Rational Survivability

This is Part I of a two-part series on a topic for which I coined the phrase "Unified Risk Management"
The second part of this paper will be out shortly.   You can download this paper as a .PDF from here

NOTE: This is a little long for a blog post, but it should make for an interesting read.

Abstract

Managing risk is fast becoming a lost art. As the pace of technology’s evolution and
adoption overtakes our ability to assess and manage its impact on the business,
the overrun has created massive governance and operational gaps resulting in
exposure and misalignment. This has
caused organizations to lose focus on the things that matter most: the
survivability and ultimate growth of the business.

Overwhelmed with the escalation of increasingly complex
threats, the alarming ubiquity of vulnerable systems and the constant onslaught
of rapidly evolving exploits, security practitioners are forced to choose the
unending grind of tactical practices – focused on deploying and managing
security infrastructure –  over the
strategic art of managing and institutionalizing risk-driven architecture as a business
process.

In order to understand the nature of this problem and its
resolution we have separated this discussion into two separate papers:

· In Part One (this paper), we analyze the gap between
pure technology-focused information security infrastructure and
business-driven, risk-focused information survivability
architectures.

· In Part Two (a second paper), we show how this
gap is bridged using sound risk management practices in conjunction with best
of breed consolidated Unified Threat Management (UTM) solutions as the
technology foundation of a consolidated risk management model. We will also
show how governance organizations, business stakeholders, network and security
teams can harmonize their efforts to produce a true business protection and
enablement strategy that delivers security as an on-demand service layer at the
speed of business. This is a process we
call Unified Risk Management or URM.

The Way Things Are

Today’s constantly expanding chain of technically-complex security
point solutions do not necessarily reduce or effectively manage risk; they
mitigate threats and vulnerabilities in the form of products produced by
vendors to solve specific technical problems but without context for the assets
which they are tasked to protect and at a cost that may outweigh the protected
assets’ value.

But how does one go about defining and measuring risk?

Spire Security’s Pete Lindstrom best defines being able to
measure and manage risk by first describing what it is not:

· Risk is not static; it is dynamic and fluctuates
constantly with potentially high degrees of variation.

· Risk is not about the possibility that something
bad could happen; it is about the probability that it might happen.

· Risk is not some pie-in-the-sky academic
exercise; you have all of the necessary information available to you today.

· Risk is not a vague, ambiguous concept; it is a
continuum along which you can plot many levels of tolerance and aversion.

It is clear that based upon research available today, most
organizations experience difficulty aligning threats, vulnerabilities and
controls to derive the security posture of the organization (defined as
acceptable or not by the business itself.) In fact, much of what is referred to as risk management today is
actually just complex math in disguise indicating an even more complex
extrapolation of meaningless data that drives technology purchases and
deployments based upon fear, uncertainty and doubt. Nothing sells security like a breach or new worm.

As such, security practitioners are typically forced into
polarizing decision cycles based almost exclusively on threat and vulnerability
management and not a holistic risk management approach to deploying security as
a service. They are distracted by the
market battles to claim the right to the throne of Network Security Supremacy
to the point where the equipment and methodology used to fight the war has
become more attractive than the battle itself.

In most cases, these security products are positioned as
being either integrated into the network infrastructure such as routers or
switches or bolted onto it in the form of single vendor security suite
appliances. These products typically do
not collaborate, interoperate, communicate or coordinate their defensive
activities with solutions not of a like kind.

Realistically, there is room for everyone at the
table. Network vendors see an
opportunity to continue to leverage their hold on market share by adding value
in the form of security while pure-play security vendors continue to innovate
and bring new products and solutions to market that address acute needs that
the other parties cannot. Both are
needed but for different reasons.

Neither of the extremes represents an ultimate answer. Meeting in the middle is the best answer with
an open, extensible, and scaleable network security reference architecture that
integrates as a network switch with all of the diversity and functionality
delivered by on demand best of breed security functions.

As the battle rages, multiple layers of overlapping
proprietary technologies are being pressed into service against risks which are
often not quantified, threats that are not recognized and attempt to defend
against vulnerabilities which within context may have little recognized
business impact.

In many cases, these solutions are marketed as new
technology when in fact they exist as re-badged products with additional
functions cobbled together onto outdated or commoditized hardware and software
platforms, polished up and marketed as UTM or adaptive security solutions.

