High-performance CPU-GPU heterogeneous computing method for 9-component ambient noise cross-correlation

Abstract

Ambient noise tomography is an established technique in seismology, where calculating single- or nine-component noise cross-correlation functions (NCFs) is a fundamental first step. In this study, we introduced a novel CPU-GPU heterogeneous computing framework designed to significantly enhance the efficiency of computing 9-component NCFs from seismic ambient noise data. This framework not only accelerated the computational process by leveraging the Compute Unified Device Architecture (CUDA) but also improved the signal-to-noise ratio (SNR) through innovative stacking techniques, such as time-frequency domain phase-weighted stacking (tf-PWS). We validated the program using multiple datasets, confirming its superior computation speed, improved reliability, and higher signal-to-noise ratios for NCFs. Our comprehensive study provides detailed insights into optimizing the computational processes for noise cross-correlation functions, thereby enhancing the precision and efficiency of ambient noise imaging.

Keywords

Nine-component NCFs / Heterogeneous computing / Ambient noise tomography / CUDA / tf-PWS

CRediT authorship contribution statement

Jingxi Wang: Writing - review & editing, Writing - original draft, Visualization, Validation, Software, Methodology, Investigation, Formal analysis, Data curation. Weitao Wang: Supervision, Software, Methodology, Funding acquisition, Conceptualization. Chao Wu: Supervision, Software, Methodology, Conceptualization. Lei Jiang: Software, Methodology. Hanwen Zou: Validation. Huajian Yao: Writing - review & editing, Supervision, Resources, Funding acquisition. Ling Chen: Supervision.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Author agreement and acknowledgment

All author agree for this publication. We would like to thank the re-viewers for their valuable comments, which have greatly improved the quality of the manuscript. We also appreciate the editor's patient assis-tance and professional guidance. This study was supported by the Key Research and Development Program of China (2021YFC3000704), Institute of Geophysics, China Earthquake Administration Grant DQJB23R18, and the USTC Research Funds of the Double First-Class Initiative (YD2080002012), NSFC Grant(U2239206). All topographic maps presented in this paper were created using GMT (Generic Mapping Tools).

References

[1] Bensen G.D., Ritzwoller M.H., Barmin M.P., Levshin A.L., Lin F., Moschetti M.P., Shapiro N.M., Yang Y., 2007. Processing seismic ambient noise data to obtain reliable broad-band surface wave dispersion measurements. Geophys. J. Int. 169 (3), 1239-1260. https://doi.org/10.1111/j.1365-246X.2007.03374.x[Article].

[2] Clements T., Denolle M.A., 2020. SeisNoise.jl: ambient seismic noise cross correlation on the CPU and GPU in julia. Seismol Res. Lett. 92 (1), 517-527. https://doi.org/10.1785/0220200192.

[3] Cook S., 2012. CUDA Programming: a Developer's Guide to Parallel Computing with GPUs. Newnes.

[4] Feng J., Yao H., Wang Y., Poli P., Mao Z., 2021. Segregated oceanic crust trapped at the bottom mantle transition zone revealed from ambient noise interferometry. Nat. Commun. 12 (1), 2531. https://doi.org/10.1038/s41467-021-22853-2.

[5] Fichtner A., Ermert L., Gokhberg A., 2017. Seismic noise correlation on heterogeneous supercomputers. Seismol Res. Lett. 88 (4), 1141-1145. https://doi.org/10.1785/0220170043.

[6] Fujita K., Yamaguchi T., Kikuchi Y., Ichimura T., Hori M., Maddegedara L., 2023. Calculation of cross-correlation function accelerated by TensorFloat-32 tensor core operations on NVIDIA's ampere and hopper GPUs. Journal of Computational Science 68, 101986. https://doi.org/10.1016/j.jocs.2023.101986.

[7] Google Cloud, 2024a. GPUs pricing. Retrieved 0629 from. https://cloud.google.com/compute/gpus-pricing.

[8] Google Cloud, 2024b. VM instance pricing. Retrieved 0629 from. https://cloud.google.com/compute/vm-instance-pricing.

[9] Huang H., Yao H., van der Hilst R.D., 2010. Radial anisotropy in the crust of SE Tibet and SW China from ambient noise interferometry. Geophys. Res. Lett. 37 (21). https://doi.org/10.1029/2010GL044981.

[10] Jin J., Luo S., Yao H., Tian X., 2023. Dense array ambient noise tomography reveals the shallow crustal velocity structure and deformation features in the Weifang segment of the Tanlu fault zone. Chin. J. Geophys. 66 (2), 558-575.

[11] Li G., Niu F., Yang Y., Xie J., 2017. An investigation of time-frequency domain phaseweighted stacking and its application to phase-velocity extraction from ambient noise's empirical Green's functions. Geophys. J. Int. 212 (2), 1143-1156. https://doi.org/10.1093/gji/ggx448.

