원문정보
초록
영어
Taking cognizance of various attacks on MDx family of Hash function including SHA-0 and SHA-1, National Institute of Standards and Technology came up with a public competition for new Hash standard SHA-3 to augment or replace the current Hash standard SHA-2. NIST provided a reference platform (Wintel machines) and compiler (Microsoft Visual Studio) on which the submitted algorithms were to be evaluated. The five algorithms that reached the final round of SHA-3 public competition were evaluated by us on ARM Cortex A8 architecture (architecture other than the reference platform). Performance evaluation of these final round candidate algorithms were done on Cortex-A8 based OpenBoard AM335x from PHYTEC running Linux 3.2.0. The results in Cycles consumed per byte are shared in this paper.
목차
1. Introduction
2. Selection of Target Platform and the Rationale behind It
2.1. Decision – 1: Going for Embedded and Mobile Platform
2.2. Decision – 2: Zeroing Down to ARM Cortex Application Series Architecture
3. Algorithms
3.1. Keccak
3.2. Skein
3.3. Grøstl
3.4. Blake
3.5. JH
4. Methodology and Tools Used
4.1. Hardware and Software Used
4.2. CPB as Performance Parameter
4.3. Hash Function as a Whole rather than Compression Phase alone
4.4. Computation of Cycles using System Control Coprocessor
4.5. Averaging the Cycle Count and Subtracting the Overhead
4.6. Optimized 32 bit Implementation of Hash Algorithms Used
5. Results and Discussion
5.1. Results for Short Messages
5.2. Results for Long Messages
5.3. Few Common Observations for Short as well as Long Messages
6. Conclusion and Future Work
References