LiTIV: A Lightweight Traceable Data Integrity Verification Scheme for Version Control Systems


연구 분야: Software Development



학회: 2022 International Conference on Computer Communications and Networks (ICCCN)


초록

Conventional data integrity solutions for Version Control Systems (VCSs) include hash verification, commit signature and log-based schemes. However, existing schemes either bring heavy burden to the VCS client, or place high requirements on the VCS server. These schemes are incapable of offering a lightweight service. Moreover, existing schemes cannot offer corruption provenance that facilitates corruption location and version recovery. To address these challenges, we present LiTIV, a lightweight traceable data integrity verification scheme for VCS. LiTIV introduces a third-party verification server, namely Data Verifier (DV), which effectively reduces impacts on the VCS client and server. LiTIV exploits index history table (IHT) along with signed commit information to achieve both integrity and traceability. Besides, a balanced Merkle hash tree (BMT) is designed and adopted to achieve fast data updating and efficient integrity verification. In case of malicious tampering of V CS file system disks, LiTIV can rely on the BMT structure to effectively detect implicit inconsistency, and achieve backtracking for tam-pering provenance with IHT query. We also design a probabilistic sampling verification mechanism on historical commits to achieve high performance under security premise. Extensive evaluations show that LiTIV achieves traceable integrity security against various attacks and is highly efficient.


Author Profile
Jingqiang Lin

School of Cyber Science and Technology University of Science and Technology of China Hefei China

Andorra
Author Profile
Xin Xu

State Key Laboratory of Information Security CAS Institute of Information Engineering Beijing China

China
Author Profile
Wei Wang

State Key Laboratory of Information Security CAS Institute of Information Engineering Beijing China

China

📄 논문 정보

발행 연도 2022년
인용수 213
출판 국가 Andorra, China
사이트 IEEE
좋아요 수 0

연관 논문 목록 (4건)