{"_id":"67494","user_id":"81636","author":[{"full_name":"Januszewski, Fabian","last_name":"Januszewski","first_name":"Fabian"},{"full_name":"Heinrichs, Christoph","last_name":"Heinrichs","first_name":"Christoph"}],"year":"2026","title":"CUDA-MPQS: A GPU-Resident Self-Initializing Quadratic Sieve, and the Factorization of RSA-155","status":"public","date_updated":"2026-10-11T10:09:12Z","date_created":"2026-10-11T10:08:48Z","external_id":{"arxiv":["2610.07126"]},"type":"preprint","citation":{"mla":"Januszewski, Fabian, and Christoph Heinrichs. “CUDA-MPQS: A GPU-Resident Self-Initializing Quadratic Sieve, and the Factorization of RSA-155.” <i>ArXiv:2610.07126</i>, 2026.","bibtex":"@article{Januszewski_Heinrichs_2026, title={CUDA-MPQS: A GPU-Resident Self-Initializing Quadratic Sieve, and the Factorization of RSA-155}, journal={arXiv:2610.07126}, author={Januszewski, Fabian and Heinrichs, Christoph}, year={2026} }","ama":"Januszewski F, Heinrichs C. CUDA-MPQS: A GPU-Resident Self-Initializing Quadratic Sieve, and the Factorization of RSA-155. <i>arXiv:261007126</i>. Published online 2026.","ieee":"F. Januszewski and C. Heinrichs, “CUDA-MPQS: A GPU-Resident Self-Initializing Quadratic Sieve, and the Factorization of RSA-155,” <i>arXiv:2610.07126</i>. 2026.","apa":"Januszewski, F., &#38; Heinrichs, C. (2026). CUDA-MPQS: A GPU-Resident Self-Initializing Quadratic Sieve, and the Factorization of RSA-155. In <i>arXiv:2610.07126</i>.","chicago":"Januszewski, Fabian, and Christoph Heinrichs. “CUDA-MPQS: A GPU-Resident Self-Initializing Quadratic Sieve, and the Factorization of RSA-155.” <i>ArXiv:2610.07126</i>, 2026.","short":"F. Januszewski, C. Heinrichs, ArXiv:2610.07126 (2026)."},"publication":"arXiv:2610.07126","abstract":[{"lang":"eng","text":"The quadratic sieve (QS) is an irregular, cache-hostile, branch-heavy integer factorization algorithm; prior GPU work accelerates individual stages of it. We present \\swname{}, a self-initializing quadratic sieve in which \\emph{every} stage --- polynomial initialization, sieving, relation post-processing, large-prime matching, \\GFtwo{} matrix construction, Block Wiedemann linear algebra and the square root --- executes on the GPU, the CPU restricted to orchestration, setup and I/O, the sieve window 99.9\\% GPU-busy with no device-wide or event synchronization and no synchronous copy. With it we factored the 512-bit RSA-155 challenge modulus end to end on GPUs: a 64$\\times$H100 sieve feeding a single-H100 \\GFtwo{} solve, 700.6~GPU-h and 242~kWh in total, 98.4\\% of it sieving. To our knowledge this is the largest integer factored by the quadratic sieve, exceeding the prior QS record RSA-150 (Buhrow, YAFU, 2025); all comparable general factorizations since 1996, RSA-155's own in 1999 included, used the number field sieve. On a single device we factor RSA-100 in 29.2~s on an H100 and 51.0~s on a consumer RTX~5070~Ti; in a controlled same-modulus measurement, with energy measured on both sides, the H100 is faster than the fastest CPU quadratic sieve, YAFU, on 96 cores of an EPYC~9655 (106--122~s), and than CADO-NFS's SIQS (292~s) and GNFS (263~s). Across four discrete GPUs (a 3.7$\\times$ range in CUDA cores) RSA-100 throughput is close to proportional to core count, memory bandwidth being a weaker predictor, and measured cost over seven benchmark factorizations from RSA-100 to RSA-155 grows as $\\LN{1/2}{1}^{1.04}$ ($R^2=0.993$). On RSA-150, the identical modulus on which the prior QS record was set, we measure 302.9~GPU-h against that run's 11{,}664 core-hours."}]}