Browse > Article

A Study on the Performance Improvement of Software Digital Filter using GPU  

Yeom, Jae-Hwan (Radio Astronomy Division, Korea Astronomy and Space Science Institute)
Oh, Se-Jin (Radio Astronomy Division, Korea Astronomy and Space Science Institute)
Roh, Duk-Gyoo (Radio Astronomy Division, Korea Astronomy and Space Science Institute)
Jung, Dong-Kyu (Radio Astronomy Division, Korea Astronomy and Space Science Institute)
Hwang, Ju-Yeon (Radio Astronomy Division, Korea Astronomy and Space Science Institute)
Oh, Chungsik (Radio Astronomy Division, Korea Astronomy and Space Science Institute)
Kim, Hyo-Ryoung (Radio Astronomy Division, Korea Astronomy and Space Science Institute)
Publication Information
Journal of the Institute of Convergence Signal Processing / v.19, no.4, 2018 , pp. 153-161 More about this Journal
Abstract
This paper describes the performance improvement of Software (SW) digital filter using GPU (Graphical Processing Unit). The previous developed SW digital filter has a problem that it operates on a CPU (Central Processing Unit) basis and has a slow speed. The GPU was introduced to filter the data of the EAVN (East Asian VLBI Network) observation to improve the operation speed and to process data with other stations through filtering, respectively. In order to enhance the computational speed of the SW digital filter, NVIDIA Titan V GPU board with built-in Tensor Core is used. The processing speed of about 0.78 (1Gbps, 16MHz BW, 16-IF) and 1.1 (2Gbps, 32MHz BW, 16-IF) times for the observing time was achieved by filtering the 95 second observation data of 2 Gbps (512 MHz BW, 1-IF), respectively. In addition, 2Gbps data is digitally filtered for the 1 and 2Gbps simultaneously observed with KVN (Korean VLBI Network), and compared with the 1Gbps, we obtained similar values such as cross power spectrum, phase, and SNR (Signal to Noise Ratio). As a result, the effectiveness of developed SW digital filter using GPU in this research was confirmed for utilizing the data processing and analysis. In the future, it is expected that the observation data will be able to be filtered in real time when the distributed processing optimization of source code for using multiple GPU boards.
Keywords
East Asian VLBI Network(EAVN); Software Digital Filter; GPU distributed processing; Tensor Core;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Thompson, A. R., Moran, J. M., and Swenson, G. W.Jr., Interferometry and Synthesis in Radio Astronomy, 2nd Ed., (New York:John Wiley & Sons), 2001, pp. 289.
2 Duk-Gyoo Roh, Se-Jin Oh, Jae-Hwan Yeom et al, "Report on Development of Korea-Japan Joint VLBI Correlator and Receiver," KASI, pp. 3-100, 2008.
3 Jae-Hwan Yeom, Se-Jin Oh, Duk-Gyoo Roh et al., "A Study on the Digital Filter Design using Software for Analysis of Observation Data in Radio Astronomy," KICSP, Vol. 16, No. 4, pp. 175-181, 2015.
4 Mark Harris, "Optimizing Cuda," SC07: High Performance Computing With CUDA, 2007.
5 Haidar, A., S. Tomov, J. Dongarra, and N. J. Higham, "Harnessing GPU Tensor Cores for Fast FP16 Arithmetic to Speed up Mixed-Precision Iterative Refinement Solvers," The International Conference for High Performance Computing, Networking, Storage, and Analysis (SC18), 2018.
6 http://www.nvidia.com//ko-kr/titan/titan-v/?nvid=nv-int-tnvptlh-29191
7 A. R. Whitney, "Mark 5 Disk-Based Gbps VLBI Data System," in New Technologies in VLBI, edited by Y. C. Minh, ASP Conference Series, Vol. 306, Astronomical Society of the Pacific, San Francisco, CA, 2003.
8 http://www.vlbi.org/vdif
9 https://www.gnu.org/software/octave
10 Jae-Hwan Yeom, Se-Jin Oh, Duk-Gyoo Roh et al(2009, December), Development of Software Correlator for KJJVC, Journal of Astronomy and Space Science, 26(4) pp. 567-588.   DOI
11 http://www.aips.nrao.edu/index.shtml