with Tunable Encrypted Deduplication Jingwei Li * , Zuoru Yang # , - - PowerPoint PPT Presentation

β–Ά
with tunable encrypted deduplication
SMART_READER_LITE
LIVE PREVIEW

with Tunable Encrypted Deduplication Jingwei Li * , Zuoru Yang # , - - PowerPoint PPT Presentation

May 09, 2023 294 likes β€’456 views

Balancing Storage Efficiency and Data Confidentiality with Tunable Encrypted Deduplication Jingwei Li * , Zuoru Yang # , Yanjing Ren * , Patrick P. C. Lee # , Xiaosong Zhang * * University of Electronic Science and Technology of China (UESTC) # The

slide-1
SLIDE 1

Balancing Storage Efficiency and Data Confidentiality with Tunable Encrypted Deduplication

Jingwei Li*, Zuoru Yang#, Yanjing Ren*, Patrick P. C. Lee#, Xiaosong Zhang*

*University of Electronic Science and Technology of China (UESTC) #The Chinese University of Hong Kong (CUHK)

EuroSys 2020

1

slide-2
SLIDE 2

Deduplication

  • Deduplication οƒ  coarse-grained compression
  • Units: chunks (fixed- or variable-size)
  • Stores only one copy of duplicate chunks

2

Storage space saved by 5/12 = 42%!

slide-3
SLIDE 3

Encrypted Deduplication

  • Augments deduplication with encryption for data confidentiality
  • Application: outsourced storage

3

Encryption Encryption

πŸ”“

Deduplication

πŸ”“ Storage Provider

Which crypto primitive should be used?

slide-4
SLIDE 4

Encryption Primitives

  • Symmetric-key encryption (SKE)
  • Derives a random key for chunk encryption/decryption
  • Ensures confidentiality, but prohibits deduplication of duplicate chunks
  • Message-locked encryption (MLE) [Bellare et al., Eurocrypt’13]
  • Derives a deterministic key from chunk content
  • Supports deduplication, but leaks frequency distribution of plaintext

chunks [Li et al., DSN’17]

4

Pose a dilemma of choosing the right cryptographic primitive

slide-5
SLIDE 5

Our Contributions

  • TED: a tunable encrypted deduplication primitive for balancing

trade-off between storage efficiency and data confidentiality

  • Includes three new designs
  • Minimizes frequency leakage via a configurable storage blowup factor
  • TEDStore: encrypted deduplication prototype based on TED
  • TED incurs only limited performance overhead
  • Extensive trace-driven analysis and prototype experiments

5

slide-6
SLIDE 6
  • Key derivation with three inputs: chunk M, current frequency f, and

balance parameter t

  • f: cumulative and increases with number of duplicates of M
  • t: controls maximum allowed number of duplicate copies for a ciphertext chunk
  • Special cases:
  • t = 1 οƒ  SKE
  • t β†’ ∞ οƒ  MLE

Main Idea

6

K = H(M || ⌊f/tβŒ‹)

Hash Function

M f t

πŸ”’ K

slide-7
SLIDE 7

Design Overview

  • TED builds on server-aided MLE architecture in DupLESS

[Bellare et al., Security’13]

  • Key generation by key manager to prevent offline brute-force attacks

7

Deduplication Provider Key Manager Clients Chunk … Chunk

slide-8
SLIDE 8

Questions

  • Q1: How does the key manager learn chunk frequencies?
  • Low overhead required even for many chunks
  • Q2: How does the key manager generate keys for chunks?
  • Distinct sequences of ciphertext chunks required for identical files
  • Q3: How should the balance parameter t be configured in practice?
  • Adaptive for different workloads

8

slide-9
SLIDE 9

Sketch-based Frequency Counting

  • Key manager estimates f via Count-Min Sketch [Cormode 2005]
  • Fixed memory usage with provable error bounds
  • Client sends short hashes {Hi(M)} to key manager
  • Key manager cannot readily infer M from short hashes

9

w counters per row M

Count-Min Sketch

r rows

+1 +1 +1 +1 H1(M) Hr(M)

f = minimum counter indexed by (i, Hi(M))

slide-10
SLIDE 10

Probabilistic Key Generation

  • Selects K uniformly from candidate keys derived from 0, 1,…, ⌊f/tβŒ‹
  • Enables probabilistic encryption on identical files
  • Maintains deduplication effectiveness
  • Reason: f is cumulative; keys derived from 0, 1,…, ⌊f/tβŒ‹-1 have been used to encrypt

some old copies of M

10

M M M M … … … … M M M M … … … ………

Processing sequence

πŸ”“ πŸ”’ K0 πŸ”“ πŸ”’ K0 πŸ”“ πŸ”’ K1 πŸ”“ πŸ”’ K1 πŸ”“ πŸ”’ K2 πŸ”“ πŸ”’ K2 πŸ”“ πŸ”’ K3 πŸ”’ K ← {K0, K1, K2, K3} Already encrypted chunks

slide-11
SLIDE 11

Automated Parameter Configuration

  • Configure t by solving optimization problem, given:
  • Frequency distribution for a batch of plaintext chunks
  • Affordable storage blowup b over exact deduplication
  • Goal: minimize frequency leakage
  • Quantify frequency leakage by Kullback-Leibler distance (KLD)
  • KLD: relative entropy to uniform distribution
  • A lower KLD implies higher robustness against frequency analysis
  • Configure t from the returned optimal frequency distribution of ciphertext

chunks

11

slide-12
SLIDE 12

Evaluation

  • TEDStore realizes TED in encrypted deduplication storage
  • ~4.5K line of C++ code in Linux
  • Trace analysis
  • FSL: file system snapshots (42 backups; 3.08TB raw data)
  • MS: windows file system snapshots (30 backups; 3.91TB raw data)
  • Prototype experiments
  • Local 10 GbE cluster

12

slide-13
SLIDE 13

Trade-off Analysis (FSL Dataset)

  • Basic TED and Full TED effectively balance trade-off
  • Full TED readily configures actual storage blowup

13

  • Schemes
  • MLE
  • SKE
  • MinHash [Li et al., DSN’17]
  • Basic TED (varying t)
  • Full TED (varying b)
slide-14
SLIDE 14

Prototype Experiments

14

Computational time per 1MB of uploads

TED operations

Steps Fast (MD5, AES-128) Secure (SHA-256, AES-256)

Chunking 0.8ms Fingerprinting 1.7ms 2.6ms Hashing 0.4ms Key Seeding 0.01ms 0.04ms Key Derivation 0.07ms 0.1ms Encryption 3.7ms 4.9ms

  • TED incurs limited overhead (7.2% for Fast; 6.1% for Secure)
  • More results in paper:
  • TED achieves ~30X key generation speedup over existing approaches
  • Multi-client upload/download performance
slide-15
SLIDE 15

Conclusion

  • TED: encrypted deduplication primitive that enables controllable

trade-off between storage efficiency and data confidentiality

  • Sketch-based frequency counting
  • Probabilistic key generation
  • Automated parameter configuration
  • Source code: http://adslab.cse.cuhk.edu.hk/software/ted

15