Additional Considerations Regarding Application and Data Integration

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 have I failed to ask you that you think we need to consider with regard to application and data integration?" 

• What are the challenges around data sovereignty in an open world? Data integration can be affected by data sovereignty requirements.
• Emphasize that integration is changing from a data exchange to integration with systems that perform like IoT (edge computing).
• Where do people see the hybrid cloud going? Will containers change the cloud decisions in the modern world? Are people mixing and matching cloud, on-premise, and hybrid data? Look at the distribution of mobile, IoT, and multinational going into the same datacenter.
• It’s important for people to evaluate the expense of a solution relative to the value provided. Look at the results rather than the cost or the image of the provider.
• We are in the API economy but there’s a lot of hype that APIs are the "be all and end all." A number of customers think that a vendor that provides a prebuilt API is equivalent to easy integration. It’s what connects front doors that determine the success of how well applications work together.
• Take this to another level. Are people in different organizations structuring data entry for data flow and data in motion? Are there new roles and responsibilities? Centers of Excellence have happened in other layers of the data center, should it happen for data flow given the movement of data between nodes?
• Look at Open Source and proprietary management systems and decide what’s right based on the culture and nature of the enterprise. Swift and Ceph are open source protocols.
• People don’t think enough about efficiency. We are generating and consuming more data every day but the network is not growing. How do we keep up? Why pay for the distribution of data that you will not need? Data is the key component of everything we do. Don’t overlook infrastructure. Load balancers are expensive. Rest polling, request/response headers, payload with redundant data. We can reduce the infrastructure load by four to five times  which also reduces the Capex. All data has electricity costs; therefore, companies can save dollars and cents by being more efficient with data thus reducing operating expense as well.
• In the integration space EDI is still around after all these years. What will be the enduring technologies? Client server and mainframe companies are being disrupted by start-ups and changes in technology.
• 1) One of the biggest areas to consider is the overall integration strategy.  Integrating solutions together without a larger vision increases the likelihood of vendor lock-in.  The more we move towards a microservices approach, the more increasingly important this strategy becomes in order to avoid architecture death by a thousand cuts. 2) This is also something that is commonly missed - as your integration strategy shouldn’t be just to connect devices, but reduce vendor lock-in, increase agility, and increase efficiency (including when replacing existing systems or migrating data). 3) As mentioned earlier, this is where advancements in hypermedia and code on demand will help, but not eliminate (as systems may need to be pulled out completely).  For this reason, it’s important to go in with an API-led approach to building your infrastructure as an application network.
• Integrations across applications from multiple generations with associated generations of data (e.g. mainframe, RDBMS, NoSQL, Columnar, document, etc.).

What other issues do you see affecting application and data integration that we have not covered over the past two weeks?