Security Metrics What Can We Measure

Download as pdf or txt
Download as pdf or txt
You are on page 1of 19

Security Metrics

What Can We Measure?

Zed Abbadi
The Public Company Accounting
Oversight Board
What is a “Metric”
† A metric is a system of
related measures enabling
quantification of some
characteristic. A measure is
a dimension compared
against a standard.*
† Security metric is a system
of related dimensions
(compared against a
standard) enabling
quantification of the degree
of freedom from possibility
of suffering damage or loss
from malicious attack.*
*QoP’ 06, Oct. 2006
Do We Really Need Metrics?
"If you cannot measure it, you cannot
improve it."
"In physical science the first essential step in the
direction of learning any subject is to find
principles of numerical reckoning and practicable
methods for measuring some quality connected
with it. I often say that when you can measure Lord Kelvin
what you are speaking about, and express it in
numbers, you know something about it; but when
you cannot measure it, when you cannot express
it in numbers, your knowledge is of a meagre and
unsatisfactory kind; it may be the beginning of
knowledge, but you have scarcely in your
thoughts advanced to the state of Science,
whatever the matter may be." [PLA, vol. 1,
"Electrical Units of Measurement", 1883-05-03]
“You cannot manage what you cannot measure"
Drivers For Metrics
† Money matters
„ Asset vs. liability
† Governance
† You claim it is a
science?
„ Do as good as math,
physics and astrology
† Decision aid
† How are we doing with
security
Good Metrics. vs. Metrics
† Quantitative
† Objective
† Based on a formal model
† Has a time dimension
† Universally acceptable
† Has ground truth
† Inexpensive
† Obtainable
† Repeatable
Data Collection
† Vulnerabilities,
exploits and attacks
† Organization vs.
industry vs. everyone
else
† Disclosure Policies
† Accuracy
† Statistical Significance
Attempts at Measuring Security
† TCSEC (Orange book)
† ITSEC (Europe’s
Orange book)
† CTCPEC (Canada’s
Orange book)
† Common Criteria
(everyone’s Orange
book)
„ Framework rather than
a list of requirements
† SSE-CMM
† NIST FIPS-140 series
† NIST SP 800-55
Security Metrics Types
† Process Security Metrics
† Network Security Metrics
† Software Security Metrics
† People Security Metrics
† Other
Process Security Metrics
† Measure processes and procedures
† Imply high utility of security
policies and processes
† Relationship between metrics and
level of security not clearly defined
† Compliance/Governance driven
† Generally support better security
† Actual impact hard to define
Examples
† No. of Policy Violations
† % of systems with formal risk
assessments
† % of system with tested security
controls
† % of weak passwords (non-
compliant)
† No. of identified risks and their
severity
† % of systems with contingency plans
Network Security Metrics
† Driven by products
(firewalls, IDS etc)
† Readily available
† Widely used
† Gives sense of
control
† Nice charts and
interfaces
† Can be misleading
Examples
† Successful/unsuccessful
logons
† No. of incidents
† No. of viruses blocked
† No. of patches applied
† No. of spam blocked
† No. of virus infections
† No. of port probes
† Traffic analysis
Software Security Metrics
† Software measures are
troublesome (LOC, FPs,
Complexity etc)
† “Laws of Physics” are
missing
† Metrics are context
sensitive and
environment-dependent
† Architecture dependent
† Aggregation may not
lead to strength
Examples
† Size and complexity
† Defects/LOC
† Defects (severity,
type) over time
† Cost per defect
† Attack surface (# of
interfaces)
† Layers of security
† Design Flaws
People Security Metrics
† Relevance
† Unique
characteristics
„ Risk perception
skewed “optimism
Bias”
„ Limited memory and
attention span
„ Behavior modeling is
difficult
† Awareness training?
Reliability vs. Security
† Similar but different
„ We care more about
reliability
„ Different adversary
model
† Reliability models
exist, but…
† Security is a moving
target
Most Common Security Metric
† Risk- We love this thing!
„ Source for profit
† Where is the data?
„ Non monetary consequences
„ Adversary behavior models
„ Accuracy against ground truth
„ Mission/system/support models
† Dynamic in nature
Future Of Security Metrics
† Consumers demand better
security metrics
† Government involvement is
increased
† Science evolves to provide
better measures
† Vendors volunteer (forced
to) develop universal accurate
metrics
† Some vendors cheat, a
watchdog is created
† Security problems continue,
no change in level of risk

You might also like