Sharpening Your Security Intelligence Ax with Big Data
Dale Carnegie reportedly told a story once of two woodsmen who were spending the day chopping down trees. The first man worked hard ’round the clock chopping away, taking no breaks, and only took a short lunch. The second man took many breaks during the day, and even stopped once to take a nap. At the end of the day, the first man was quite dismayed to discover that the second man had out-chopped him by a wide margin. “I don’t understand, every time I looked over at you you were taking a break – how did you chop more wood than I did”, he said. The second man smiled at him and asked, “did you also notice that while I was sitting down, I was sharpening my ax?
If you are the typical enterprise, you potentially have tens of thousands of software vulnerabilities spread across thousands and thousands of machines in your network. You also know full well that patching potentially means expense and downtime. Looking at this forest of things that need to be “chopped down” can be daunting for even the most experienced woodsman. Which ones are the ones that need to be patched yesterday, and why, and how are you ever going to get on top of this? Which things are relevant to your business right now, and which can be addressed later?
This is where the need for a vulnerability management tool begins to intersect with the need for a Security Intelligence platform. The more intelligent information something can weigh into this decision making process, the sharper your “ax” will become.
First, we have the obvious questions, the ones that take the ax from dull to straight.
- What is the required security and compliance posture of the host? Is it on an isolated development test network, or is it sitting in your DMZ, or is it sitting in your PCI compliant data center?
- What is the overall severity of the vulnerability? Is it remotely exploitable or does it require local access?
Next, we have the less obvious questions, the ones that the ax from straight to sharp.
- Is the vulnerability relevant to this asset? A zero-day vulnerability in a web browser on a data server that has never before had outbound web traffic is not as high of a patching priority as one on a client machine that has such traffic on a daily basis.
- Is the vulnerability actively exploitable from the Internet, or from your DMZ, or do you have firewall rules or IPS signatures in place that block this vulnerability before it even reaches the host?
- Does this asset present a threat to other assets? Which critical assets or services does it have access to that might also be threatened should this vulnerability be exploited?
If you have a platform that can take all of this into account (such as the IBM Security QRadar platform with QRadar Vulnerability Manager and Risk Manager), your ax is already pretty sharp indeed. But a question that you should ask is, can your ax become a razor? Are there are other pieces of data that already exist in your company, that you might be able to bring to bear as well? This is where Big Data (from a platform such as IBM InfoSphere Big Insights), when closely integrated with a Security Intelligence solution, can make a real grindstone for your ax.
The possibilities for using existing data to enhance your vulnerability prioritization are immense, but here are a couple of such examples:
- Assume your security ax can also access information your employment database, and know which user accounts are valid and which ones should not be valid. Have there been suspicious logins using accounts that should have been disabled on this host?
- Suppose you analyze DNS lookups at your company, and can correlate domain registrant information with other domains known to host malware. Your Security Intelligence platform should already know which hosts have communicated with those domains in the past, and what vulnerabilities those hosts have. By combining these three pieces of information, you now have a list of hosts that have vulnerabilities which should probably be the top of any priority list to address.
- Next add a history of links embedded in email messages to users. Combined with information regarding malware hosts and your Security Intelligence platform’s knowledge of user activity, you can begin to address risky users and concentrate on their assets.
The best way to deal with any challenge that seems insurmountable is to take in as much information as possible and create a plan of attack. Trying to tackle vulnerabilities across an organization without such a properly prioritized plan can quickly become a lost cause. The more information you can take into account, the better.
To conclude, let me wrap it up with another ax-inspired quote, one often attributed to Abraham Lincoln.[testimonial author=”Abraham Lincoln”]If I had 60 minutes to cut down a tree, I would spend 40 minutes sharpening the ax and 20 minutes cutting it down.[/testimonial]
Let’s be more like Lincoln, and less like the man in the first story who took no breaks.