WP_Data Management Best Practices

12 / 96

W H I T E P A P E R
© 2017 Persistent Systems Ltd. All rights reserved. 12
www.persistent.com
For companies that are just starting to use cloud computing, most of the usage will be for new applications, with stand-
alone being the low-hanging fruit. Those that have already critical cloud apps in use (e.g., ), have
already solved the cultural and some technical problems in transitioning to cloud computing, so data management in
cloud databases will make more sense for them.
Much of the cloud integration challenge centers on connecting cloud and on-premises applications (not just moving
data from an on premise database to a cloud data warehouse), which is developed in our next section.
3.1.3 Cloud Application Integration
In the early days, the first cloud application integration tools focused largely on integrating the
earliest successful cloud application, with packaged ERs providers (SAP, Oracle) and other on premise applications,
usually by moving files in batch from the cloud to on premise systems.
Today’s organizations have adopted a variety of cloud and on premise applications to manage their business. IT
departments are now expecting that hybrid system landscapes will become the standard way to manage the
6 organization’s IT assets in the future . In fact, cloud applications will probably be favored over on premise applications
7 when there is a choice, but on premise applications will not go away anytime soon . On these hybrid landscapes,
8 process and data integration across the cloud / on premise boundaries is a must to prevent data silos and allow
seamless end-to-end business processes.
To illustrate this trend with a specific example, let’s use the CRM (Customer Relationship Management) paradigm.
When they were first introduced, the basic thinking was that the company owns the relationship with the customer. An
efficient CRM was one that would allow to easily create new fields and do better customer segmentation. Data quality
management was focused on rules and control for create-read-update-delete (CRUD) operations. CRM is now
changing towards a dynamic, highly personalized and multi-channel collection of interactions where customers enter
in different kinds of engagements with a company. In this new setting, customers tend to be more self-enabled and
interact with the company through different applications: first maybe a marketing app, then an onboarding app, then
one or more services apps; some are deployed in the cloud, some on premise. It is no longer a CRM package that
owns the relationship. It now becomes necessary to seamlessly manage these interactions. In this setting, best-in-
breed customer MDM systems are not those that just govern customer master data through CRUD operations but
those that are focused on automated completion of customer master data along the paths a customer takes in his
journey with the company.
The following requirements on data integration and governance on these hybrid landscapes were drafted in large
enterprise projects we have participated in the past, and they include:
— Real-Time Business Process Integration
— Agility, in particular in integrating new sources/targets
— Responsiveness to new functionality and technical changes
— Batch and real time data movements on both individual records and collections of records
6According to Gartner, 70% of enterprises will pursue a hybrid cloud hosting model by 2015
7A recent TechTarget surve shows that more than three-quarters of organizations (77%) have either all or most applications running on-premises.
8See the Glossary sectio for a definition of this term.