FPGA design and performance analysis of SHA-512, whirlpool and PHASH hashing functions

Show full item record

Redirect: RIT Scholars content from RIT Digital Media Library has moved from http://ritdml.rit.edu/handle/1850/7740 to RIT Scholar Works http://scholarworks.rit.edu/theses/3213, please update your feeds & links!
Title: FPGA design and performance analysis of SHA-512, whirlpool and PHASH hashing functions
Author: Zalewski, Przemyslaw
Abstract: Hashing functions play a fundamental role in modern cryptography. Such functions process data of finite length and produce a small fixed size output referred to as a digest or hash. Typical applications of these functions include data integrity verification and message authentication schemes. With the ever increasing amounts of data that needs to be hashed, the throughput of hashing functions becomes an important factor. This work presents and compares high performance FPGA implementations of SHA- 512, Whirlpool and a recently proposed parallelizable hash function, PHASH. The novelty of PHASH is that it is able to process multiple data blocks at once, making it suitable for achieving ultra high-performance. It utilizes the W cipher, as described in Whirlpool, at its core. The SHA (SHA-0, SHA-1, SHA-2) family of functions is one of the first widely accepted standards for hashing. According to currently published literature, the fastest SHA-512 (a variant of SHA-2) implementation achieves a throughput of 1550 Mbps. A recently introduced hashing function, Whirlpool, provides comparable security to SHA- 512 and is able to achieve much better performance. According to currently published literature, the fastest Whirlpool implementation achieves a throughput of 4896 Mbps. The proposed PHASH hash function greatly outperforms both SHA-512 and Whirlpool. All implementations are targeted for the state-of-the-art Xilinx Virtex-5 LX330 FPGA. The SHA-512 implementation attains a throughput of 1828 Mbps, and Whirlpool attains 7687 Mbps. PHASH achieves a throughput over 15 Gbps using a singleWcipher instance. Using 8 W cipher instances a throughput over 100 Gbps is achieved and 16 instances provide a throughput over 182 Gbps.
Record URI: http://hdl.handle.net/1850/7740
Date: 2008-05

Files in this item

Files Size Format View
PZalewskiThesis05-2008.pdf 1.602Mb PDF View/Open

The following license files are associated with this item:

This item appears in the following Collection(s)

Show full item record

Search RIT DML

Advanced Search