Cloud security

Is The Cloud secure?

We talk with a lot of customers and the two questions, or opinions, we meet frequently is; The Cloud is expensive and The Cloud is not secure. We cover the economical part of The Cloud in a separate blog post, this post will dive into the question whether or not The Cloud is secure. It’s an interesting question, but in general wrong. A better question is; “Is The Cloud suitable for me?”

When facing the question or the statement “The Cloud is not secure!”, meant to close the discussion. I try avoid answering. Instead I turn the question around. “Do you think that your business will benefit from shorter time-to-market, higher speed in business innovation and meeting customer expectations, less upfront investment for IT equipment and a competitive edge towards the rest of your industry?” If only the answer hints off a Yes. I reply; “Then we make The Cloud secure for you!”

The Cloud is secure enough

Normally when talking about The Cloud we often mean the big Public Clouds provided by companies like Microsoft, Google, AWS, AliBaba etc. Today they all have built in services similar to traditional on-premises security controls but with other names. In many cases with built in capability to provide traceability and transparency that facilitates monitoring and compliance evidence.

Source: http://www.eventid.net/docs/onprem_to_cloud.asp

This makes it possible to get at least the same security level in the The Cloud as on-premises. What’s needed is likely a changed skill set in the organization when operating in The Cloud.Another very important thing to understand about The Cloud is that the responsibility “up” there is shared. But it’s not shared in an obscure way. It is very well defined. The cloud provider is responsible for the security OF The Cloud and the customer is responsible what’s IN The Cloud. Shared, and crystal clear. If google “shared responsibility model” you got millions of hits and can check yourself what it means for a specific cloud provider. I give AWS view as an example below.

Source: AWS Shared Responsibility Model

If we apply the above on two use cases, the responsibility works according to the following. If you choose to use The Cloud as a:

  • virtual datacenter and install some virtual computer with relational databases and additional softwares. You are responsible for the security of the operating system on the virtual computers you installed, the update of the database engine and additional softwares as well as the data you are putting into the database. The cloud provider is responsible for the virtual Datacenter and you for everything you put into it.
  • virtual service center and selects a high level database service or a full fledge cloud based business application (SaaS), such as Salesforce. You are responsible for the data you put into the services. Not the computers and the software the services are running on. That is the the cloud provider’s responsibility.

The consequence is that the customer decides and are in total control of how secure their part of the responsibility shall be and the cloud provider about their part. The way the cloud provider commit is via certification towards recognized standards such as ISO-27001/2, HIPAA etc etc. and that they maintain their certifications and continuously publish reports of compliance.

The responsibilities are clear and the tools are there to make The Cloud as secure as needed.

If The Cloud is suitable, is a business decision

But in order to understand why I think the question should be “Is The Cloud suitable for me?” We need to take one step back. What is it that actually shall be secured? It’s normally not the computers in the data centers or the software on them we talk about.

We want to secure the information we put into them, and the capability to process it and utilize it for the benefit of our business. We want to make public only that information that shall be public, and keep the rest hidden for the public. But accessible for the right audience at the right time. 

The question left to answer is. Are we allowed to put the data in The Cloud. If the answer is “No” then the The Cloud is not suitable even if it can be secure enough from a technical perspective. But if the answer is “Yes”, partly or fully, then The Cloud is a competitive candidate towards on-premises.

What kind of requirements can forbidd a organization to put their data in The Cloud. I see to major groups of requirements:

  • Legal requirements
  • Contractual requirements

Legal requirements are hard to challenge. But contractual ones can be discussed with the other parties. So in order to answer the question if the The Cloud is suitable, we need to know what the law say and what contractual requirement we have for the specific set of data we plan to move to The Cloud, and of course the processing of it.

When we know that, we not only understand if The Cloud is suitable. We also have information enough to select a suitable cloud provider. For the set of data that is not allowed to be moved to The Cloud. On-premises operation might be the only choice left.

An embryo to a Cloud First strategy

Doing this exercise we have turned the question of “Is The Cloud Secure?” to a normal business decision about balancing business benefits and business risks. At the same time we have created the first embryo of a Cloud First strategy. That’s not bad.

Besides security, another common opinion about The Cloud is that it’s expensive. Read Anders Erikssons blog post “Cloud is expensive”.