GeantV: From CPU to accelerators

Journal Article · Fri Jan 01 00:00:00 EST 2016 · Journal of Physics. Conference Series

DOI:https://doi.org/10.1088/1742-6596/762/1/012019· OSTI ID:1346932

Amadio, G. ^[1]; Ananya, A. ^[2]; Apostolakis, J. ^[3]; Arora, A. ^[2]; Bandieramonte, M. ^[3]; Bhattacharyya, A. ^[4]; Bianchini, C. ^[5]; Brun, R. ^[3]; Canal, P. ^[6]; Carminati, F. ^[3]; Duhem, L. ^[7]; Elvira, D. ^[6]; Gheata, A. ^[3]; Gheata, M. ^[8]; Goulas, I. ^[3]; Iope, R. ^[1]; Jun, S. ^[6]; Lima, G. ^[6]; Mohanty, A. ^[4]; Nikitina, T. ^[3] ; Novak, M. ^[3]; Pokorski, W. ^[3]; Ribon, A. ^[3]; Sehgal, R. ^[4]; Shadura, O. ^[3]; Vallecorsa, S. ^[3]; Wenzel, S. ^[3]; Zhang, Y. ^[3]

---

1. Univ. Estadual Paulista (Brazil)
2. Indian Institute of Technology (India)
3. European Organization for Nuclear Research (CERN), Geneva (Switzerland)
4. Bhabha Atomic Research Center (India)
5. Univ. Estadual Paulista (Brazil); Mackenzie Presbyterian Univ. (Brazil)
6. Fermi National Accelerator Lab. (FNAL), Batavia, IL (United States)
7. Intel Corp., Santa Clara, CA (United States)
8. European Organization for Nuclear Research (CERN), Geneva (Switzerland); Institute of Space Sciences (Romania)

The GeantV project aims to research and develop the next-generation simulation software describing the passage of particles through matter. While the modern CPU architectures are being targeted first, resources such as GPGPU, Intel© Xeon Phi, Atom or ARM cannot be ignored anymore by HEP CPU-bound applications. The proof of concept GeantV prototype has been mainly engineered for CPU's having vector units but we have foreseen from early stages a bridge to arbitrary accelerators. A software layer consisting of architecture/technology specific backends supports currently this concept. This approach allows to abstract out the basic types such as scalar/vector but also to formalize generic computation kernels using transparently library or device specific constructs based on Vc, CUDA, Cilk+ or Intel intrinsics. While the main goal of this approach is portable performance, as a bonus, it comes with the insulation of the core application and algorithms from the technology layer. This allows our application to be long term maintainable and versatile to changes at the backend side. The paper presents the first results of basket-based GeantV geometry navigation on the Intel© Xeon Phi KNC architecture. We present the scalability and vectorization study, conducted using Intel performance tools, as well as our preliminary conclusions on the use of accelerators for GeantV transport. Lastly, we also describe the current work and preliminary results for using the GeantV transport kernel on GPUs.

• Research Organization: Fermi National Accelerator Lab. (FNAL), Batavia, IL (United States)
• Sponsoring Organization: USDOE Office of Science (SC), High Energy Physics (HEP)
• Grant/Contract Number: AC02-07CH11359
• OSTI ID: 1346932
• Report Number(s): FERMILAB-CONF-16-625-CD; 1499965
• Journal Information: Journal of Physics. Conference Series, Vol. 762, Issue 1; ISSN 1742-6588
• Publisher: IOP Publishing
• Country of Publication: United States
• Language: English

Related Subjects

43 PARTICLE ACCELERATORS
97 MATHEMATICS AND COMPUTING