Accelerating simulation of nanodevices based on 2D materials by hybrid CPU-GPU parallel computing

We describe our new C-based software that combines atomistic quantum transport with the solution of the 2D Poisson’s equation for nanodevice simulation. A significant acceleration of about ~100× is demonstrated in comparison to our old Matlab code, by using numerical libraries BLAS and LAPACK, shared-memory parallelization with OpenMP, and GPU acceleration based on CUDA libraries. The new code has enabled the analysis of 10 nm-gate length nanotransistors based on silicene nanoribbons, for which we report electronic, transport and device properties.

[1]  M. Sancho,et al.  Highly convergent schemes for the calculation of bulk and surface Green functions , 1985 .

[2]  A. V. van Duin,et al.  A roadmap for electronic grade 2D materials , 2019, 2D Materials.

[3]  Charles L. Lawson,et al.  Basic Linear Algebra Subprograms for Fortran Usage , 1979, TOMS.

[4]  Y. Kamakura,et al.  Theoretical performance estimation of silicene, germanene, and graphene nanoribbon field-effect transistors under ballistic transport , 2014 .

[5]  S. Datta.  Quantum Transport: Atom to Transistor , 2004 .

[6]  Giuseppe Iannaccone,et al.  Electronics based on two-dimensional materials. , 2014, Nature nanotechnology.

[7]  Dapeng Yu,et al.  Tunable bandgap in silicene and germanene. , 2012, Nano letters.

[8]  Diana Adler,et al.  Electronic Transport In Mesoscopic Systems , 2016 .

[9]  L. Dagum,et al.  OpenMP: an industry standard API for shared-memory programming , 1998 .

[10]  Joost VandeVondele,et al.  Pushing back the limit of ab-initio quantum transport simulations on hybrid supercomputers , 2015, SC15: International Conference for High Performance Computing, Networking, Storage and Analysis.

[11]  Wei Liu,et al.  2D Semiconductor FETs—Projections and Design for Sub-10 nm VLSI , 2015, IEEE Transactions on Electron Devices.

[12]  Robert A. van de Geijn,et al.  BLAS (Basic Linear Algebra Subprograms) , 2011, Encyclopedia of Parallel Computing.

[13]  M. Luisier,et al.  Atomistic simulation of nanodevices , 2016, 2016 International Conference on Simulation of Semiconductor Processes and Devices (SISPAD).

[14]  H. Tsuchiya,et al.  Non-equilibrium Green function simulations of graphene, silicene, and germanene nanoribbon field-effect transistors , 2015 .

[15]  M. Luisier,et al.  First-principles simulations of 2-D semiconductor devices: Mobility, I-V characteristics, and contact resistance , 2016, 2016 IEEE International Electron Devices Meeting (IEDM).

[16]  J. Demmel,et al.  Sun Microsystems , 1996 .

[17]  Patrick Vogt,et al.  Silicene: compelling experimental evidence for graphenelike two-dimensional silicon. , 2012, Physical review letters.

[18]  Madan Dubey,et al.  Silicene field-effect transistors operating at room temperature. , 2015, Nature nanotechnology.

[19]  Kevin Skadron,et al.  Scalable parallel programming , 2008, 2008 IEEE Hot Chips 20 Symposium (HCS).

[20]  K. Jacobsen,et al.  The Computational 2D Materials Database: high-throughput modeling and discovery of atomically thin crystals , 2018, 2D Materials.

[21]  Guo Wang.  Do silicene nanoribbons have high carrier mobilities? , 2013 .

[22]  F. Schwierz.  Graphene transistors. , 2010, Nature nanotechnology.

[23]  Andre K. Geim,et al.  Electric Field Effect in Atomically Thin Carbon Films , 2004, Science.