Cryptology ePrint Archive: Report 2014/397

Cofactorization on Graphics Processing Units

Andrea Miele and Joppe W. Bos and Thorsten Kleinjung and Arjen K. Lenstra

Abstract: We show how the cofactorization step, a compute-intensive part of the relation collection phase of the number field sieve (NFS), can be farmed out to a graphics processing unit. Our implementation on a GTX 580 GPU, which is integrated with a state-of-the-art NFS implementation, can serve as a cryptanalytic co-processor for several Intel i7-3770K quad-core CPUs simultaneously. This allows those processors to focus on the memory-intensive sieving and results in more useful NFS-relations found in less time.

Category / Keywords: Cofactorization, GPUs, Number Field Sieve

Original Publication (with minor differences): IACR-CHES-2014

Date: received 30 May 2014, last revised 28 Jan 2015

Contact author: andrea miele at epfl ch

Available format(s): PDF | BibTeX Citation

Note: We fixed a typo

Version: 20150128:161853 (All versions of this report)

Short URL: ia.cr/2014/397

Discussion forum: Show discussion | Start new discussion