continuous monitoring: diagnostics & mitigation [email protected] october 24, 2012

23
Continuous Monitoring: Diagnostics & Mitigation [email protected] October 24, 2012

Upload: chester-powell

Post on 29-Dec-2015

217 views

Category:

Documents


0 download

TRANSCRIPT

Continuous Monitoring: Diagnostics & Mitigation

[email protected] 24, 2012

2

CXOs are accountable for IT security

BUT.

directly supervise onlya small part of the

systems actually in use.OBS

TACL

E

3

Nature of Attacks

80% of attacks leverage known vulnerabilities and

configuration management setting weaknesses

4

LOWERING RISK ACHIEVED BY:• Correcting for “tunnel vision” seen

in physiological studies of pilots

• Using math and statistics to accelerate corrective action

• Adapting market economics to daily risk calculation/priorities

• Automated patch distribution

5

WHILE NOT CHANGING:• Structure of departments or agencies• Decentralized technology management • Structure of security program

RATHER: Focus on Return on Investment (ROI) Integrate cyber security, operations, and top

to bottom work force decisions

6

“Attack Readiness” .

• What time is spent on• Faster action =

lower potential risk

7

Organizations, Major SystemsContractor Performance

8

AddressingInformation Overload

List Dominant Percentages of Risk

9

Results First 12 Months

0.0

200.0

400.0

600.0

800.0

1,000.0

1,200.0

6/1/2008 7/21/2008 9/9/2008 10/29/2008 12/18/2008 2/6/2009 3/28/2009 5/17/2009 7/6/2009 8/25/2009

Domestic Sites

Foreign Sites

89% Reduction

90% Reduction

Personal Computers and Servers

10

1/3 of Remaining Risk Removed

7/15/2

009

7/31/2

009

8/16/2

009

9/1/2

009

9/17/2

009

10/3/2

009

10/19/2

009

11/4/2

009

11/20/2

009

12/6/2

009

12/22/2

009

1/7/2

010

1/23/2

010

2/8/2

010

2/24/2

010

3/12/2

010

3/28/2

010

4/13/2

010

4/29/2

010

5/15/2

010

5/31/2

010

6/16/2

010

7/2/2

0100

20

40

60

80

100

120

140

160

180

200

DomesticForeign

[Year 2: PC’s/Servers]

11

4/1/2010 5/1/2010 6/1/2010 7/1/2010 8/1/2010 8/31/2010 10/1/20100

20

40

60

80

100

120

140

160

180

200

DomesticPolynomial (Domestic)OverseasPolynomial (Overseas)

Time

Risk

Poi

nts

whe

re 1

0 Po

ints

= 1

maj

or V

ulne

rabi

lity

per

mac

hine [Year 2: PC’s/Servers]

12

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

MS10-042 – August 2010Percent of applicable devices patched

Expected Value (Based on all reporting machines)

Lower Bound (Assumes all non-reporting machines are non-compliant)

Efficiency is Repeatable & Sustained

.

when charging 40 points0 - 84% in seven (7) days0 - 93% in 30 days

13

Case Study Results• 89% reduction in risk after 12 months

– personal computers & servers• Mobilizing to patch worst IT security risks first

– Mitigation across 24 time zones – Patch coverage 84% in 7 days; 93% in 30 days

• Outcome: – Timely, targeted, prioritized information– Actionable– Increased return on investment compared to an

earlier implementation of FISMA

14

Lessons Learned• When continuous monitoring augments

snapshots required by FISMA:– Mobilizing to lower risk is feasible & fast (11 mo)– Changes in 24 time zones with no direct contact– Cost: 15 FTE above technical management base

• This approach leverages the wider workforce• Security culture gains are grounded in

fairness, commitment and personal accountability for improvement

Development Phase

Federal CIO and CISO Cyber Goals

• Protect information assets of the US gov’t– Availability, integrity and confidentiality

• Lower operational risk and exploitation of– national security systems

– .gov networks, major systems & cloud services• Increase situational awareness of cyber status• Improve ROI of federal cyber investments • Fulfill FISMA mandates

20 C

ritica

l Con

trol

s1. Inventory of Authorized and Unauthorized Devices 2. Inventory of Authorized and Unauthorized Software 3. Secure Configurations for Hardware and Software on Laptops,

