Efficient Content Verification in Nam ed Data Netw orking 2015. 10. - - PowerPoint PPT Presentation

2nd ACM Conference on Information-Centric Networking in San Francisco

Efficient Content Verification in Nam ed Data Netw orking

• 2015. 10. 2.

Dohyung Kim1, Sunwook Nam2, Jun Bi3, Ikjun Yeom1 mr.dhkim@gmail.com

1Sungkyunkwan University 2Korea Financial Telecommunications and Clearing Institute 3singhua University

2nd ACM Conference on Information-Centric Networking in San Francisco

Nam ed Data Netw orking ( NDN)

• Name-based consumer-driven content delivery

Request to Where  Request for What

I nternet

R2 R1

2nd ACM Conference on Information-Centric Networking in San Francisco

Nam ed Data Netw orking ( NDN)

• Name-based consumer-driven content delivery

Request to Where  Request for What

I nternet

R2 R1

PIT entry is created at routers

Interest

2nd ACM Conference on Information-Centric Networking in San Francisco

Nam ed Data Netw orking ( NDN)

• Name-based consumer-driven content delivery

Request to Where  Request for What

I nternet

R2 R1

• PIT-based Content Delivery
• In-network Caching

Response

2nd ACM Conference on Information-Centric Networking in San Francisco

Nam ed Data Netw orking ( NDN)

• Name-based consumer-driven content delivery

Request to Where  Request for What

I nternet

R2 R1

Interest

Content is served from in-network cache

2nd ACM Conference on Information-Centric Networking in San Francisco

Secure Com m unication

• In IP networks

R1 R2

2nd ACM Conference on Information-Centric Networking in San Francisco

Secure Com m unication

• In IP networks

End-to-end secure channel

R1 R2

2nd ACM Conference on Information-Centric Networking in San Francisco

Secure Com m unication

• In IP networks
• In NDN

End-to-end secure channel

R1 R2 R1 R2

2nd ACM Conference on Information-Centric Networking in San Francisco

Secure Com m unication

• In IP networks
• In NDN

End-to-end secure channel

R1 R2 R1 R2

Content itself should be secure

2nd ACM Conference on Information-Centric Networking in San Francisco

• Fabricated content is placed in the content store

 Router compromise

Content Poisoning Attack

2nd ACM Conference on Information-Centric Networking in San Francisco

• Fabricated content is placed in the content store

 Router compromise  Injection from attackers’ server

Content Poisoning Attack

2nd ACM Conference on Information-Centric Networking in San Francisco

• Fabricated content is placed in the content store

 Router compromise  Injection from attackers’ server

Content Poisoning Attack

2nd ACM Conference on Information-Centric Networking in San Francisco

Content Poisoning Attack

• Distribution of the fabricated content

I nternet

R2 R1

Poisoned content

2nd ACM Conference on Information-Centric Networking in San Francisco

Content Poisoning Attack

• Distribution of the fabricated content

I nternet

R2 R1

Poisoned content

Interest

2nd ACM Conference on Information-Centric Networking in San Francisco

Content Poisoning Attack

• Distribution of the fabricated content

I ntenet

R2 R1

• Poisoned content is distributed

by the system itself

2nd ACM Conference on Information-Centric Networking in San Francisco

Content Poisoning Attack

• Distribution of the fabricated content

I ntenet

R2 R1

• Poisoned content is distributed

by the system itself

• Users are separated from

valid content sources

Interest

Not forwarded

Poisoned response

2nd ACM Conference on Information-Centric Networking in San Francisco

NDN Content Verification

Signature verification

2nd ACM Conference on Information-Centric Networking in San Francisco

NDN Content Verification

Signature verification incurs huge computational overhead

2nd ACM Conference on Information-Centric Networking in San Francisco

Related W ork

• Probabilistic caching
• Bianchi, Giuseppe, et al. "Check before storing: What is the

performance price of content integrity verification in LRU caching?." ACM SIGCOMM Computer Communication Review 43.3 (2013): 59-67.

• Verification overhead is controlled by caching probability

2nd ACM Conference on Information-Centric Networking in San Francisco

Related W ork

• Probabilistic caching
• Bianchi, Giuseppe, et al. "Check before storing: What is the

performance price of content integrity verification in LRU caching?." ACM SIGCOMM Computer Communication Review 43.3 (2013): 59-67.

• Verification overhead is controlled by caching probability
• Limitation
• Recency problem under dynamic content popularity
• Limited application
• Strongly bounded with random caching policy

2nd ACM Conference on Information-Centric Networking in San Francisco

Motivations

• Why do we verify even the content that is not actually

served?

2nd ACM Conference on Information-Centric Networking in San Francisco

Motivations

• Why do we verify even the content that is not actually

served?

