Is Your Data Ready for a Multicloud Strategy? Four Steps to Get You on the Right Track

By Special Guest
Kendall Clark, Stardog CEO
  |  April 05, 2021

Every few weeks for the past year, there’s another “are you ready for multicloud?” checklist. These specs are very thorough and often include support, devops, policy, governance, and risk considerations. Good and necessary stuff, not to mention a great management tool for scalability. But when they’re missing key strategic considerations, they may well lead to a false sense of security. After all, if we’ve attended to everything on the checklist, one could assume they must be ready, right?

The problem is that none of these multicloud worksheets address multicloud’s impact on data strategy. Multicloud is not just about the existence of more computing environments, it’s also about more places for data to hide and to remain disconnected. If we care about digital transformation at all, we have to start thinking more carefully about the impact of multicloud on data and how we manage, integration, and connect it to achieve data-driven insights.

Why Multicloud Makes Data Integration Harder, not Easier

Hybrid, multicloud and cloud repatriation all put immense pressure on conventional data management products and techniques, because conventional data management uses data location in physical storage as the primary leverage point. Perhaps the key fact of modern enterprise data management is that some data always evades the system. People used to say information wants to be free, but now we should be saying that data is rebellious and hates being contained. But even setting that aside, the storage-location-of approach has the unavoidable requirement that data is moved and copied between different physical locations, all of which takes place over networks. 

This is as true for the cloud era as it was in the pre-cloud era, and it’s even more true in the multicloud era since, simplistically, multicloud means there is an increasing, rather than decreasing, number of data environments. There are fundamental obstacles to continuing on the present course of only moving and copying data in order to manage it, including data volumes increasing while networking performance improvements are slowing down. At the same time, macro-trends like multicloud mean that there are an increasing number of places where data can live, which would seemingly require us to more frequently move and copy data! Something has to give.

Four Steps to Ensuring You are Ready for Multicloud

There are four courses of action to consider after concluding that cloud reality impacts data management strategy and practice downstream. After all, cloud environments are IT environments and perhaps the main function of those is to drive insights, efficiencies, and ROI with data-backed decisions. Organizations do not compute in the cloud entirely to transform the business into data-driven systems, but that is clearly one of the fundamental motivations.

The first step is to recognize and explicitly account for the fact that multicloud impinges upon data strategy in a very impactful way. Until we realize that, there isn’t even a hope of responding rationally and productively. The key takeaway is everything we do in IT impacts data strategy and data management, especially in multicloud. The natural state of enterprise data is to be siloed and disconnected, fragmented, and decontextualized. Multicloud is going to increase these challenges just by virtue of the physics involved.

Second, having recognized the connections and dependencies between data management strategy and cloud strategy -- including hybrid, multicloud, repatriation, etc. -- let’s rethink these checklists and supplement them with the right kinds of considerations such as:
 

  • Is our data management strategy ready for multicloud, especially when it comes to data integration?
  • How will multicloud adoption or growth impact data movement and data copying?
  • Do we have a strategy or plan for connecting data across public cloud vendors without running costly, and politically fraught, system and app consolidation plays?

Third, flip this discourse on its head and start asking, not so much how does multicloud impinge upon your data management strategy, but rather how are we going to integrate data in the hybrid, multicloud world? This shift in perspective moves organizations from a reactive to a proactive posture. There are huge opportunities for competitive advantage by moving first in the new world, recognizing the opportunities to accomplish digital and data transformation first. If that is to be accomplished, it will have to be done in a multicloud world.

The final step is to acknowledge that multicloud strategy isn’t necessarily a distributed strategy. Many of the biggest cloud native vendors are monoliths that, even if they could be ported or moved to other public cloud systems, are not in fact distributed in nature. But your enterprise data landscape is. The data silos are an existing, inherently distributed data placement plan, however imperfect or duplicative. 


Diverse approaches to data management exist in a growing tranche of startups vendors that are innovating around distributed, federated, and virtualized data management techniques, all of which are inherently better adapted for the hybrid multicloud world than existing approaches. A fundamental revolution of data management is coming, and it will be anchored and framed around the needs of the distributed world that hybrid multicloud now portends.

Managing the Checklist to Ensure Proper Data Management

The new multicloud world will increasingly demand that data be managed, integrated, connected, and queried where it lives. This net new capability will not replace existing storage-based systems like data lakes, data warehouses, and the like. However, it will serve as a strategically essentially additional capability. The era of data integration by storage is not over, but the era of data integration by storage only is very near to its end.


Key to this end will be the cloud itself; hybrid, multicloud, repatriation, and the like will cause all of us to reckon deeply with how we manage data. Checklists that are innocent of these facts and trends are not the problem, they are an opportunity to drive the conversation and strategic work to the right outcomes.

About The Author: Kendall Clark is founder and CEO of Stardog, the leading Enterprise Knowledge Graph (EKG) platform provider. For more information visit www.stardog.com or follow them @StardogHQ.


 


 




Edited by Maurice Nagle