It is important to make clear the definition of UTM within
the context of the mainstream security solution space offered by most vendors
today. UTM solutions are those which provide an aggregate of security
functionality comprised of at least network firewall, network intrusion
detection and prevention, and
gateway anti-virus. UTM solutions are
often extended to offer additional functionality such as VPN, URL filtering,
and anti-spam capabilities with a recognized benefit of squeezing as much
functionality from a single product offering in order to maximize the
investment and minimize the number of arterial insertion points throughout the
network.

Most of the UTM solutions on the market today provide a
single management interface which governs the overall operation of many
obfuscated moving parts which deliver the functionality advertised above.

In many cases, however, there are numerous operational and
functional compromises made when deploying typical single application/multiple
function appliances or embedded security extensions applied to routers and
switches. These compromises range from
poor performance to an inability to scale based on emerging functionality or
performance requirements. The result is what some hope is “good enough” and
implies a tradeoff favoring cost over security.

Unfortunately, this model of “good enough” security is
proving itself not good enough as these solutions can lead to cost and
management complexities that become a larger problem than the perceived threat
and vulnerabilities the solutions were designed to mitigate in the first place.

So what to do? Focus
on risk!

Prudent risk management strategy dictates that the best
method of securing an organization’s most critical assets is the rational
application of policy, technology and processes where ultimately the risk
justifies the cost.

It is within this context that the definition of
information survivability demands an introduction as it bears directly on the
risk management processes described in this paper. In their paper titled “Information
Survivability: Required Shifts in Perspective,” Allen and Sledge introduce the
concept of information survivability as a discipline which is defined as “…the
capability of a system to fulfill its mission, in a timely manner, in the
presence of attacks, failures, or accidents.”

They further juxtapose information survivability against
information security by illustrating that information security “…takes a
technology centric point of view, with each technology solving a specific set
of issues and concerns that are generally separate and distinct from one
another. Survivability takes a broader,
more enterprise-wide point of view looking at solutions that are more pervasive
than point-solution oriented.”

Information survivability thus combines elements of
business impact, continuity, contingency and disaster recovery planning with
the more narrowly-focused and technical information security practices, thereby
elevating the combined foundational elements to an enterprise-wide risk
management concern.

From this perspective, risk management is not just about
the latest threat. It is not just about
the latest vulnerability or its exploit. It is about how, within the context of the continued operation of the
business and even while under duress, the organization’s mission-critical
functions will be sustained and the most important data will be appropriately
protected.

The language of risk

One obvious illustration of this risk gap is how
disconnected today’s enterprise security and networking staffs remain even when
their business interests should be so very much closely aligned. Worse yet is the resultant misalignment of
both teams with the enterprises’ mission and appetite for risk.

As an example, while risk analysis is conducted on one side
of the house with little understanding of the network and all its moving parts,
the device sprinkling of network and security appliances are strung together on
the other side of the house with little understanding of how these solutions
will affect risk or if they align to the objectives or matters to the business
at all.

To prove this point, ask your network team if they know
what OCTAVE or CoBIT frameworks are and how current operational security
practices map to either of them. Then, ask the security team if they know how MPLS
VRF, BGP route reflectors or the spanning tree protocol function at the network
level and how these technologies might affect the enterprise’s risk posture. 

Then, ask representative business stakeholders if they can
articulate how the answers given by either of the parties clearly maps to their
revenue goals for the year and how their regulatory compliance requirements may
be affected. Where are the metrics to
support any assertion?

Thus, while both parties seek to serve the business with a
common goal of balancing security with connectivity neither speaks a common
language that can be used to articulate the motivation, governance or value of
each other’s actions to the business.

At the level of network security integration, can either
team describe the mapping of asset-based risk categories across the enterprise
to the network infrastructure? Can they tell you tomorrow what the new gaps are
at each risk category level and provide a quantifiable risk measurement across the
enterprise of the most critical assets in a matter of minutes?

This illustration defines the problem at hand; how do we
make sure that we deliver exactly what the business requires to protect the
most critical assets in a manner fitting the risk profile of the organization
and no more.

Interestingly, from an economic point of view, the failure
to create a tightly integrated risk management ecosystem results almost by
definition in a completely inefficient and ineffective solution. Without risk
management basics such as asset and data classification and zoned network
segmentation by asset class, the network has the very real potential to actually
be over-defended at risk boundaries and thus drive costs and complexity much
higher than they need to be.