• ns-3 simulation for estimating the amount of serving contents

Cache hit rate Proportion of serving content in the CS

2nd ACM Conference on Information-Centric Networking in San Francisco

Objective

• Reduce verification overhead while preserving

functionality of the built-in signature verification

2nd ACM Conference on Information-Centric Networking in San Francisco

The Proposed Schem e

 Verify serving contents only

2nd ACM Conference on Information-Centric Networking in San Francisco

The Proposed Scheme

• Verify Serving Contents Only

2nd ACM Conference on Information-Centric Networking in San Francisco

The Proposed Scheme

• Verify Serving Contents Only

2nd ACM Conference on Information-Centric Networking in San Francisco

Signature verification

The Proposed Scheme

• Verify Serving Contents Only

2nd ACM Conference on Information-Centric Networking in San Francisco

The Proposed Scheme

• Verify Serving Contents Only
• In the proposed scheme, poisoned content is either
• Evicted from the content store without any damages to the

network

• Discarded by the verification mechanism before being brought
• ut to the network

2nd ACM Conference on Information-Centric Networking in San Francisco

The Proposed Schem e

 Flag for the already verified content

C1Name

name data

C2Name … …

Content store structure

verify

True False

2nd ACM Conference on Information-Centric Networking in San Francisco

The Proposed Schem e

 Favor the already-verified content in the content store

• Segmented LRU prevents serving content from being evicted by

by-passing content in the content store

2nd ACM Conference on Information-Centric Networking in San Francisco

Efficiency Analysis

• Efficiency metric
• : the number of examined poisoned contents
• : the number of verifications

2nd ACM Conference on Information-Centric Networking in San Francisco

Efficiency Analysis

• In the basic scheme, corresponds to the proportion
• f the requests for the poisoned contents,

2nd ACM Conference on Information-Centric Networking in San Francisco

Efficiency Analysis

• In the basic scheme, corresponds to the proportion
• f the requests for the poisoned contents,
• In the proposed scheme,
• is the request arriving rate
• is the hit ratio for the unverified contents in the CS

2nd ACM Conference on Information-Centric Networking in San Francisco

Efficiency Analysis

• Hit ratio for the unverified contents

Proportion of requests for content i

2nd ACM Conference on Information-Centric Networking in San Francisco

Efficiency Analysis

• Hit ratio for the unverified contents

Cache-miss probability for the content i

2nd ACM Conference on Information-Centric Networking in San Francisco

Efficiency Analysis

• Hit ratio for the unverified contents

Cache-hit probability for the content i

2nd ACM Conference on Information-Centric Networking in San Francisco

Efficiency Analysis

• Hit ratio for the unverified contents
• According to Che approximation
• is the size of CS, and t is the residing time in the CS

Cache-hit probability for the content i

2nd ACM Conference on Information-Centric Networking in San Francisco

Analytic Results

• without SLRU

2nd ACM Conference on Information-Centric Networking in San Francisco

Analytic Results

• with SLRU

2nd ACM Conference on Information-Centric Networking in San Francisco

Analytic Results

• In the proposed scheme without SLRU
• In the proposed scheme with SLRU
• When is close to 0, the proposed scheme achieve a 10
• r 20 time larger value of
• The value of is changed according to the amount of

poisoned content,

2nd ACM Conference on Information-Centric Networking in San Francisco

Evaluation

• Ns-3 simulation with
• 106 Contents whose popularity follows Zipf-Mandelbrot

distribution function

• youTube trace from UMASS Campus during Mar. 11-17 in 2008

2nd ACM Conference on Information-Centric Networking in San Francisco

Results

• Poisoned contents

2nd ACM Conference on Information-Centric Networking in San Francisco

Results

• Effect of Segm ented LRU

2nd ACM Conference on Information-Centric Networking in San Francisco

Results

• youTube Trace

2nd ACM Conference on Information-Centric Networking in San Francisco

Results

• youTube Trace

2nd ACM Conference on Information-Centric Networking in San Francisco

Discussion

• The access delay is increased due to the verification process
• Limited to the first access to the content

2nd ACM Conference on Information-Centric Networking in San Francisco

Discussion

• The access delay is increased due to the verification process
• Limited to the first access to the content
• Multiple pending interests may distribute the poisoned content
• Verification for the content that is matched with multiple

pending interests

2nd ACM Conference on Information-Centric Networking in San Francisco

Discussion

• The access delay is increased due to the verification process
• Limited to the first access to the content
• Multiple pending interests may distribute the poisoned content
• Verification for the content that is matched with multiple

pending interests

• Cache is attacked by using unverified data
• Might show abnormal cache-hit pattern, that is, different value
• f (Hit rate / amount of hit data)

2nd ACM Conference on Information-Centric Networking in San Francisco

Conclusion

 We look at the content poisoning attack in NDN

• Implementation and its effects

2nd ACM Conference on Information-Centric Networking in San Francisco

Conclusion

 We look at the content poisoning attack in NDN

• Implementation and its effects

 We present an efficient way to reduce overhead of content

verification at routers

• Verification of the “serving content” only

2nd ACM Conference on Information-Centric Networking in San Francisco

Conclusion

 We look at the content poisoning attack in NDN

• Implementation and its effects

 We present an efficient way to reduce overhead of content

verification at routers

• Verification of the “serving content” only

 We minimize verification overhead by favoring the serving contents

in the CS

• Flag and Segmented LRU

2nd ACM Conference on Information-Centric Networking in San Francisco

Q and A

Thank you