People often talk about data and cloud infrastructure as if they are two different things—and in a strict technical sense, they are. But in my mind, data platforms and cloud infrastructure are so dependent upon one another that they are essentially one and the same.

When it comes to ensuring optimal performance and the highest levels of security for your IT systems, data platforms and cloud infrastructure need to be evaluated in tandem. Let’s take a look at the reasons why.

Effective Big Data Collection and Analysis

Think about all the data—both structured and unstructured—that gets passed between devices, networks, and cloud infrastructures. It amounts to mountains of data, and those who collect and analyze it correctly often find there’s gold in those mountains.

The cloud is the best place to collect, store, analyze, and ultimately gain valuable business insights from big data. The cloud offers several benefits over managing big data analysis on premises, including greater agility and elasticity—you can quickly spin up or spin down as needed—and far lower management costs.

It’s also difficult to beat the level of security the cloud provides. Oracle and KPMG recently surveyed 456 cybersecurity and IT professionals in North America, Western Europe, and Asia and found that a whopping 72 percent believe the cloud offers a more secure environment than they can provide on premises.  

But to collect and analyze big data effectively and efficiently, organizations require a cloud infrastructure with globally-distributed collection points designed to verify the integrity of data and get it into your private network as quickly as possible. The farther data has to travel to get into your network, the greater the chances are that something could go wrong in terms of its integrity and security.

Suppose a user in Hong Kong initiates a transaction that ultimately hits your network in Ohio. If that data first zigzags through networks that aren’t controlled by you or your trusted cloud provider, the chances of bad things happening rise significantly. For example, the data could fall victim to packet loss and other latency issues, BGP route hijacking, and distributed denial-of-service attacks. Any of these things can negatively impact the amount of insight you’re able to get from data once it is analyzed.

If that same data travels from the endpoint in Hong Kong directly to one of your cloud provider’s points of presence located just a few blocks away, however, the chances of experiencing those problems significantly goes down.

When it comes to big data analysis, it’s important to use a cloud that keeps data as close to your private network as humanly possible.

Hyper Defense in Depth

I travel to many IT industry conferences and attend lots of meetings with security experts. During these sessions, the topic of defense-in-depth IT security models comes up often. A defense-in-depth approach to cybersecurity is based on the notion, originated in the military, that it’s harder to breach complex, multilayered defenses than it is to penetrate a single barrier.

In IT, a typical defense-in-depth strategy involves the use of antivirus software, firewalls, strong password policies, identity and access management software, and ongoing cybersecurity training for employees.

Those tactics are all important, but they mostly deal with the defenses that can help protect your cloud deployment from bad actors attempting to access it from the outside. A good defense-in-depth strategy—or a hyper-defense-in-depth strategy, as I like to call it—should also take into account the defense layers available within the cloud infrastructure itself.

For example, organizations should consider whether their cloud provider has multilayered security features and proven processes in place to protect against attack internal threats like corporate theft, unauthorized access, and software vulnerabilities that could lead to data loss or exposure. They need to protect against application, data store, and hardware and infrastructure threats in addition to external threats.

It’s important to think about the design of your cloud infrastructure—and precisely how it works to protect your data—if you truly want to achieve hyper defense in depth.

Database Optimization

Data management software is often optimized for a particular cloud infrastructure with specific hardware.

If the cloud provider is using servers, processors, and drivers that work best with your database software, the results will be an improved security posture, better price-performance, and better overall performance. And if the database is a service inside the cloud platform, even better. That’s the approach we take with the Oracle Autonomous Database — a first-class service that enables the best data management across the board.

When evaluating cloud infrastructure providers, be sure to gauge how your data management software runs inside that cloud. If you find that another cloud does it better, then it’s likely time to move on.