Keys To Integrating Applications and Data
APIs, standards, knowing the business objectives, providing an excellent UX, and testing securely.
Join the DZone community and get the full member experience.Join For Free
To gather insights on the state of application and data integration, we spoke with 18 executives from 15 companies who are involved in the integration of applications and data.
Here’s who we talked to:
Shawn Ryan, V.P. Marketing Digital as a Service, Axway | Kurt Collins, Director of Technology Evangelism and Partnership, Built.io | Thomas Hooker, V.P. of Marketing, CollabNet | Piyush Mehta, CEO, Data Dynamics | Daniel Graves, VP of Product Management, Delphix | Samer Fallouh, V.P. of Engineering, and Andrew Turner, Senior Solutions Engineer, Dialexa | Andrew Leigh, V.P. of Marketing and Alliances, Jitterbit | Trevor Hellebuyck, CTO, Metalogix | Mike Stowe, Developer Relations Manager, MuleSoft | Zeev Avidan, V.P. Product Management, Open Legacy | Sean Bowen, CEO, Gordon McKinney, Senior Solution Architect, Ross Garrett, Product Marketing, Push Technology | Joan Wrabetz, CTO, Quali | Razi Sharir, V.P. of Products, Robin Systems | Girish Pancha, CEO, StreamSets | Bob Brodie, CTO, SUMOHeavy |
And here’s what they told us when we asked them, "What are the keys to integrating applications and data?"
It’s all about the RESTful API. It has high top-of-mind awareness. RESTful APIs are key for modern integration. Service-oriented architecture owners are not sure what to replace with a native API approach. More education is needed on how to have an API strategy.
Leveraging standards. Five to six years ago integration required leveraging proprietary APIs. Microsoft has embraced open standards. You must understand the limitations, scalability, and throughput of APIs. Customers care about speed, real-time replication, and synchronization of data.
- Ensure you use one data format to connect everything – JSON. It’s proving to be what XML wanted to be. Come up with a new kind of standard for exchanging data with REST-based APIs. REST provides the flexibility to structure data without corrupting it. Real-time needs to become the standard. Now it’s a “nice to have.” The new version of HTTP will drive us closer to real-time becoming the standard.
- Ability to isolate components from automated computing, through the network, to storage with the ability to snapshot and clone.
- Understand the landscape of options. While it’s easy to set up and collect data, it can get very expensive very quickly. Segment.io is an awesome product with a landscape of options. Omniture was missing 33% of a client’s orders until we built a back end to integrate with Omniture. Put one data tag to Data IO. It’s easier to know the landscape of options as a consultant rather than if you are competing to provide the same services. You get a clearer picture of what everyone has to offer and you can objectively evaluate each one.
- It starts with a clear plan based on the objectives of the business. This requires an up-front dialogue with the business owner(s) followed by the team, the processes, and then the tools. The tools must be able to help the team achieve the business objectives. Know what data is pertinent and relevant across the tool chain. What’s the collaboration layer that brings everything together, integrate, test, and build the package. Making people part of the culture chain is key. 74% of organizations now involve business people in IT. They want innovation with data quickly. Fix fast, test, and deploy and then know how many times it took to get it right. They want a closed-loop process with data shown in a dashboard. Everything needs to be able to be customized for the customer.
- Integration needs to be the responsibility of everyone. Connectivity is more important than anything else. It’s more important to connect than what you choose to connect with. You must provide a good digital, connected experience for customers, partners, and employees. There’s not one single solution that’s right for every situation. You need to be able to connect and orchestrate multiple applications like Uber did with GPS, CRM, automatic billing, scheduling, mobile connections to redefine an industry.
- 1) Define what the integration contract will look like by documenting the API and how it works against the specifications as it’s being developed. People typically don’t test code against documentation, especially for APIs. Documentation comes first then build the code to the documentation. This results in a more stable and predictable API. 2) Consistently predict the data structure and response codes so everything is predictable and results in better built APIs.
- Think of application data integration as a full lifecycle operation. Runtime has sufficient operational control as you gain a better understanding of how to upgrade and evolve processes with agility. This makes developers more productive with runtime controls and agility.
- It’s a challenge since the application owner cannot see the infrastructure of the data storage. We provide the ability to create storage silos and manage them. Users cannot see inside third-party cloud hosting solutions like Azure and AWS. They’re providing a utility. Google provides analytics and a data mining platform so they’re further ahead in providing customer insight into their data.
- There's a lot going on in mobile with tactical differences in understanding how data moves inside and outside the organization. Exposure to the outside increases the need for security. First generation integration is very tactical, not following best practices. Look to the broad layer to decouple the frontend and backend. APIs become defacto standards but one size does not fit all.
- The integration space is evolving. 1) The technology of integration to REST, JSON, and API. 2) The fast pace of integration of bi-model systems. APIs enable companies to move fast.
- Deploying full-stack in the sandbox – developing, testing, and staging. Telcos are testing applications to deploy that interacts with their billing system. In the sandbox, clients can get access to real things in the network. There’s a push towards virtualizing data and services to put into sandboxes. More automation is the challenge. Telcos create fake records to see property updates but that's still very risky. You can create virtual copies with the same APIs.
- The key to integrating applications and data is building a seamless application network. An application network is a network of applications, data and devices connected with APIs to make them pluggable and create reusable services. By wrapping applications and data as APIs, any digital asset can be discovered and reused by consumers on the network. Approaching integration in this way enables organizations to leverage existing and new technologies to drive innovation and agility at scale, launch new products and revenue opportunities, and improve customer experiences.
- Ability to gather data without disrupting ongoing operations. Need to gather data from heterogeneous data sources. Need to synchronize data to the same point in time for developing/testing composite or federated applications. Streamlining the process of provisioning dev/test environments and doing so in a secure manner. Ability to provision full-stack environments (the applications and their associated data) to developers, testers and analysts.
What are your keys to integrating data and applications that we haven't covered here?
Opinions expressed by DZone contributors are their own.