In the Age of Multi-Cloud, Is Data Governance Receiving the Attention It Deserves?

Despite all the hype, the public cloud is far from being the default hosting environment for enterprises in 2021. 

In 2019, a essidsolutions survey found that a staggering 98% of businesses continue to host some of their workloads on on-premise servers. Last year, this had to necessarily change due to the pandemic, but despite the shift, less than 4 in 10Opens a new window companies said they are mostly or all in the cloud. The reality is, no one hosting environment can fulfill all the requirements of a growing business, and enterprises are increasingly looking at more hybrid options to keep up with a dynamic, post-pandemic climate. 

That’s why multi-cloud has emerged as such a popular option. 

ResearchOpens a new window suggests that multi-cloud environments are on the rise, with the average organization using 2.2 public cloud services and 2.2 private cloud services, respectively. There are several reasons for this, including the fact that some workloads requiring low latency are simply more suited for private hosting, while others requiring dynamic scalability are best left in public, multi-tenant resources. But one of the most important considerations IT decisions makers must keep in mind when transitioning to the public cloud is data governance. 

A fully private or fully public (single vendor) cloud infrastructure is relatively simpler in terms of data visibility and governance. There is a centralized dashboard, and applications don’t have to be portable, reducing the need for data testing. All of this changes when IT decision-makers start embracing multi-cloud. Here are the top considerations to remember during the shift. 

Learn More: How to Improve Data Governance in Your Organization 

1. Paying attention to standards and compliance 

Data security laws differ across the world, and companies must be mindful of adhering to local and international regulations when storing data in a multi-cloud environment. Standards like PCI/DSS for the payment industry, HIPAA for healthcare data, and GDPR for serving EU citizens should be at the top of your list. The ownership for conducting audits and demonstrating compliance is typically shared by the cloud vendor and the enterprise owning the data, depending on the SLAs. 

IT decision-makers must carefully assess their end-to-end multi-cloud footprint, documenting dependencies on in-house, private, and public stakeholders to gain full clarity on data compliance, working closely with local regulatory authorities. 

2. Reducing the possibility of data duplication 

As the number of buckets in your storage capacity grows, the risk of duplication will multiply accordingly. And there is no denying it – duplicate data poses a real risk to an enterprise, in the form of:

• • Exponential storage costs to house redundant data
  • Security vulnerabilities, if the location of duplicate data isn’t mapped 
  • A dip in data integrity, leading to poor quality analytics 
  • Additional de-duplication costs and efforts 

It is crucial that you avoid duplicating data stacks when porting applications from one cloud to another and maintain a single source of truth. IT decision-makers need to establish strict controls to store the right data in the right place (without forming duplicate silos), preferably through interconnectivity between multi-cloud sources to centralize visibility. 

3. Keeping track of co-location centers 

There is a saying that cloud-hosted data is nothing but storing your data on someone else’s computer. For multi-cloud, this means your data and data processes will be spread across multiple computing systems, servers, and data warehouses, possibly worldwide. Companies – particularly large enterprises – should keep a close watch on the co-location center’s geolocation. Public cloud vendors like AWS let you chooseOpens a new window the specific location for hosting many of its services. IT decision-makers can factor in security policies, latency needs, and geopolitical relationships to choose the ideal co-location center for their data resources. 

Learn More: 3 Questions To Drive Better Data Governance 

4. Maintaining cloud data backup 

Cloud data backup is a double-edged sword. On the one hand, it is hosted in a location far from your primary center of operations, making it resilient to any natural disasters or outages you might face at the HQ. But on the other hand, the shared nature of cloud resources (which is compounded by a multi-cloud environment) means that it will be more exposed to cyber-risk that a shared tenant might be facing. That’s why cloud data backup has to be a top priority when transitioning to multi-cloud. 

The movement has been slowOpens a new window in this regard, with the number of companies using the cloud for data backup recovery or archiving growing by just 12 percentage points between 2016 and 2019. 

IT decision-makers can turn the multi-cloud nature of their data environments into an advantage by leveraging it for the 3-2-1 backup rule. Create 3 copies of your data, store in 2 formats (preferably one in a backup cloud and one in a private server), and maintain 1 primary copy on a different cloud than your backup environment. 

5. Allocating data and workloads to internal or external cloud services 

Multi-cloud gives you a variety of options when it comes to hosting your data and workloads. You could choose between on-premise servers, multi-tenant public clouds, non-shared private clouds managed by a public cloud vendor like AWS, or a remotely located private server. Companies increasingly realize that the public cloud may not be the perfect, one-size-fits-all solution for every type of data, and multi-cloud offers a smarter alternative. A recent research reportOpens a new window found that 84% of organizations on the cloud have moved select workloads from a public entity to a different provider or even a non-cloud location. 

It is key to judiciously allocate your datasets across the available hosting environments, ensuring there are no duplication or security vulnerabilities. For low latency workloads where you might need to fetch data multiple times in the day/hour, an on-premise environment or a local public cloud co-location center might be a better idea. For historical data records, archives, and long-term analysis, a secure remote private server is recommended. 

Remember, multi-cloud isn’t a set-and-forget tool. 

IT decision-makers can regularly assess, revisit, and reconfigure their data hosting mix to meet evolving business requirements and newly emerging performance trends – and it is highly recommended that they do so. You could also take help from a managed services provider, like 84% of organizations in the last report we cited. 

Learn More: Cultural Change Through Integrated Data Governance and Data Quality  

6. Avoiding budget overruns 

This is an important consideration when moving to multi-cloud, as additional cloud resources and potentially unmapped shadow resources could inadvertently exceed your cost estimates. It is not uncommon to leave a static data stack tucked away in a subscription-based public cloud resource without clear documentation. When there are changes in the IT team, the data stack is overlooked and possibly duplicated in a different location, even as the enterprise continues to pay for the original hosting resource. 

This has become so common that 1 in 3 organizationsOpens a new window are now exceeding their cloud budgets by up to 40% in 2021. 64% of organizations cite cloud cost management and containment as their biggest concern, primarily due to big data proliferation. The first step to avoiding budget overruns is getting visibility. Connectivity between multi-cloud resources and centralized governance dashboards can help map utilization against costs. You could also leverage dedicated cloud use optimization tools to make the most of your investment without exceeding your initial cost estimates by a wide margin. 

There is no around it – multi-cloud environments are full of complexity, yet they offer purpose-built solutions that are simply not possible in a homogenous hosting environment. As enterprises reach new levels of digital maturity, the adoption of multi-cloud is inevitable. However, adequate data governance must be accompanied by good data governance, factoring in compliance, duplication risks, physical locations, backup, budgets, and appropriate data allocation to make it sustainable in the long term. 

Do you agree that the multi-cloud strategy is here to stay? Comment below or let us know on LinkedInOpens a new window , TwitterOpens a new window , or FacebookOpens a new window . We would love to hear from you!