{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T14:16:51Z","timestamp":1778595411096,"version":"3.51.4"},"reference-count":18,"publisher":"MDPI AG","issue":"5","license":[{"start":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T00:00:00Z","timestamp":1778544000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Computation"],"abstract":"<jats:p>The Biconjugate Gradient Stabilized (BiCGStab) algorithm is a widely used iterative method for solving large, sparse, and non-symmetric linear systems in scientific and engineering applications. While efficient, its performance is constrained by high iteration costs, memory bandwidth limitations, and synchronization overheads in CPU implementations. This paper investigates GPU-based acceleration of BiCGStab, with particular emphasis on the use of CUDA streams to optimize kernel concurrency and improve resource utilization. A structured hepta-diagonal matrix format is adopted to ensure efficient memory access across both CPU and GPU executions. Performance evaluations are conducted across problem sizes ranging from 1 to 64 million unknowns, comparing single-threaded and multi-threaded CPU baselines against GPU implementations with and without CUDA streams. The results demonstrate that GPU acceleration achieves up to 30\u00d7 speedup relative to single-threaded CPU execution and up to 5\u00d7 compared to the best OpenMP configuration (16 threads), with CUDA streams providing an additional 10\u201320% performance improvement through intra-iteration kernel overlap. Scalability analysis reveals that GPU performance advantages increase with problem size, underscoring the effectiveness of CUDA streams in minimizing idle GPU time and enhancing throughput. These findings highlight the potential of stream-optimized GPU solvers for large-scale scientific simulations and provide a foundation for future extensions incorporating CUDA graphs and multi-GPU environments.<\/jats:p>","DOI":"10.3390\/computation14050110","type":"journal-article","created":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T12:33:37Z","timestamp":1778589217000},"page":"110","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Intra-GPU Concurrency in BiCGStab Solvers: Leveraging CUDA Streams for Kernel-Level Parallelism"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1167-7319","authenticated-orcid":false,"given":"Ayaz H.","family":"Khan","sequence":"first","affiliation":[{"name":"Computer Engineering Department, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,5,12]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"75","DOI":"10.1007\/BF02141261","article-title":"BiCGstab(l) and other hybrid Bi-CG methods","volume":"7","author":"Sleijpen","year":"1994","journal-title":"Numer. Algorithms"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Aliaga, J., Quintana-Ort\u00ed, E.S., Dufrechou, E., and Ezzatti, P. (2018, January 1\u20135). Extending ILUPACK with a GPU Version of the BiCGStab Method. Proceedings of the 2018 XLIV Latin American Computer Conference (CLEI), Sao Paulo, Brazil.","DOI":"10.1109\/CLEI.2018.00092"},{"key":"ref_3","unstructured":"Yamazaki, I., Abdelfattah, A., Ida, A., Ohshima, S., Tomov, S., Yokota, R., and Dongarra, J. (2026, February 15). Analyzing Performance of BiCGStab with Hierarchical Matrix on GPU Clusters. IEEE International Parallel and Distributed Processing Symposium (IPDPS), Available online: https:\/\/par.nsf.gov\/biblio\/10065618."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Yamazaki, I., Abdelfattah, A., Ida, A., Ohshima, S., Tomov, S., Yokota, R., and Dongarra, J. (2018, January 21\u201325). Performance of Hierarchical-matrix BiCGStab Solver on GPU Clusters. Proceedings of the 2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS), Vancouver, BC, Canada.","DOI":"10.1109\/IPDPS.2018.00102"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"510","DOI":"10.31577\/cai_2020_3_510","article-title":"Investigation of Parallel Data Processing Using Hybrid High Performance CPU + GPU Systems and CUDA Streams","volume":"39","author":"Czarnul","year":"2020","journal-title":"Comput. Inform."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1477","DOI":"10.1007\/s40819-017-0403-8","article-title":"Optimized BiCGStab Based GPU Accelerated Computation of Incompressible Viscous Flows by the \u03c8\u2013v Formulation","volume":"3","author":"Kalita","year":"2017","journal-title":"Int. J. Appl. Comput. Math."