Workstations, and Servers 4. Continuous Vulnerability Assessment and Remediation5. Malware Defenses 6. Application Software Security 7. Wireless Device Control 8. Data Recovery Capability (validated manually)9. Security Skills Assessment and Appropriate Training to Fill Gaps (validated manually) 10. Secure Configurations for Network Devices such as Firewalls, Routers, and Switches 11. Limitation and Control of Network Ports, Protocols, and Services 12. Controlled Use of Administrative Privileges 13. Boundary Defense 14. Maintenance, Monitoring, and Analysis of Security Audit Logs 15. Controlled Access Based on the Need to Know 16. Account Monitoring and Control 17. Data Loss Prevention 18. Incident Response Capability (validated manually)19. Secure Network Engineering (validated manually)20. Penetration Tests and Red Team Exercises (validated manually)

Continuous Diagnosis and Mitigation (CDM)“Full Operational Capability” (FOC) / Desired State:

• Minimum Time to FOC for CDM: 3 years;• CDM Covers 80-100% of 800-53 controls;• Smaller attack surface/“risk” for .gov systems;• Weaknesses are found and fixed much faster;• Replaces much 800-53 assessment work ($440M)

– And most of the POA&M process ($1.05 B)

• Risk scores reflect: threat, vulnerability and impact – Used to make clear, informed risk-acceptance decisions

• Economies reduce total cost yet improve security.

Selection of First Year Priorities

• Implement CMWG focus areas for controls– NSA and CMWG collaboration put in pilots– Complete baseline survey of highest D/A risks

• Award task orders for sensors and services tailored to agency needs and risk profile

• Connect initial controls to dashboard– HW/SW asset management/white listing;

vulnerability; configuration settings; anti-malware

Use of DHS Appropriated Funds

• Strategic Sourcing to buy– Sensors (where missing)– A Federal Dashboard– Services to operate the sensors and dashboard in

the D/As• Labor to mentor and train D/As to use the

dashboard to reduce risk efficiently• Processes to support CMWG (continuous C&A)

Stakeholder Consultation

• DHS and CMWG will consult on program direction and reflect stakeholder concerns of:– CIO Council/ISIMC, ISPAB– NSS, EOP, NIST, NSA– D/As and components– Industry– FFRDCs– Others

22

Continuous Monitoring (CM) Contract Element

Beneficiary for FY13 Networks & COTS CM

Software ($202M)

Tools /Services as options for internal

use

Use diagnostic standards

but may or may not purchase

1. DashboardDHS pays for all government

Department and Agencies Security reporting

to Cyber Scope

Can Purchase off of federal contract:•.mil, Defense Industrial Base;• others who use federal $;• plus State, local gov’t

Cloud Service providers for direct support of government dedicated cloud clients with

cost embedded.CSP ‘s could buy dashboard.

2. Continuous Monitoring Tool Bundles

(Multiple Award)

DHS Pays for initial .gov Agencies

& Departments who choose diagnostic capabilities

Can Purchase off of federal contract:•.mil, Defense Industrial Base;• others who use federal $;• plus State, local gov’t

Cloud Service Providers offer direct support of government dedicated cloud clients with

cyber testing cost embedded.CSP ‘s could buy tools.

3. Continuous Monitoring as a Service (CMaaS)

DHS pays for initial .gov Agencies & Department who

may choose a diagnostic service provider

Department & Agencies (or others) pay for custom systems CM using internal C&A report money (diagnostics and feeds to Cyber Scope)

Department & Agency custom systems using

internal funds.CSP ‘s could buy CMaaS for use as 3rd party Assessors.

4. Continuous monitoring data integration

DHS pays to prepare .gov diagnostic reports &

CyberScope feeds

Department & Agencies (or others) pay Using DHS

published standards using internal funds

Using DHS published standards using internal

funds

[email protected]

$440

M/y

r

Cloud

1

2,5

3, 10, 114

9

18, 8

Change to “plan for events” and “respond to events”.

12, 15

13, 17

14, 16, 20

19

6,7 are assets:They require all the other capabilities applied to them. For an application, it’s HW, SW etc. must be managed, inside a boundary, configured and relatively free of vulnerabilities.

One delta would be the extra analysis SW needs

pre-operations.

a

bcde

i

h

g

jk

l mn

o

p

P1

P1

P1

P1P2

P2

P2

P2

P2

P3

P3

P3

P3/4

P3/4

P3/4