A Parallel GPU Implementation of SWIFFTX

2020-01-01
Ulu, Metin Evrim
Cenk, Murat
The SWIFFTX algorithm is one of the candidates of SHA-3 Hash Competition that uses the number theoretic transform (NTT). It has 256-byte input blocks and 65-byte output blocks. In this paper, a parallel implementation of the algorithm and particular techniques to make it faster on GPU are proposed. We target version 6.1 of NVIDIA®CUDA™compute architecture that employs an ISA (Instruction Set Architecture) called Parallel Thread Execution (PTX) which possesses special instrinsics, hence we modify the reference implementation for better results. Experimental results indicate almost 10x improvement in speed and 5 W decrease in power consumption per 216 hashes.
8th International Conference on Mathematical Aspects of Computer and Information Sciences, MACIS 2019

Suggestions

A high-speed asic implementation of the RSA cryptosystem
Yeşil, Soner; Aşkar, Murat; Department of Electrical and Electronics Engineering (2003)
This thesis presents the ASIC implementation of the RSA algorithm, which is one of the most widely used Public Key Cryptosystems (PKC) in the world. In RSA Cryptosystem, modular exponentiation of large integers is used for both encryption and decryption processes. The security of the RSA increases as the number of the bits increase. However, as the numbers become larger (1024-bit or higher) the challenge is to provide architectures, which can be implemented in hardware, operate at high clock speeds, use a m...
A High throughput FPGA implementation of markov chain monte carlo method for mixture models
Bozgan, Caner; Ulusoy, İlkay; Department of Electrical and Electronics Engineering (2019)
Markov Chain Monte Carlo (MCMC) is a class of algorithms which can generate samples from high dimensional and multimodal probability distributions. In many statistical and control applications, MCMC algorithms are employed widely thanks to their ability to draw sample from arbitrary distribution regardless of dimension or complexity. However, as the complexity of the Bayesian models and the computational load of the MCMC algorithm increase, performing MCMC inference becomes impractical or too time consuming...
Speeding up Curve25519 using Toeplitz Matrix-vector Multiplication
Taskin, Halil Kemal; Cenk, Murat (2018-01-24)
This paper proposes a new multiplication algorithm over F-2(255)-19 where the de-facto standard Curve25519 [2] algorithm is based on. Our algorithm for the underlying finite field multiplication exploits the Toeplitz matrix-vector multiplication and achieves salient results. We have used a new radix representation that is infeasible when used with schoolbook multiplication techniques but has notable advantages when used with Toeplitz matrix-vector multiplication methods. We present the new algorithm and dis...
Performance analysis of elliptic curve multiplication algorithms for elliptic curve cryptography
Özcan, Ayça Bahar; Yücel, Melek D; Department of Electrical and Electronics Engineering (2006)
Elliptic curve cryptography (ECC) has been introduced as a public-key cryptosystem, which offers smaller key sizes than the other known public-key systems at equivalent security level. The key size advantage of ECC provides faster computations, less memory consumption, less processing power and efficient bandwidth usage. These properties make ECC attractive especially for the next generation public-key cryptosystems. The implementation of ECC involves so many arithmetic operations; one of them is the ellipt...
A new scheme for off-line handwritten connected digit recognition
Arica, N; Yarman Vural, Fatoş Tunay (1998-08-20)
A new scheme is proposed for off-line handwritten connected digit recognition, which uses a sequence of segmentation and recognition algorithms. First, the connected digits are segmented by employing both the gray scale and binary information. Then, a new set of features is extracted from the segments. The parameters of the feature set are adjusted during the training stage of the Hidden Markov Model (HMM) where the potential digits are recognized. Finally, in order to confirm the preliminary segmentation a...
Citation Formats
M. E. Ulu and M. Cenk, “A Parallel GPU Implementation of SWIFFTX,” presented at the 8th International Conference on Mathematical Aspects of Computer and Information Sciences, MACIS 2019, Gebze, Türkiye, 2020, Accessed: 00, 2021. [Online]. Available: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85083701099&origin=inward.