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"3761","DOI":"10.1007\/s11227-017-1972-3","article-title":"SpMV and BICG-stab optimization for a class of hepta-diagonal-sparse matrices on GPU","volume":"73","author":"Khan","year":"2017","journal-title":"J. Supercomput."},{"key":"ref_8","unstructured":"Gray, A. (2025, August 19). Getting Started with CUDA Graphs | NVIDIA Technical Blog. Available online: https:\/\/developer.nvidia.com\/blog\/cuda-graphs\/."},{"key":"ref_9","unstructured":"Werneck, L.F., de Freitas, M.M., da Silva, H.G., de Souza, G., and Souto, H.P.A. (2016, January 6\u20139). An OpenMP Parallel Implementation for Numerical Simulation of Gas Reservoirs Using Intel Xeon Phi Coprocessor. Proceedings of the XXXVII Iberian Latin-American Congress on Computational Methods in Engineering (CILAMCE), Bras\u00edlia, Brazil."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Rocki, K., Van Essendelft, D., Sharapov, I., Schreiber, R., Morrison, M., Kibardin, V., Portnoy, A., Dietiker, J.F., Syamlal, M., and James, M. (2020, January 9\u201319). Fast stencil-code computation on a wafer-scale processor. Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, Atlanta, GA, USA.","DOI":"10.1109\/SC41405.2020.00062"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"2574","DOI":"10.1016\/j.camwa.2019.11.025","article-title":"Revisiting performance of BiCGStab methods for solving systems with multiple right-hand sides","volume":"79","author":"Krasnopolsky","year":"2020","journal-title":"Comput. Math. Appl."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"127868","DOI":"10.1016\/j.amc.2023.127868","article-title":"Communication-hiding pipelined BiCGSafe methods for solving large linear systems","volume":"449","author":"Huynh","year":"2023","journal-title":"Appl. Math. Comput."},{"key":"ref_13","first-page":"4","article-title":"Evaluating Performance of Mixed Precision Linear Solvers with Iterative Refinement","volume":"8","author":"Krasnopolsky","year":"2021","journal-title":"Supercomput. Front. Innov."},{"key":"ref_14","first-page":"129341E","article-title":"Fast multilevel B-spline approximation for scattered data on GPU architecture using CUDA","volume":"Volume 12934","author":"Wang","year":"2023","journal-title":"Proceedings of the Third International Conference on Computer Graphics, Image, and Virtualization (ICCGIV 2023), Nanjing, China, 16\u201318 June 2023"},{"key":"ref_15","unstructured":"Oyarzun, G., Mira, D., and Houzeaux, G. (2021). Performance assessment of CUDA and OpenACC in large scale combustion simulations. arXiv."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"106957","DOI":"10.1016\/j.anucene.2019.106957","article-title":"Implementation of the CPU\/GPU hybrid parallel method of characteristics neutron transport calculation using the heterogeneous cluster with dynamic workload assignment","volume":"135","author":"Song","year":"2020","journal-title":"Ann. Nucl. Energy"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"123","DOI":"10.1007\/s42514-020-00034-9","article-title":"Towards a heterogeneous architecture solver for the incompressible Navier\u2013Stokes equations","volume":"2","author":"Wang","year":"2020","journal-title":"CCF Trans. High Perform. Comput."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"4119","DOI":"10.1109\/TKDE.2020.3038109","article-title":"CuWide: Towards Efficient Flow-Based Training for Sparse Wide Models on GPUs","volume":"34","author":"Miao","year":"2022","journal-title":"IEEE Trans. Knowl. Data Eng."}],"container-title":["Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2079-3197\/14\/5\/110\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T13:28:48Z","timestamp":1778592528000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2079-3197\/14\/5\/110"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,12]]},"references-count":18,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,5]]}},"alternative-id":["computation14050110"],"URL":"https:\/\/doi.org\/10.3390\/computation14050110","relation":{},"ISSN":["2079-3197"],"issn-type":[{"value":"2079-3197","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,12]]}}}