoperability experiments and working groups that contribute to the development of standards.
From the standpoint of the individual member organisation and the world community, this is an extremely efficient way to reduce costs, develop new business and accomplish societal goals.