Scale and performance in a distributed file system
ACM Transactions on Computer Systems (TOCS)
Fast and secure distributed read-only file system
ACM Transactions on Computer Systems (TOCS)
Unifying File System Protection
Proceedings of the General Track: 2002 USENIX Annual Technical Conference
Caches and Hash Trees for Efficient Memory Integrity Verification
HPCA '03 Proceedings of the 9th International Symposium on High-Performance Computer Architecture
Protecting the Integrity of an Entire File System
IEEE-IWIA '03 Proceedings of the First IEEE International Workshop on Information Assurance (IWIA'03)
HOTOS '01 Proceedings of the Eighth Workshop on Hot Topics in Operating Systems
Towards Constant Bandwidth Overhead Integrity Checking of Untrusted Data
SP '05 Proceedings of the 2005 IEEE Symposium on Security and Privacy
Plutus: Scalable Secure File Sharing on Untrusted Storage
FAST '03 Proceedings of the 2nd USENIX Conference on File and Storage Technologies
Efficient Authentication of Large, Dynamic Data Sets Using Galois/Counter Mode (GCM)
SISW '05 Proceedings of the Third IEEE International Security in Storage Workshop
How to build a trusted database system on untrusted storage
OSDI'00 Proceedings of the 4th conference on Symposium on Operating System Design & Implementation - Volume 4
Cryptographic Security for a High-Performance Distributed File System
MSST '07 Proceedings of the 24th IEEE Conference on Mass Storage Systems and Technologies
Integrity checking in cryptographic file systems with constant trusted storage
SS'07 Proceedings of 16th USENIX Security Symposium on USENIX Security Symposium
An efficient way to build secure disk
ISPEC'06 Proceedings of the Second international conference on Information Security Practice and Experience
Hi-index | 0.00 |
Hash tree based verification can give solid data integrity protection to disk data; however, it suffers from performance penalty and consistency difficulty. AFI-HTree is proposed to solve such problems. To optimize performance, it utilizes hot-access-windows to buffer the frequently used hash tree nodes to quicken the checking speed. To maintain consistency without compromising security and performance, it fixes the structure of hash tree to make it very regular; then, it applies incremental hash to reduce the update cost of synchronization between the tree and the data; at the end, it records any possible inconsistent states to make fast recovery. In such way, AFI-HTree realizes both high performance and fine consistency, while preserving the required security at the same time. Related approach is elaborated, as well as experiment result. Theoretical analysis and experimental simulation show that it is an optimized way to protect disk data integrity.