In recent years, the question has been, “Should we move to the cloud and, if so, how and when?”  Today, the questions are more granular: “What kind of cloud or clouds should we leverage?”, and “How can we integrate multiple cloud technologies into our existing infrastructure and workflow?”

Most data architectures today involve a mix of on-premises and cloud technologies, and many companies use a variety of cloud platforms, such as Google (News - Alert) Cloud for advanced analytics and Microsoft Azure Cloud to run enterprise applications.

Rightscale, in its 2017 State of the Cloud Report based on a survey with over one thousand respondents, stated that “85 percent of enterprises have a multi-cloud strategy, up from 82 percent in 2016.”  Meanwhile, companies are not simply divesting their on-premises data assets but are seeking ways to integrate these multi-cloud architectures with their existing data sources.

Data Access Challenges

This approach creates a data integration challenge due to very large volumes of data, and for use cases where the data needs to be integrated in hours, if not minutes, or in real-time.  Traditionally, data has been integrated by extracting it from the various source systems, transforming it as necessary, and loading it into a central data warehouse via extract, transform, and load (ETL) processes.  However, these are batch-oriented processes, so they are not feasible to support real-time use cases.

This data integration challenge creates at least two critical data-access obstacles: How do users know what data is available, and how do they find and access it?

Enterprises are moving away from legacy, monolithic applications and suites deployed on-premises, and toward specialized SaaS (News - Alert) applications in the cloud.  Their key objectives in doing so are modernization and substantial cost savings.  However, multiple SaaS applications create additional challenges such as:

• How do stakeholders access the data in SaaS applications?
• How do users get a holistic view of data across multiple, specialized applications?
• How do users get data into the SaaS apps?
• Will organizations provide users with access to each SaaS application?

A Novel Approach

While other options exist, one technology provides a completely different data integration strategy.  Data virtualization directly mitigates the challenges of multiple data sets stored across heterogeneous sources as well as the challenges of multiple SaaS apps.  Rather than creating copies of the data and attempting to house these in single, monolithic repositories, data virtualization sits in a layer above all the data sources and creates pointers to the data, enabling data consumers to access the data in real time.

Users don’t need to know the details of where the data is stored, whether its on-premises or in the cloud, or in one SaaS app or another.  Users, and applications that leverage the data, simply query the data virtualization layer which then takes care of the rest.

Data virtualization also enables organizations to leverage all the advantages of hybrid, multi-cloud architectures, while avoiding all the potential pitfalls.  It accomplishes this in many ways, including:

• Migrating data to the cloud – Because the data virtualization layer masks the complexities of accessing the data, companies can migrate data to cloud on their own schedule, without incurring downtime, and without users even knowing that a migration is taking place.
• Moving from on-premises apps to SaaS apps – On-premises applications tend to be monolithic suites that centralize information for companies in a reliable way.  When companies start using individual SaaS apps, they give up that centralized view.  Data virtualization restores this view across apps, and easily accommodates new apps into the mix
• Providing real-time data access across hybrid and multi-cloud architectures – With data virtualization, heterogeneity is not a problem.  Data virtualization works with any existing architecture, no matter how diverse, and provides real-time access to all available data as if it were stored in a single repository.

Leveraging the True Power of Cloud Technologies

Cloud computing is here to stay and will continue to grow in importance, as organizations seek to leverage the agility and flexibility that it provides.  For the foreseeable future, though, these hybrid information architectures based on a mix of on-premises and cloud-based data sources will continue to cause challenges.  Organizations need to consider a technology that specifically addresses this mixed environment by creating a hybrid data fabric that spans both cloud and on-premises solutions.  Technologies such as data virtualization provide a unified data access and security and governance layer, which then provides this “unified data fabric” to support use cases, including cloud-to-cloud integration (e.g., SaaS applications), cloud-to-ground integration, migration of data sources/ applications to cloud, and analytics in the cloud.

About the Author: