Fast Integrity Verification for High-Speed File Transfers
share
The amount of data generated by scientific and commercial applications is growing at an ever-increasing pace. This data is often moved between geographically distributed sites for various purposes such as collaboration and backup which has led to significant increase in data transfer rates. Surge in data transfer rates when combined with proliferation of scientific applications that cannot tolerate data corruption triggered enhanced integrity verification techniques to be developed. End-to-end integrity verification minimizes the likelihood of silent data corruption by comparing checksum of files at source and destination servers using secure hash algorithms such as MD5 and SHA1. However, it imposes significant performance penalty due to overhead of checksum computation. In this paper, we propose Fast Integrity VERification (FIVER) algorithm which overlaps checksum computation and data transfer operations of files to minimize the cost of integrity verification. Extensive experiments show that FIVER is able to bring down the cost from 60 solutions to below 10 operations and enabling file I/O share between them. We also implemented FIVER-Hybrid to mimic disk access patterns of sequential integrity verification approach to capture possible data corruption that may occur during file write operations which FIVER may miss. Results show that FIVER-Hybrid is able to reduce execution time by 20 compromising the reliability of integrity verification.
READ FULL TEXT