High throughput long integer multiplication using Fast Fourier Transform on parallel workstation
Jitendra V. Tembhurne, S. R. Sathe
Abstract
Jitendra V. Tembhurne, S. R. Sathe
Abstract
In this paper, we have proposed high throughput parallel long integer multiplication algorithm on parallel workstation. In integer arithmetic operations, long integer multiplication is the most time consuming and key operation. In public-key cryptography such as RSA, Diffie-Hellman and so on long integer multiplication is required. Long integer multiplication operation is performed heavily for the computation of exponentiation. This computation affects and bottlenecks the overall performance of cryptography process. We used floating point Fast Fourier Transform (FFT) for parallel implementation of long integer multiplication to accelerate these cryptography processes. This paper aims on high throughput and practical parallel implementation of long integer multiplication on newly available parallel processors such as multiple cores CPUs and many cores Graphics Processing Units (GPUs). We analyze and present the results on FFT benchmarks fast libraries for long integer multiplication on parallel workstation.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In this paper, we have proposed high throughput parallel long integer multiplication algorithm on parallel workstation. In integer arithmetic operations, long integer multiplication is the most time consuming and key operation. In public-key cryptography such as RSA, Diffie-Hellman and so on long integer multiplication is required. Long integer multiplication operation is performed heavily for the computation of exponentiation. This computation affects and bottlenecks the overall performance of cryptography process. We used floating point Fast Fourier Transform (FFT) for parallel implementation of long integer multiplication to accelerate these cryptography processes. This paper aims on high throughput and practical parallel implementation of long integer multiplication on newly available parallel processors such as multiple cores CPUs and many cores Graphics Processing Units (GPUs). We analyze and present the results on FFT benchmarks fast libraries for long integer multiplication on parallel workstation.
Key concepts: Fast Fourier transform, Multiplication (music), Computer science, Parallel computing, Integer (computer science), Exponentiation, Throughput, Cryptography