Mission to Helsinki
I grew up in Northern New York State, so a trip to Helsinki in the middle of February held no fears for me. Interesting things are going on in Finland from a cybersecurity point of view, so I jumped at the chance to speak to the Security Day conference in Finland’s capital city. The conference appearance was actually one stop on an itinerary that took me to three countries, two press conferences, and four customer visits…in five days.
In some ways, it’s a tribute to globalization that audiences all over world share the same concerns about cybersecurity. Mobility, identity, explosive growth of an Internet of Things, and an increasingly malicious threat environment are as much on the minds of the people I met in Finland as they are in every part of the world I have traveled. I also found it notable that the Security Day conference celebrated its 12th anniversary this year with the largest number of attendees in its history. My talk centered on three kinds of methods that can make it harder for cybersecurity adversaries to succeed. First, I recommend doing the basics—patching, asset inventories, identity management, visibility into device and user behavior—and doing them well. Here it is particularly important to eliminate any dark space in an infrastructure. It’s the assets and users that you don’t know about that will oftentimes create our largest risks.
Second, the security community has been innovating some delightful ways to lead adversaries on merry, frustrating chases. Virtualization, honey pots, software-defined network configuration changes, and systems set up to act as mineshaft canaries, can be used to bring frustration and confusion to the working lives of adversaries.
Third, I shared my thoughts on developing new kinds of metrics designed to reflect changing definitions of security effectiveness. These include heightened ability to measure…
Adversarial Dwell Time—Time required to detect an adversary entering a system.
Compromise Speed—Time required for an adversary to perform their mission.
Unmitigated Attack Duration—Time an attack operates before stopping it.
Adversarial Confusion Ratio 1—Ratio of time an adversary appears confused to the total time of an attack.
Adversarial Confusion Ratio 2—Number of incorrect adversary decisions to the number of correct decisions.
Cost Effectiveness Ratios—Cost of protecting an infrastructure and/or service to cost of losses, and cost of protecting an infrastructure to cost of restoring a service.
These proposed metrics probably justify a free-standing blog post in their own right, so stay tuned for that.
In summing up, I described the above methods as steps along the path of building a condition of information superiority over security adversaries. This means knowing more about the infrastructure, services, and users you protect than your adversaries as a precondition for the ability to act effectively.
There’s a lot more that can be said about this, and the more I talk to customers and security practitioners, the more I’m learning and processing to take these concepts further. That alone is one of the factors that makes cybersecurity so fascinating. There’s something new to learn and think about every day.