• Success was benchmarked using a vulnerability-to-host ratio.
• The performance of sites and teams was mapped directly to this ratio.
• Organizationwide ratios were measured from quarter to quarter.
Dramatic improvements were seen after just a few quarters, and NASA concluded that the effort cost about $30 per system annually. As this illustrates, organizations with a good handle on the management of their vulnerabilities can more cost-effectively deal with current threats.
Money Talks
Vulnerability-assessment software, asset-tracking systems and patch-management tools are not just beneficial. They're necessary. But these are still tactical solutions to a very strategic problem: the staggering number of vulnerabilities in mainstream OSs, services and applications. The answer--as any hardened security professional will tell you--is in the hands of developers. The world needs better software engineering, or we'll never escape today's patch-and-pray model.
But better software engineering is not going to come overnight, and it's certainly not something the enterprise customer can control. So what can be done? Start making security a key factor in product purchasing and deployment decisions, and seek out products with better security track records. ICAT is a good place for the motivated consumer to start, but those who subscribe to SAC, Bugtraq or VulnWatch will also be able to identify common offenders.
Philosophical debates aside, products that have better vulnerability track records, and thus have required less patching, not only reduce an organization's day-to-day risk, they also reduce the TCO (total cost of ownership) via lower support costs. For example, consider choosing Apache over IIS, Postfix instead of Sendmail, Tinydns instead of Bind, and Opera instead of IE. Obviously, these are not isolated decisions; security and patching requirements are just one point among many criteria. But it's a factor that is growing in importance.
If consumers begin factoring security track records into their purchasing decisions, it will send a clear message to vendors: Those that design and implement more secure products will be rewarded, and those that don't will be penalized. Think of how much easier and cheaper it would be to operate a computing environment that didn't require patching.
What Lies Ahead
Unfortunately, we suspect that the vulnerability landscape will get worse before it gets better. Over the past few years the number of discovered vulnerabilities in commercial products has risen dramatically.
Complicating matters, the management lines between application development and system, network and database administration continue to blur, particularly in regard to zones of security control. For example, administration and security of the OS (and the subsequent patching) still clearly falls under the jurisdiction of the system administrator, but his or her security efforts can be completely foiled by a single bad application; if an application developer places a vulnerable CGI form on a previously secure Web server, much of the system administrator's security controls may be bypassed. The network administrator can't be held responsible for insecure systems, and the security-conscious application developer can still be thwarted by a careless database administrator. The dependencies among administrative teams are growing ever more web-like, and these areas of authoritative haze will only be complicated by the adoption of new technologies, such as Web services.
Most organizations have two choices: Continue what they are doing and continue operating with large risk/exposure profiles, or invest in more mature vulnerability-management efforts. Those efforts must include both the tools and processes to quickly and effectively identify, and respond to, an ever-evolving set of threats. Forward-thinking organizations will not only build out better vulnerability-management systems, they will also become more security-conscious in their purchasing decisions. The safety of their data, and their businesses, depends on it.
Greg Shipley is the CTO for Chicago-based security consultancy Neohapsis. Write to him at gshipley@neohapsis.com.
Post a comment or question on this story.
RSS
