Why CIOs Shouldn’t Race to Move all Data to the Public Cloud

In this article Jon Toor, CMO of Cloudian discusses the race to move all data to the public cloud may seem like the right move given widespread work from home and an increasingly cloud-centric industry, but CIOs may soon realize it’s not the smartest choice.

The move to the public cloud is nothing new for most enterprises. In fact, IDG’s 2020 cloud computing surveyOpens a new window found that 59% of enterprise tech buyers said they planned to be “mostly” or “all” in the cloud within 18 months, up from 38% in 2019.

With the sharp increase in remote work, companies have increased their use of the public cloud. A survey from Maria DBOpens a new window taken in May of this year, after the initial impact of the pandemic, showed that cloud migration had risen. The survey found that 40% of enterprise tech managers in France, Germany, the U.S., and the U.K were accelerating their move to the cloud, and 39% expected to be 100% in the cloud.

Important Criteria and Best Path Forward

Despite this trend, there are fundamental realities that may make moving all your data to the public cloud a

poor decision in the long run. 

To choose the best path forward, it’s important to evaluate your situation: how much data do you have, how is that data used, and what growth do you expect? 

Other important criteria include your particular security and regulatory compliance exposures. Fortunately, the options to address these needs are now broader than ever, with regional service providers offering robust and cost-effective offerings, plus vendors with on-prem, private cloud offerings. The best path forward maybe some combination of these.   

Moving entirely to the public cloud means turning control and security of all your data to a third party and often proves to be more expensive than an on-premises or hybrid approach. Soaring costsOpens a new window associated with cloud are already being reported, and companies are scrambling to get it under control as they try to remain conservative with spending during COVID-19. 

While it may be tempting to move all data to the public cloud, there are critical factors to consider that will undoubtedly impact your organization’s performance, budget and security. 

Learn More: Legacy to Cloud Transformation – Are You Invested Enough?

Performance and Agility

First, it is important to consider the impact public cloud performance has on overall application performance. 

Available WAN bandwidth and the number of workloads running on each cloud provider are significant factors impacting public cloud performance. The higher the workload capacity, the more time will be needed to move its data in and out of the cloud.

For example, it needs about 180 minutes to transfer 1 TB of data over a 1 GB link. Application requirements must also be factored in, potentially adding on additional time. If you consider backing up data to the cloud, these two factors will be critical in understanding if your backup performance objectives can be met. 

On-premise object storage is a much more flexible choice for many use cases since it offers local access to data through multiple high-speed network connections that run concurrently. Organizations can quickly deploy the space they need to run data-intensive workloads with low latency because on-prem object storage is infinitely scalable.

Learn More: How To Implement Chaos Engineering Using Cloud To Improve On-Prem Systems

Cost

As mentioned previously, many companies have already started to see the negative financial impact of the public cloud, with larger-than-expected monthly bills coming in since the start of the pandemic. The issue is that there are many hidden costs associated with using the public cloud. This unpredictability is particularly concerning amid an economic downturn and tighter budget constraints. 

Apart from storage costs, data egress and data replication fees are associated with utilizing the public cloud. These typically account for 15-20% of an organization’s cloud bill, but they can be higher in storage-intensive workloads.

This is on top of the cost of the cloud’s WAN connection. When performing computational tasks on massive data sets, network bandwidth and data access charges appear to rise significantly. Companies using artificial intelligence and machine learning workloads and basic analytics operations on a massive database are increasingly using this approach. The final bill is influenced by the volume of data accessed by the user, which can vary drastically on a month-by-month basis.  

On-prem object storage is preferred over the public cloud if storage usage exceeds a few hundred terabytes. This helps to avoid data access/egress fees, expensive network bandwidth and hidden charges that come with running high-capacity workloads (such as AI or ML).

Learn More: How Cloud Tech Helps “Everyday Life” Companies

Security and Control

Security and control are becoming increasingly as significant as, if not more than, performance and cost considerations, as security can impact both if not done correctly. When it comes to encryption in the public cloud, there is a lot of doubt about who is responsible for it. Many public cloud users believe that their cloud provider protects their data. However, this is not the case, and as a result of this confusion, many organizations have become vulnerable.

Something to keep in mind, switching from securing public cloud deployments to securing on-prem deployments makes new public cloud customers vulnerable to breaches. Because of this, CIOs need to prioritize security to keep data protected. 

CIOs can maintain data on-prem or in a private cloud to have complete control over security and data access.

Alas, the public cloud is still an appropriate option for some use cases, such as disaster recovery, which can be done cost-effectively in the public cloud. For the reasons outlined here, the public cloud is not the best choice for all workloads and sectors, despite common belief.  A hybrid cloud model is a better approach, and because object storage is based on the S3 API – the de facto language of public and private cloud storage – it’s an ideal platform for a hybrid cloud. 

Did you enjoy reading this article? Let us know your thoughts on LinkedInOpens a new window , TwitterOpens a new window , or FacebookOpens a new window . We would love to hear from you!