Consequently, most, if not all, security controls and
prescribed protective technologies are applied somewhat indiscriminately across
the enterprise as a whole. Either too
much security is applied or many of the features of the solution are disabled
since they are not needed. Where is the
return on investment there? Do you need
URL filtering in a DMZ? Do you need
SOA/XML schema enforcement applied across user desktops? No. So
why deploy complex blanketed security technology where it is neither needed nor
justified?

For example, since all assets and the data they contain are
not created equal, it is safe to assume that the impact to the business caused
by something “bad” happening to any two assets of different criticality would
also not be equal. If this is an
accepted corollary, does it make sense to deploy solutions that provide
indiscriminant protective umbrellas over assets that may not need any
protection at all?

In many cases, this issue also plays out in a different
direction as security architectures are constrained based on the deployment of
the physical wiring closets and switch and router infrastructures. Here, the ability or willingness to add one
after the other of point solution devices in-line between key network arteries,
incrementally add specialized security blades into core network components or
even forklift switching and routing infrastructure to provide for “integrated
security” is hideously problematic.

In these cases, overly-complex solutions consist of devices
sprinkled in every wiring closet because there will probably be a
representative computing resource of every risk category in that area of the
network.

Here we are being asked to change the network to fit the
security model rather than the other way around. If the network was built to accommodate the
applications and data that traverse it, should we not be just as nimble, agile
and accommodating in our ability to defend it?

Referring back to the definition of risk management, the
prudent answer is to understand exactly where you are at risk, why, the
business impact, and exactly what is needed from a control perspective to
appropriately manage the risk. In some
cases the choice may be to assert no control at all based upon the lack of
business impact to the organization.

One might ask if the situation is not better than it was
five years ago. The answer to this question is unclear – the effects of the
more visible and noisy threats such as script kiddies have been greatly
mitigated. On the other hand, the emergence of below-the-radar,
surgically-focused, financially motivated cyber-criminals has exposed business
assets and data more than ever. The net effect is that we are not, in fact,
safer than we were because we focus only on threats and vulnerabilities and not
risk.

Security is in the network…or is it in the appliance over
there?

Let us look for a moment at how technology visions spiral
out of control when decoupled from risk in a technology centric perspective. The most
blatant example is the promise of security embedded in the network or
all-in-one single vendor appliances.

On the one hand, we are promised a technically-enlightened,
self-defending network that is resilient to attack, repels intruders,
self-heals when infected and delivers security as a service as applications and
data move about fluidly pursuant to policies enforced across every platform and
network denizen.

We also are told to expect intelligent networks that offer
solution heterogeneity irrespective of operating system or access modality,
technology agnosticism, and completely integrated identity management as a way
to evolve from being data rich but information poor, providing autonomic
response when bad things happen.

Purveyors of routing and switching products plan to branch
out from the port density penetration
foothold they currently enjoy to deliver end-to-end security functionality
embedded into the very fabric of the machinery meant to move bits with the
security, reliability and speed it deserves and which the business demands.

At the other end of the spectrum, vendors who offer
single-sourced, proprietary security suites utilizing integrated functions by
way of appliances integrated into the network suggest that they will provide
the architecture of the future.

They both suggest they will provide host-based agents that
provide immune system-like responses to attempted “infection” and will take
their orders from a central networked “nervous system” that coordinates the
activities of the various security “organs” across the zones of trust defined
by policy.

They propose the evolution of the network into a sentient
platform for the delivery of business in all its forms, aware of and able to
interact with and control the applications and data which travel over it.

Data, voice, video and mobility with all of the challenges
posed by the ubiquity of access methodologies – and of course security – are to
be provided by the network platform as the launch pad for every conceivable
level of service. The network will take the place of complex business logic
such as Extraction/Transform/Load (ETL) layers and it will deliver applications
directly and commit and retrieve data dynamically and ultimately replace tiers
of highly-specialized functions and infrastructure that exist today.

All the while, as revolutionary technology and
architectures such as web services emerge, new standards compete for relevancy
and the constant demand for increased speeds and feeds continue to evolve, the
network will have to magically scale both in performance and functionality to
absorb this change while the transparency of applications, data and access
modality blurs.

