Advanced Computing in the Age of AI | Thursday, April 18, 2024

At the ‘Cloudification’ Tipping Point, Application Integration Matters More Than Ever 

As Internet-connected, smart devices increasingly meld into all aspects of daily life, people are starting to expect the same connectedness and speed in everything that they do. Companies are discovering that response times that were adequate a few years ago now trigger annoyance and frustration, threatening customer loyalty. To meet today’s demands you have to speed up transactions and business processes and make virtually every product or service you offer available over the Internet.

This market imperative, together with the ease of adopting cloud-based applications, has created a tipping point. Businesses are embracing “cloudification”: shifting apps to cloud-based systems unless there’s a compelling reason to keep them running on-premises. They are migrating their applications from local installations to cloud platforms, or replacing them with cloud-based equivalents, such as Salesforce or Splunk. IDC predicts that, “over the 2013 to 2017 forecast period, public IT cloud services will have a compound annual growth rate [CAGR] of 23.5 percent, five times that of the IT industry as a whole and PaaS will lead IaaS and SaaS with a CAGR of 29.7 percent.” This phenomenon applies both to traditional and advanced scale applications handling, for example, real-time simulation, modeling and data analytics based on extreme scale data sets in financial services, manufacturing and other verticals.

SnapLogic's Darren Cunningham

SnapLogic's Darren Cunningham

Some traditional IT organizations are eager to move apps to the cloud. Some younger companies have only used cloud-based apps and nothing else. And those who haven’t yet embraced a cloud solution are likely to find themselves with no choice but to move apps to the cloud, in order to take advantage of the same speed, simplicity, and flexibility that their competitors are enjoying.

The cloud does indeed eliminate many of the costs and concerns of on-premises IT, but the challenge of enterprise data and application integration persists. In fact, cloudification can make solving that challenge more urgent than ever. The right approach to integration is the key to unlocking the transformative potential of your cloud investments.

Integration in the era of cloudification

With all of the well documented benefits of adopting the cloud it's easy to see why you might think you don’t have to worry about integration. But there you’d be wrong. Despite its many advantages, cloudification is not appropriate for everything. About one third of businesses run major mission-critical applications that are more than 10 years old, and one in five have at least one major application that’s largely gone unchanged for 15 years or more, according to a recent survey. These legacy systems can’t readily meet the connectivity and integration demands of the mobile, cloud, and other web applications that a company may be using. They can’t simply be moved to a cloud platform and they can’t just be written off, at least not yet, because they contain too much IP in the form of data and business processes. In addition, because of regulatory or other security requirements, some systems still need to be based on premises.

Organizations need a strategy for determining which apps will gain the most from a move to the cloud. Some will decide it’s not yet worthwhile to carry out a wholesale replacement of all legacy systems, so integrating them with newer, cloud-based apps is the cost-effective option.

Traditional integration approaches can’t keep up

Integration isn’t just a short-term challenge that will disappear when you move all your apps to the cloud. If you move siloed on-premises apps to cloud-based versions, you’ll create siloed SaaS, also called SaaS sprawl. SaaS sprawl may be even worse than traditional app sprawl, because SaaS makes it easy for business users to slip the apps under IT’s radar. Even when IT knows about all the apps that are in use, it will likely still struggle to integrate new apps with existing cloud-based or on-premises solutions. All the problems of traditional app sprawl pertain: unintegrated data and processes reduce your business’ efficiency and its ability to gain a current, 360 degree view of customers, suppliers, and partners.

In fact, cloudification dramatically increases the challenges of integration — and creates an expectation that integration should happen almost immediately. A business unit manager can buy an app online in an instant with a credit card, but it might take days, weeks or months to get the right data into the new app, or to get data gathered by the new app into your other corporate systems. This can create serious disconnects.

The old approach to integration would have been to create brittle point-to-point connections and cobble together component solutions for apps and data, but these old on-premises tools can’t keep up with the world of social, mobile, analytics, cloud and the Internet of Things (SMACT): they were not designed to handle an IT ecosystem that expands beyond traditional confines of the firewall.

To leverage the elasticity and efficiency of the cloud, and to operate at the peak performance levels that your customers expect and demand, you’ll need to take a different approach to solving integration. A cloud iPaaS provides a modern and scalable architecture that seamlessly handles cloud constructs of JSON, REST, and APIs. It’s also easy for less technical users to use. In the past, integration tools required a “black belt” in the technology and a lot of code. Modern approaches are cloud-based, with the flexibility to perform integrations either in the cloud, on-premises, or even natively in a Spark or Hadoop cluster. If you want to retain some of your data on-premises, you can fully cloudify your technology and still keep your ties to hardware when necessary.

Adobe recently adopted an iPaaS approach to integration. When the company adopted a “cloud first” strategy for its internal business applications, it had to slash integration time for new applications from weeks to days. To meet this requirement, the company decided to provide a common hybrid integration platform that not only cut time and costs, but provided self-service capabilities, allowing its nearly 400 non-specialists to deliver more than 600 integration interfaces and more than 200 APIs. This would have taken years using the company’s previous, more traditional approach.

Conclusion:

Businesses that haven’t yet adopted cloud-based IT are beginning to feel the pressure: to keep up with the speed of their markets, they’ll probably need to cloudify at least some of their systems. You might not yet need a fully cloud-based IT infrastructure -- maybe you can get by with a mix of systems -- but every business, no matter what its mix of cloud-based or on-premises systems, needs seamless integration and a level of self-service capabilities. By delivering that integration today, iPaaS enables businesses to move to cloudification in their own way and at their own pace, while eliminating the damage of disconnects between systems.

Darren Cunningham, VP of Marketing at SnapLogic, which provides Integration Platform as a Service (iPaaS) tools for connecting cloud data sources, SaaS applications and on-premises business software applications.

EnterpriseAI