Cloud computing has revolutionized the ICT landscape and is a catalyst for digital transformation. But the adoption of Cloud GIS has been slow and GIS deployments on desktops and enterprise servers remain the norm. Will this change with the advent of big data analytics, IoT and AI?
The article below reviews the essentials of cloud computing and how they play out in the context of GIS. Illustrative references are made to ArcGIS Online, Esri’s SaaS solution, and cloud-based deployments of Esri’s ArcGIS platform. Still Cloud GIS is a vibrant and open market where traditional GIS companies and emerging niche players are growing the market for GIS with new and innovative solutions.
What is Cloud GIS?
Cloud GIS deployments are hosted in the cloud and embrace the essentials of Cloud Computing. There are competing definitions on what Cloud Computing entails, but one effective definition comes from the National Institute of Standards and Technology (NIST). In their definition the cloud model has five essential characteristics, three service models, and four deployment models.
Five Essential Characteristics
The five essential characteristics of the cloud model and how they are implemented in ArcGIS Online and other Cloud GIS solutions are listed in the following table.
| Cloud Model Essential Characteristic | Cloud GIS Implementation |
|
On-demand self-service – A consumer can provision additional computing capabilities as needed without requiring human interaction or permission from the service provider. This makes cloud solutions highly responsive to the evolving needs of an organization. |
In ArcGIS Online, any number of Named User subscriptions can be added at any time. Each subscription includes 500 Service Credits per year and these are used for storage, analytics and content services. More Service Credits can be bought at any time. |
| Broad network access – Capabilities are available over the network through an Internet connection by thin or thick client (e.g., mobile phones, tablets, laptops, and workstations). Hence cloud services can be accessed virtually anytime, anywhere, and with any device. |
ArcGIS Online is a platform for mapping and analysis which is accessible through a web browser making it available anytime, anywhere. The platform comes with specialized applications for field and office, and many of them run on mobile devices. |
|
Resource pooling – The provider’s computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to consumer demand. |
ArcGIS Online uses the cloud infrastructure of Microsoft Azure and Amazon Web Services. By default, resources are shared with other users, but dedicated storage can be purchased at a premium. |
|
Rapid elasticity – Cloud computing services can be scaled up and down, in some cases automatically, to match spikes in demand and reduce wastage. |
ArcGIS uses a Names User subscription model, but the computing services and resources are availed to match demand. The speed of computing is primarily constrained by the speed of the Internet connection and the capabilities of the client device. |
| Measured service – Cloud computing services use a pay for use model based on resource usage (e.g., storage, processing, bandwidth, active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency for both the provider and consumer of the utilized service. |
ArcGIS Online subscriptions are charged by the number of named users, and these are (re)assigned to people within the organization by an administrator. Service Credit consumption for the different services is available beforehand and actual usage over time, by user, and type of service, can be monitored by the administrator. |
Three Service Models
The three models for service delivery are Software as a Service (SaaS), Platform as a Service (PaaS) and Infrastructure as a Service (IaaS).
- IaaS – IaaS cloud services comprise storage, processing, and network. This market is growing steadily and dominated by Amazon Web Services and Microsoft Azure. ArcGIS Server instances can be run on both platforms using a per usage payment model and this offers flexibility and cost savings to both large and small organizations. Esri also uses Amazon Web Services and Microsoft Azure to host ArcGIS Online.
- PaaS – PaaS adds an application framework to the cloud solution stack that supports the deployment and development of applications. Esri’s ArcGIS platform for example can be accessed through a variety of APIs (e.g. ArcGIS REST API), App Builders (e.g. and SDKs (e.g. ArcGIS Runtime SDK for .NET). Other GIS niche players providing access to JavaScript libraries through APIs are Mapbox, OpenLayer and Leaflet.
- SaaS – SaaS acts like plug and play and offers immediate value benefits once you are signed up and connected. As an end-user you probably need to upload your data and configure applications according to your specific needs, but even this can be managed by a service provider. The market for Cloud GIS as a SaaS offering is dynamic and combines offerings of established GIS software vendors like Esri (ArcGIS Online), Hexagon (Smart M.Apps), and PItney Bowes (MapInfo Stratus) with niche players like Mango, CARTO and Cloud GIS.
Four Deployment Models
According to the NIST definition, cloud computing can be deployed as Private, Community, Public and Hybrid cloud.
- Public cloud – a public cloud is publicly accessible through a Service Level Agreement, but the cloud infrastructure is owned by a third-party cloud service provider like Amazon Web Services or Microsoft Azure. PaaS and SaaS cloud service providers may act as intermediary between consumers and IaaS cloud service providers and they can opt to do so in a transparent or opaque manner.
- Community cloud – a community cloud is very similar to a public cloud, but access to the cloud is restricted to a group of companies that have a common business interest. An example is the Salesforce Community Cloud which is restricted to Salesforce employees, partners, and customers.
- Private cloud – the cloud infrastructure is owned by a single organization or business entity. Most often the infrastructure is on-premise, but it can also be kept at an off-premise data center or an overseas location.
- Hybrid cloud – a hybrid cloud implementation is a composition of public/community cloud and private cloud. The two can be coupled to permit cloud bursting which pushes services to the public cloud during high demand. The two may also run as disparate instances, which are used for the sharing of private and public information respectively.
Future of Cloud GIS
Cloud GIS is seen by many at the future of GIS, since it democratizes GIS services through an on-demand service model and per usage payment model. The booming cloud market has set the utmost standards for service delivery and cloud service providers go out of their way to maintain high levels of security and trust with their customer.
Despite this, traditional paradigms and deployments of GIS remain entrenched as common practice. GIS experts acting like gatekeepers view universal access to GIS services in the cloud as a threat to their existence. Organizations still pride themselves in acquiring GIS labs of brick and mortar and some lock the door to keep the lab in its pristine state.
Many of today’s GIS experts, including myself, have been trained in the Esri school of GIS and carefully watched the keynote presentations of Jack Dangermond to stay abreast with the latest GIS technology. Nonetheless Google Maps and open source GIS began challenging Esri’s position of GIS leadership founded on proprietary and complicated software. Interestingly Jack himself is now an advocate for what he calls the Geospatial Cloud. This is what he says about the combined forces of computing power and the cloud in an article for Forbes.
As a software executive, I have never been more energized by the growth and impact these forces are having on my field, geographic information systems (GIS). And, my sense is that it’s just beginning.
Promoting the Benefits
Reading through the lines above you will discover many benefits of Cloud GIS. Let’s recap the important ones from a user perspective.
- Reduced investment – Cloud GIS drastically reduces the upfront investment cost, since there is no need to buy servers, storage solutions, and permanent software licenses. Neither does one have to invest in a server room with network infrastructure, air-conditioning and power backup. Some of these savings can be invested in high-spec laptops or desktops, good internet connectivity and handheld GPS for field data collection.
- Quick payback – A cloud GIS instance can be up and running in a short time and yields immediate organizational benefits with ubiquitous access to relevant data and information services that could reduce cost and increase revenue. Cost savings and new revenue streams would ideally be plowed back to widen and deepen the deployment of Cloud GIS in the organization.
- Operational cost – Cloud GIS is virtually maintenance free since the cloud infrastructure is maintained by the cloud service provider. Software upgrades and incompatibility arising from use of different software versions will no longer present a cost, since the cloud is always updated to the latest version. Part of these savings should be directed to the administration of the cloud instance to ensure that GIS services are well organized, kept up and running and actively maintained.
- Service availability – Cloud services are available anytime, anywhere, from any device and this offers a tremendous benefit to the users who can now access critical GIS services in the field, the office, or at home without having to worry about the device that they are using. Most cloud service providers guarantee uptime of 99.9%, a service level that is difficult to attain with the on-premise deployment of GIS.
- Service accessibility – Accessibility takes the concept of availability a step further and makes it easier for everyone in an organization to connect to relevant GIS services through focused maps and apps. For one all GIS services are hosted in the cloud and available through a portal, so data and information services can easily be found. In addition, cloud GIS services are easily embedded in focused and easy to use end-user applications.
- Enhanced analytics – Cloud infrastructure is massively scalable, so cloud GIS facilitates big data analytics on huge repositories of satellite data, multidimensional data cubes and live streams of IoT data. Furthermore, cloud GIS workflows tend to be intuitive and wizard-driven, so people in the organization will feel empowered to start using cloud GIS as a self-service tool.
Dealing with Objections
Cloud computing has met fierce opposition, particularly in organizations or cultures that are resistant to change. Let us find out how the most common objections can be addressed.
- Security – Occasionally we get to hear that massive amounts of data is lost or stolen from the cloud. Such security breaches attract media attention, but they are getting rarer. Cloud service providers now use very high security standards due to fierce competition and stringent legislation. So, the risk of security breaches is higher for on-premise deployment.
- Internet – Many organizations complain about poor Internet connectivity making it impossible to connect to cloud GIS services. Nowadays most places have good Internet coverage and poor connectivity often results from poor investment decisions and pursuit of other goals and priorities.
- Annual fee – Organizations working with annual budgets argue that they might not be able to renew their annual cloud subscription when faced with budget cuts. But technology changes fast and procured assets may become obsolete or unproductive in a period of 1 to 3 years. This is particularly true when there is no budget for repair and maintenance.
- Data – It has been argued that upload and download of massive amount of data is an impediment to cloud GIS, particularly when Internet connectivity is poor. The truth is that geospatial is increasingly maintained and accessed from online stores. The satellite imagery that you procured 2 years ago is likely to be outdated and no longer useful.
- Control – Cloud architectures are like a black box and integration with legacy applications and other cloud platforms can be cumbersome. This is true but the days of extensive customizations that are difficult to maintain are over. Instead configure the out of the box cloud applications or build your own which has never been easier.
Wrapping Up
Cloud GIS exhibits the essential characteristics, service models, and deployment models of cloud computing and presents a compelling value proposition to old and new GIS users alike. Cloud GIS is often associated with SaaS offerings like ArcGIS Online, but PaaS offerings consisting of APIs and SDKs present exciting opportunities to GIS developers.
The benefits of Cloud GIS that were identified from a user perspective include reduced investment cost, quick payback period, low operational costs, increased service availability and accessibility, and enhanced analytics. Common objections to the use of Cloud GIS are high security risk, low internet speeds, need for annual payments, download/upload constraints, and lack of control, but none of these appear substantial.
You have just learned the characteristics of Cloud GIS and the benefits that if offers. Has this influenced your view towards Cloud GIS and has it motivated you to advocate for its deployment in your organization? Feel welcome to share your views and opinions and please contact us for further help and information.