These vendors claim that security will simply be subsumed
by the “network” as a function of the delivery of the service since the
applications and data will be provided by a network platform completely aware
of that which traverses its paths. It
will be able to apply clearly articulated business processes and eliminate
complex security problems by mitigating threats and vulnerabilities before they
exploit an attack surface.

These solutions are to be “open,” and allow for
collaboration across the enterprise, protecting heterogeneous elements up and
down the stack in a cooperative defense against impact to the delivery of
applications and data.

These solutions promise to be more nimble and will be
engineered to provide adaptive security capabilities in software with hardware
assist in order to keep pace with exponential increases in requirements. These solutions will allow for quick and easy
update as threats and vulnerabilities evolve. They will provide more deployment flexibility and allow for greater
coverage and value for the security dollar as policy-driven security is applied
across the enterprise.

What’s Wrong with These Answers? Mr. Fox, meet Ms. Chicken

Today’s favorite analogy for security is offered in direct
comparison to the human immune system. The immune system of modern man is indeed a remarkable operation. It is there, inside each human being, where
individual organs function independently, innocuously and in an autonomic
fashion. When employed in a coordinated fashion as a consolidated and
cooperative system, these organs are able to fight infection by adapting and
often become more resistant to attack and infection over time.

Networks and networked systems, it is promised, will
provide this same capability to self-defend and recover from infection. Networks of the future are being described as
being able to self-diagnose and self-prescribe antigens to cure their ills, all
the while delivering applications and data transparently and securely to those
who desire it.

It is clear, however, that unfortunately there are
infections that humans do not recover from. The immune system is sometimes overwhelmed by attack from invaders that
adapt faster than it can. Pathogens
spread before detection and activate in an overwhelming fashion before anything
can be done to turn the tide of infection. Mutations occur that were unexpected, unforeseen and previously
unknown. The body is used against itself
as the defense systems attack both attacker and healthy tissue and the patient
is ultimately overcome. These illnesses
are terminal with no cure.

Potent drugs, experimental treatments and radical medical
intervention may certainly extend or prolong life for a short time, but the
victims still die. Their immune systems
fail.

If this analogy is to be realistically adopted as the basis
for information survivability and risk management best practices, then anything
worse than a bad case of the sniffles could potentially cause networks – and
businesses — to wither and die if a more reasonable and measured approach is
not taken regarding what is expendable should the worst occur. Lose a limb or lose a life? What is more important? The autonomic system
can’t make that decision.

These glimpses into the future are still a narrowly-focused
technology endeavor without the intelligence necessary to make business
decisions outside of the context of bits and bytes. Moreover, the deeper and
deeper information security is pushed down into the stack, the less and less
survivable our assets and businesses will become because the security system
cannot operate independently of the organ it is protecting.

Applying indiscriminate and sometimes unnecessary layers of
security is the wrong thing to do. It
adds complexity, drives costs, and makes manageability and transparency second
class citizens.

In both cases, these promises will simply add layer upon
layer of complexity and drive away business transparency and the due care
required to maintain it further and further from those who have the expertise
to manage it. The reality is that either
path will require a subscription to a single vendor’s version of the truth. Despite claims to the contrary, innovation,
collaboration and integration will be subject to that vendor’s interpretation
of the solution space. Core
competencies will be stretched unreasonably and ultimately something will give.

Furthermore, these vendors suggest that they will provide
ubiquitous security across heterogeneous infrastructure by deploying what can
only be described as homogenous security solutions. How can that be? What possible motivation would one vendor
have to protect the infrastructure of his fiercest competitor?

In this case, monoculture parallels also apply to security
and infrastructure the same way in which they do to networked devices and
operating systems. Either of the examples referenced can potentially introduce
operational risk associated with targeted attacks against a single-vendor
sourced infrastructure that provides both the delivery and security for the
data and applications that traverse it. We have already seen recent malicious attacks surgically designed and
targeted to do just this.

What we need is perfectly described by Evan Kaplan of
Aventail who champions the notion of a “dumb” network connectivity layer with
high speed, low latency, high resiliency, predictable throughput and
reliability and an “intelligence” layer which can deliver valued added service
via open, agile and extensible solutions.

In terms of UTM, based upon a sound risk management model,
this would provide exactly the required best of breed security value with
maximum coverage exactly where needed, when needed and at a cost that can be
measured, allocated and applied to most appropriately manage risk.

