Visible to the public File preview

Photos placed in horizontal position with even amount of white space between photos and header

Pilo%ng
 a
 Secure
 System
 Design
  Compe%%on
 
Ben
 Cook
 on
 behalf
 of
 the
 FIREAXE
 Team
 
Adam
 Anderson,
 Mitch
 Adair,
 William
 Atkins,
 Alan
 Berryhill,
 Dominic
 Chen,
 Ben
 Cook,
 Jeremy
  Erickson,
 Michael
 Z.
 Lee,
 Steve
 Hurd,
 Ron
 Olsberg,
 Lyndon
 Pierson,
 Owen
 Redwood,
  Yevgeniy
 Vorobeychik
 
Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000. SAND NO. 2012-10175C


 

Origins
 of
 a
 Secure
 System
 Design
  Compe%%on
 for
 Students
 


  § 

April
 2010
 at
 Carl
 Landwehr’s
 Designing
 a
 Secure
 Systems
  Engineering
 Compe%%on
 (DESSEC)
 Workshop
  §  DESSEC’s
 Workforce
 Development
 track
 iden%fied
 needs
 and
  poten%al
 compe%%on
 “specifica%ons”
 
§  Considered
 how
 to
 aYract,
 mo%vate,
 inform,
 and
 educate
 students
 in
 cyber
  security
  §  Acknowledged
 significant
 gap
 in
 secure
 design
 educa%on
  §  Highlighted
 importance
 of
 adversarial
 mindset,
 understanding
 of
 “lore”,
 ability
  to
 convert
 aYack
 knowledge
 to
 robust
 defense,
 and
 confidence
 to
 take
 on
  real-­‐world
 system
 engineering
 problems
  §  Produced
 several
 loosely
 developed
 compe%%on
 ideas:
 Cyber
 Cup,
 Cyber
  Village,
 Cyber
 Scouts,
 Weakest
 Link
 

§  In
 late
 2010
 Doug
 Maughan
 at
 DHS
 S&T
 endorsed
 CCD
 pilot
 
2
 

The
 Compe%%on
 Se]ng:
 Sandia’s
 
  Center
 for
 Cyber
 Defenders
 (CCD)
 
§  CCD
 is
 a
 highly
 selec%ve,
 applied
 research
  internship
 ins%tute
 at
 Sandia’s
 New
 Mexico
  and
 California
 sites
 
§  Hosted
 30
 undergraduate
 and
 graduate
 students
  in
 summer
 2012
 from
 about
 20
 universi%es
  (selected
 from
 over
 300
 applicants)
  §  Offers
 collabora%ve,
 threat-­‐informed,
 project-­‐ based
 internships
 in
 cyber
 security
  §  Contributes
 enabling
 solu%ons
 to
 real
 na%onal
  security
 R&D
 projects
 in
 cyber
 security:
 examples
  include
 control
 system
 modeling,
 network
  situa%onal
 awareness,
 protocol
 analysis,
 digital
  forensics,
 and
 red
 teaming
 
3
 

The
 Compe%%on
 Specifica%on:
  A
 Stylized
 Electronic
 Vo%ng
 System
 
§  Designed
 to
 help
  students
 iden%fy
 and
  internalize
 security
  principles
 
  §  Assumed
 realis%c
 but
  limited
 threat
 model
  §  Clearly
 (or
 so
 we
  thought)
 spelled
 out
  specifica%on
 including
  requirements,
 rules,
  and
 evalua%on
 and
  scoring
 procedure
 

EVS consists of a server (election management system) and client (voting station) with sneakernet USB for data transfer

4
 

Compe%%on
 Structure
 and
 Results
 
 
§  Compe%%on
 structured
 into
 mul%ple
 rounds
 each
 having
 a
  dis%nct
 design
 and
 then
 red
 team
 phase
 
§  “Design
 and
 red
 team”
 itera%ve
 structure
 was
 chosen
 to
 cul%vate
 and
  integrate
 adversarial
 mindset
 into
 an
 evolu%onary
 design
 process
  §  Pilot
 included
 two,
 three-­‐person
 student
 teams
 and
 white
 team
 for
  oversight
  §  Constrained
 red
 teaming
 to
 predefined
 aYack
 scenarios
 with
 either
  user-­‐
 or
 root-­‐level
 access
 

§  Students
 produced
 two
 substan%ally
 different
 designs
 
§  NM
 team
 focused
 on
 customizing
 the
 kernel
 and
 produced
 very
 small,
  highly
 restricted
 OS,
 while
 CA
 team
 implemented
 limited
 user
 shell
  and
 “red
 pill”
  §  Teams
 choose
 different
 development
 plaforms,
 tools,
 and
 crypto
  libraries
 
 
 
 

5
 

Observa%ons
 
§  Students
 improved
 their
 understanding
 of
 and
 ability
 to
  ar%culate
 secure
 design
 principles
 
§  §  §  §  §  Reduce
 aYack
 surface
  Use
 exis%ng
 tools
  Enforce
 policies
 at
 lowest
 level
  Defense
 in
 depth
  Prevent
 easy
 access
 

§  Specifica%on
 of
 an
 effec%ve
 compe%%on
 is
 nontrivial:
 despite
  extensive
 pre-­‐work,
 numerous
 ambigui%es
 surfaced
 and
  unan%cipated
 issues
 arose
  §  Compe%%ons
 are
 great
 mo%vator…
 this
 was
 billed
 as
 research
  project
 but
 students
 were
 quick
 to
 forget
 
6
 

What’s
 Ahead?
 
§  Scale
 and
 sustain
 compe%%on
 
§  Need
 to
 automate
 evalua%on
 and
 find
 partners
 

§  Experiment
 with
 different
 formats
 and
 themes,
 e.g.,
 
§  Could
 compe%%on
 be
 used
 to
 familiarize
 students
 with
 new
  technologies
 and
 accelerate
 adop%on?
  §  What
 are
 realis%c
 expecta%ons
 with
 respect
 to
 innova%on
 versus
  educa%on?
 

Gratefully
 acknowledge
 DHS
 S&T
 –
 Doug
 Maughan
 and
 
  Ed
 Rhyne
 –
 for
 their
 funding
 and
 encouragement
 
  and
  Special
 thanks
 to
 Jeremy
 Epstein
 and
 Carl
 Landwehr
 for
 
  their
 technical
 guidance
 

 
7
 


 

More
 info…
 
DESSEC
 Workshop
 Report
 produced
 by
 I3P
 
www.thei3p.org/docs/publicaIons/410.pdf
 

ACSAC
 Poster
 Session
 next
 week
  Upcoming
 CSIIRW
 presenta%on
 and
 paper,
 
 
FIREAXE:
 The
 DHS
 Secure
 Design
 CompeIIon
 Pilot
 


  Connect
 with
 Sandia’s
 technical
 leads:
 
Eugene
 Vorobeychik,
 yvorobe@sandia.gov
  Will
 Atkins,
 wdatkin@sandia.gov
 
 
 
 
 
 

8