Gaining Control in a Multi-Cloud Environment

Gaining Control in a Multi-Cloud Environment

By Special Guest
Ravi Shankar, SVP and CMO of Denodo
  |  September 30, 2020

Cloud computing technology is growing more quickly than ever before, and multi-cloud architectures have become the de-facto standard for organizations looking to modernize their applications to support a data-driven architecture.

In recent years, cloud adoption has become much more manageable, as organizations have the ability to diversify spend and skills, build resiliency, and cherry-pick features and capabilities depending on each cloud service provider’s particular strengths, all while avoiding the dreaded vendor lock-in. To overcome the challenges of integrating data between the multiple clouds in a multi-cloud environment, organizations are leveraging logical data fabric, a flexible, modern data-integration approach, which seamlessly shields users from the disparities of the dissimilar clouds.

Weaving a Logical Data Fabric to Maintain Cloud Control

Many organizations struggle to gain control over their multi-cloud environments. This is because of the vast differences between different cloud service providers for not only storage, compute, and networking, but also applications, data management, and security. According to this year’s Denodo Global Cloud Survey, security and governance-related continue to top the list for cloud challenges, as indicated by close to 70 percent of the respondents.

As a result, business users do not have a unified view of the data across the applications running on the different clouds, and have to access the applications separately and then manually integrate the data. Logical data fabric removes this constraint by knitting together the data from the different applications into a combined view so that business users can access all of their data from the fabric, rather than having to go to each separate clouds. Logical data fabric abstracts users from the differences across the clouds and leverages data virtualization as the core technology to accomplish multi-cloud data management.

Data virtualization is modern data integration technology that (1) connects to disparate data sources irrespective of their location, whether in the public or private cloud, (2) combines the data from these sources into a unified view, and (3) enables business users to consume the data from this combined view. Abstraction is a fundamental principle of data virtualization. By acting as an intermediary between the various clouds and the users who access them, data virtualization liberates business users from having to learn the different methods for accessing each cloud, while enabling IT to use the right clouds for running the workloads appropriate for their organization. Today, users are looking to simplify cloud data integration in a multi-cloud environment without having to depend on heavy duty data migration or replication. This may explain why almost 50 percent of respondents in the Denodo (News - Alert) Cloud Survey said they are considering data virtualization as a key part of their cloud integration and data migration strategy.

Pitfalls to Avoid When Managing a Multi-Cloud Environment

The biggest mistake enterprises make when trying to gain better control over a multi-cloud environment is to approach cloud management as an afterthought rather than a key strategy. Organizations tend to select cloud databases and applications one-by-one; they might substitute a legacy Oracle (News - Alert) database with a NoSQL cloud equivalent, and choose Azure Cosmos DB as the replacement. Next, they might want to migrate their data lake to a cloud object storage implementation, and choose Amazon S3.

As a result, they now have two separate data stores in different clouds, and analytics users have to point their BI tools to Cosmos DB for one type of data, and S3 for another. This bifurcation creates difficulty for IT users, who need to manage the different environments, and also for business users, who have to resort to manual data integration of different data across the disparate systems and labor-intensive reconciliation of the different formats.

Corralling Your Clouds Starts with Careful Planning

As data continues to spread across disparate sources and geographic regions, it is even more important to ensure that data movement and replication is minimized, to save time and cost. Therefore, even before organizations embark on their multi-cloud journeys, it is important that they embrace logical data fabric as a fundamental part of their modern data architecture.

By first creating a modern data architecture based on logical data fabric, IT teams can enable a more strategic approach to onboarding different clouds without impacting business users in their day-to-day operations.

About the Author: Ravi Shankar is senior vice president and Chief Marketing Officer at Denodo, a leading provider of data virtualization software. For more information visit https://www.denodo.com or https://twitter.com/denodo.




Edited by Maurice Nagle