We pose the question of whether proprietary vendor-driven
threat and vulnerability focused technology solutions truly offer answers to
business problems and if this approach really makes us more secure. More importantly, we call into question the
ability for these offerings to holistically manage risk. We argue they do not and inherently
cannot.

The Solution: Unified Risk Management utilizing Unified
Threat Management

A holistic paradigm for managing risk is possible. This
model is not necessarily new, but the manner in which it is executed is. Best-of-breed, consolidated UTM provides this
execution capability. It applies
solutions from vendors whose core competencies provide the best solution to the
problem at hand. It can be linked
directly to asset and information criticality.

It offers the battle-hardened lessons and wisdom of those
who have practiced before us and adds to their work all of the benefits that
innovation, remarkable technology and the pragmatic application of common sense
brings to the table. The foundation is
already here. It does not require years
of prognostication, massive infrastructure forklifts or clairvoyant bets made
on leveraging futures. It is available
today.

This methodology, which we call Unified Risk Management
(URM), is enabled by applying a well-defined framework of risk management
practices to an open, agile, innovative and collaborative best-of-breed UTM
solution set combined in open delivery platforms which optimize the
effectiveness of deployments in complex network environments.

These tools are combined with common sense and the
extraordinary creativity and practical brilliance of leading-edge risk
management practitioners who have put these tools to work across organizational
boundaries in original and highly effective ways.

This is the true meaning of thought leadership in the high
technology world: customers and vendors working hand-in-hand to create
breakthrough capabilities without expensive equipment forklifts and without the
associated brow-beating from self-professed prophetic visionaries who
pontificate from upon high about how we have all been doing this wrong and how
a completely new upgraded infrastructure designed to sell more boxes and Ethernet
ports is required in order to succeed.

URM is all about common sense. It is about protecting the right things for
the right reasons with the right tools at the right price. It is not a marketecture. It is not a fancy sales pitch. It is the logical evolution and extension of
Unified Threat Management within context.

It is about providing choice from best-of-breed offerings
and proven guidance in order to navigate the multitude of well-intentioned
frameworks and come away with a roadmap that allows for true risk management
irrespective of the logo on the front of the machinery providing the heavy
lifting. It is, quite literally, about
“thinking outside of the box.”

URM combines risk management – asset management, risk
assessment, business impact analysis, exposure risk analytics, vulnerability
management, automated remediation –  and
the virtualization of UTM security solutions as a business process into a tight
feedback loop that allows for the precise management of risk. It iteratively feeds into and out of
reference models like Spire Security’s Pete Lindstrom’s “Four Disciplines of
Security Management” that include elements such as:

· Trust Management

· Identity Management

· Vulnerability Management

· Threat Management

This system creates a continuously iterative and highly
responsive intelligent ecosystem linked directly to the business value of the
protected assets and data.

This information provides rational and defensible metrics
that show value, the reduction of risk on investment, and by communicating
effectively in business terms, is intelligible and visible to all levels of the
management hierarchy from the compliance auditor to the security and network
technicians to the chief executive officer.

This re-invigorated investment in the practical art of risk
management holds revolutionary promise for solving many of today’s business
problems which are sadly mislabeled as information security issues.

Risk management is not rocket science, but it does take innovation,
commitment, creativity, time, the reasonable and measured application of
appropriate business-driven policy, excellent technology and the rational
application of common sense.

This tightly integrated ecosystem consists of solutions
that embody best practices in risk management. It consists of tightly-coupled
and consolidated layers of UTM-based information survivability architectures
that can apply the results of the analytics and management toolsets to
business-driven risk boundaries in minutes. It collapses the complexity of existing architectures dramatically and
applies a holistic policy driven risk posture that meets the security appetite of
the business and it does so while preserving existing investments in routing
and switching infrastructure that serves the business well.

Conclusion: On To the Recipe

In this first part of our two-part series, we have tried to
define the basis for looking at network security architectures and risk
management in an integrated way.  Key to
this understanding is a move away from processes in which disparate appliances
are thrown at threats and vulnerabilities without a rationalized linkage to the
global risk profile of the infrastructure.

In the second paper of the series we will demonstrate
exactly how the lightweight processes that form the foundation of Unified Risk
Management can be implemented and applied to a UTM architecture to create a
highly responsive, real-time enterprise fully aware of the risks to its
business and able to respond on a continual basis in accordance with the ever-changing
risk profile of its critical data, applications and assets.