Home Articles A more Elastic and Automatic Cloud

A more Elastic and Automatic Cloud

Sandeep Singhal, Director, Cloud Storage, Google

The third generation Cloud is allowing developers to focus on the value they are going to create, rather than thinking about the infrastructure they need, says Sandeep Singhal, Director, Cloud Storage, Google

How is location creating new business opportunities for organizations?

Many companies today are using geospatial information to provide services and customize their capabilities. But they don’t really consider themselves to be location-based companies; they just take location naturally. They see location as just another tool in their box of programming techniques. Location is a very natural thing. It’s like breathing air. They don’t see geoinformation as being special or unique; it’s just a data source that they can take advantage of. But, location is creating a variety of new opportunities. For example, CloudEngage, a start-up based out Portland, Oregon, US, is using location as a way to support targeted marketing on the Web. What they have discovered is that using location allows you to take your existing website experience and improve monetization by around 20-30%.

Would you say that the geospatial industry is becoming more Cloud-focused?

It is one of the big trends today. More and more companies are growing up in the Cloud: relying on the Cloud to host, manage and manipulate the data, to apply machine learning, and then to publish the results back out to the customers. Today, companies start with the Cloud and don’t even think of having an on-premise infrastructure; they rely on the Cloud as their backbone.

What are the key drivers boosting the global Cloud-storage market?

Traditional Cloud infrastructure lets you host the apps and deploy your applications, while you manage the entire infrastructure. We are starting to see a shift from that model to a much more elastic and automatic Cloud: a Cloud where you upload data and you have access to infinite storage; where you can deploy your algorithms and your analytics pipeline will automatically grow based on the capacity that you need. You don’t have to go in and manage all of that capacity; it happens automatically. So, what we are seeing with this third generation Cloud is the ability for application developers to really focus on the value that they are going to create rather than thinking about the infrastructure they need.

What are the key benefits of third generation Cloud technology?

The biggest benefit that the third generation Cloud technology offers today is that you don’t have to worry about capacity. You can deploy your data in the Cloud, set up your application in the Cloud, and if you need more capacity, you will get it. You don’t have to plan for it, deploy it or configure it. So, the idea of a Cloud which is truly elastic, which simplifies your deployment, planning, processing, management and support, and creates an automatic IT environment, is a huge win. It allows you to really focus on the places where you are creating value and eliminate the stages where you are really not creating any value.

How are Cloud and Internet of Things (IoT) impacting the geospatial industry and vice versa?

IoT is about collecting large volumes of data from lots of different sources. And the Cloud is a place where you can receive that information, process and analyze it, learn from it and deliver it to the users. IoT and Cloud go hand-in-hand while creating a backbone on which the geospatial industry can build solutions, services and new capabilities. As the geospatial industry becomes more and more solution-centric, it is able to take advantage of all of this new data and processing capabilities to really focus on the differentiated set of services customers really want.

Are businesses today rapidly adopting Cloud or do they still have their heads buried in the sand?

Companies are starting to understand what it means to be in the Cloud and they are beginning to adopt it, but their levels of adoption vary. The US and Europe have been the fastest to adopt the Cloud, but as an industry, we are still in a very nascent stage. Only a small percentage of companies have really moved to the Cloud. So, a lot of opportunity in still up for grabs in the Cloud space.

One of the big challenges is that the Cloud means different things to different companies. You could dip your toe in the water and take your existing processes and simply extend them with Cloud-based publishing or Cloud-based archival. Or you could decide to do something much more fundamental. You could do a real re-architecting of your business where you say, ‘Well, what would happen if I had no legacy, if I had no on-premise infrastructure? Can I build in the Cloud from the start?’

People understand the Cloud or have heard of the Cloud, but the actual move to executing a migration to the Cloud will take time. Companies today are at different levels of that evolution and it will take time for each company to develop its own strategy. There are many different strategies and paths available. Some of them are riskier, but may, in fact, be more rewarding than others.

Are regulations and security concerns hampering the uptake of the Cloud?

Security is certainly an issue. Customers are nervous about moving their data into a shared environment. They take time to understand the privacy and security constraints, comprehend what the Cloud providers can give them and figure out the security model. One of the things time-consuming processes is to understand and develop a security policy working closely with the Cloud provider, and make sure that the provider meets their security standard.

There’s a big opportunity in connecting public and private Clouds and making it possible for an application to move seamlessly between them

Then there are some customers who will obtain a dedicated Cloud infrastructure for their needs in order to get the most privacy possible. Meanwhile, some customers will continue to run their own infrastructure on-premise simply because they want that level of control. But, I think what all the Cloud providers have started to develop is a fairly sophisticated security privacy policy based on the access capabilities that enterprises need.

How do you see private and public Cloud platforms competing with each other?

The big opportunity is not in competition; it is in the hybrid Cloud. It is about connecting these Clouds together and making it possible for an application to move seamlessly between public and private Clouds. There is a whole generation of storage appliances now in the market which blur the line between on-premise storage and Cloud storage. They basically treat Cloud storage as the third tier of local caches and on-premise storage. So, they will automatically move cold data — data that is not being accessed — into the Cloud as kind of their archival vault storage, and then pull it out of the Cloud and bring it on-premise when it is needed. We are starting to see these kinds of solutions that really bridge public-private hosted infrastructure in a very seamless and natural way.

What is it that you see beyond the Cloud?

I think the biggest thing on the horizon is the shift toward rich data analytics and rich machine learning. We are starting to see companies, including Google, provide richer and richer services for the data, once it is actually in the Cloud. For example, we recently introduced a Cloud machine learning API that automatically does image analysis, feature extraction, text recognition and even sentiment analysis on people who are in your images. So, you start seeing these services that used to require deep knowledge, deep science, and specialized expertise become democratized — made widely available so that any application can take advantage of technology that used to be basically kept only for the elite.

IoT and Cloud go hand-in-hand while creating a backbone on which the geospatial industry can build solutions, services and new capabilities

Do you see all these technologies and Cloud finally riding on each other and moving toward the Internet of Things?

I think it is dangerous to think about any of the technologies like location, mobile, social media, earth observation, etc., as really being separate. Location is information about lots of spatial data. IoT is about collecting large amounts of location data. Geospatial is about analyzing all of this data and producing meaningful results. And Cloud is really a substratum for receiving the data, running the analysis and publishing results. So, these are not really disjointed technologies. They really build on each other. And I think any modern solution relies on a combination of all of these different technologies integrating and working together to create new and unique experiences.