Exclusive: The emergence of cyber threat modelling
James Thorpe
Share this content
Occasionally I stumble across some parts of the industry that I have failed to give due attention. I do try to stay informed. I attend events and webinars and have my peer networks like I suspect many of us have. So, when recently I was introduced to the concept of cyber threat modelling, I was at best sceptical.
I was vaguely aware of modelling techniques but not those used in a cyber context. So, I was intrigued to hear that researchers at the Swedish Royal Institute of Technology (KTH) had been developing modelling methods for the non-functional attributes relating to software architecture, system architecture and enterprise architecture for complex IT environments. The output of ten years research was the commercial release of a product provided by a Swedish company named foreseeti.
The software comes in three different flavours, Professional, Enterprise and Vanguard for AWS cloud, but I will focus on the cloud offering as that is my current focus supporting a digital transformation programme.
Firstly, lets start with the Achilles heel of any security programme and that’s asset management. Not just asset management but understanding its communication interconnectivity across the network. Threat modelling data is ingested via either vulnerability scanning tools or a replica extract of your cloud environment. This can provide you with a precise up to date network configuration and immediately highlight communication channels that are susceptible and vulnerable to attack. The methodology used from an attacker is supplemented using the Mitre Kill Chain.
The next challenge for security programmes is understanding the effectiveness of security controls. The threat modelling algorithms place the tactics, techniques and procedures used by adversaries across the vulnerabilities of your infrastructure. Drawing a clear diagram of vulnerable attack routes.
Once you have this network information you can place different controls into your threat modelling and allow it to present a host of different impacts those controls will have on your environment.
The great benefit this provides is that your infrastructure planning can be developed before any instance is actually built. You can manually replicate your target state and run these threat models against them.
If you already have an environment, you can export the configuration to again run different simulations against a number of possible controls. This in my mind is hugely significant in reducing the costs and improving effectiveness of design.
More importantly it provides clear visibility of your current infrastructure allowing you to make informed decisions that can easily be articulated up to your leadership team.
I predict, and I know that is dangerous, that the future of successful transformation projects will inevitably have to adopt modelling. Be it business, infrastructure or cybersecurity.
This article was written by Mo Ahddoud, International Security Journal’s resident cybersecurity expert and CEO of MA Consulting Ltd. To find out more, please visit: https://www.macyberuk.com/