[12] Li N., W Weitao, Wang Baoshan, 2018. Speeding the nine-component cross correlation function calculation using cloud-computing and its application on the dataset of China array-NE tibet. Earthq. Res. China 34, 244-257.

[13] Lin F.C., Moschetti M.P., Ritzwoller M.H., 2008. Surface wave tomography of the western United States from ambient seismic noise: Rayleigh and Love wave phase velocity maps. Geophys. J. Int. 173 (1), 281-298. https://doi.org/10.1111/j.1365246x.2008.03720.x.

[14] Obara K., K, K., Hori S., Okada Y., 2005. A densely distributed high-sensitivity seismograph network in Japan: hi-net by national research Institute for Earth science and disaster prevention. Rev. Sci. Instrum. 76 (21301). https://doi.org/10.1063/1.1854197.

[15] Okada Y., K, K., Hori S., Obara K., Sekiguchi S., Fujiwara H., Yamamoto A., 2004. Recent progress of seismic observation networks in Japan -Hi-net, F-net, K-NET and KiK-net. Earth Planets Space 56. https://doi.org/10.1186/BF03353076.

[16] Schimmel M., Gallart J., 2005. The inverse S-transform in filters with time-frequency localization. IEEE Trans. Signal Process. 53 (11), 4417-4422.

[17] Schimmel M., Paulssen H., 1997. Noise reduction and detection of weak, coherent signals through phase-weighted stacks. Geophys. J. Int. 130 (2), 497-505. https://doi.org/10.1111/j.1365-246X.1997.tb05664.x.

[18] Schimmel M., Stutzmann E., Gallart J., 2011. Using instantaneous phase coherence for signal extraction from ambient noise data at a local to a global scale. Geophys. J. Int. 184 (1), 494-506. https://doi.org/10.1111/j.1365-246X.2010.04861.x.

[19] Shapiro N.M., Campillo M., Stehly L., Ritzwoller M.H., 2005. High-resolution surfacewave tomography from ambient seismic noise. Science 307 (5715), 1615-1618. https://doi.org/10.1126/science.1108339.

[20] Wu C., Tan X., Li H., Sun G., 2022. An efficient ambient noise cross-correlation algorithm on heterogeneous CPU-GPU cluster conference paper.2022. IEEE 13th International Symposium on Parallel Architectures. Algorithms and Programming (PAAP), pp. 1-5. https://doi.org/10.1109/paap56126.2022.10010612.]

[21] Yang X., Bryan J., Okubo K., Jiang C., Clements T., Denolle M.A., 2022. Optimal stacking of noise cross-correlation functions. Geophys. J. Int. 232 (3), 1600-1618. https://doi.org/10.1093/gji/ggac410.

[22] Yao H., X, G., Zhu L., Xiao X., 2005. Mantle structure from inter-station Rayleigh wave dispersion and its tectonic implication in Western China and neighboring regions. Phys. Earth Planet. In. 148 (1), 39-54.

[23] Yao H., van Der Hilst R.D., de Hoop M.V., 2006. Surface-wave array tomography in SE Tibet from ambient seismic noise and two-station analysis - I. Phase velocity maps. Geophys. J. Int. 166 (2), 732-744. https://doi.org/10.1111/j.1365246X.2006.03028.x.

[24] Yao H.J., Gouédard P., Collins J.A., McGuire J.J., van der Hilst R.D., 2011. Structure of young East Pacific Rise lithosphere from ambient noise correlation analysis of fundamental- and higher-mode Scholte-Rayleigh waves. Compt. Rendus Geosci. 343 (8-9), 571-583. https://doi.org/10.1016/j.crte.2011.04.004.

[25] Zeng X., Thurber C.H., 2016. A graphics processing unit implementation for time-frequency phase-weighted stacking. Seismol Res. Lett. 87 (2A), 358-362. https://doi.org/10.1785/0220150192.

[26] Zhang Y., Yao H., Yang H.-Y., Cai H.-T., Fang H., Xu J., Jin X., Kuo-Chen H., Liang W.-T., Chen K.-X., 2018. 3-D crustal shear-wave velocity structure of the taiwan strait and fujian, SE China, revealed by ambient noise tomography. J. Geophys. Res. Solid Earth 123 (9), 8016-8031. https://doi.org/10.1029/2018JB015938.

[27] Zheng J., Liu G., 2024. GPUPRSI: GPU implementation of seismic interferometry for retrieving reflection responses from passive source seismic recordings. Comput. Geosci. 190, 105654. https://doi.org/10.1016/j.cageo.2024.105654.

[28] Zhou J., W, Q., Wu C., Sun G.,2021 A high performance computing method for noise cross-correlation functions of seismic data. In: IEEE intl conf on parallel & distributed processing with applications Big Data & Cloud Computing, Sustainable Computing & Communications, Social Computing & Networking.(ISPA/BDCloud/SocialCom/ SustainCom)*, New York City, NY, USA, 2021, pp. 1179-1182. https://doi.org/10.1109/ISPA-BDCloud-SocialCom-SustainCom52081.2021.00162.