|
The Harvard University node for the National Nanotechnology Infrastructure Network Computation Project (NNIN/C) has, since 2004, been operating to provide hardware, software, expert support and educational resources to the nanoscience community. NNIN/C at Harvard has hosted several workshops including the international conference in May 2006: Synergy Between Experiment and Computation in Nanoscale Science, which featured over 100 participants from 14 countries.
Users of NNIN/C at Harvard have access to the 6000 node odyssey cluster of the faculty of arts and science (FAS) and the priority nnin queue for the 224 node NNIN/C cluster of AMD Opterons. In addition, users may obtain access to the Orgoglio cluster of Nvidia C1060 Graphical Processing Units, a joint resource of NNIN/C and the Cyber-enabled Discovery Initiative (CDI) at Harvard. The Orgoglio cluster, with its theoretical capacity of 9.6 Tflops, was installed in the summer of 2009. In August of 2009, NNIN/C at Harvard hosted the workshop: Programming the GPU: Introduction to CUDA, a hands-on tutorial-style workshop over three days, to intoroduce users to the massively parallel structure of the GPU and its Nvidia-designed programming language.
The principal areas of expertise of the NNIN/C staff at Harvard are
- electronic structure of semiconductor heterostructures including split-gate GaAs-AlGaAs 2DEG devices and heteroepitaxially-grown semiconductor nanowires of InAs/InP or Si/Ge. The SETE code and the SETEwire code at Harvard are designed to calculate, on an inhomogeneous grid, electron states, wavefunctions and potential profiles for 3D experimental devices.
- Multiscale electronic structure of molecules and nanoparticles in a complex environment (also referred to as QM/CE, or quantum mechanics in a complex environment).
- Highly parallel computing and in particular GPU computing on the Orgoglio cluster. Dramatic advances in computational speed for mature codes for molecular dynamics or the N-body problem have been reported recently. Harvard's NNIN/C seeks to facilitate further such advances.
Hardware Facilities
- Twin AMD Opteron cluster
- 2x28 dual-processor, dual-core AMD 2.0GHZ, 4GB RAM, 64-bit Opteron blades, for a total of 224 cores. Half of them are also equipped with Infiniband interconnects.
- SUN large memory suite
- 4 units of 4-way Opterons from SUN Microsystems, two with 24 GB memory, two with 32 GB memory for a total of 112 GB RAM.
- GPU Cluster: 9.6 Tflops
- Single quad-core Xeon"Harpertownï" processors at 3 GHz
- 16 GB of EEC DDR2 800 RAM
- Two Tesla C1060 GPUs (each with 4GB of RAM)
-
(total of 24 nodes/motherboards, 96 cores, 192 GB RAM, 48 S1070 cards).
- QLogic 24-Port 9024 DDR InfiniBand networking between the nodes.
Software Facilities
- SETE - single electron tunneling elements
- socorro - Ab initio electroni structure code with exact exchange capacity from Sandia National Laboratories
- DL_POLY - General purpose molecular dynamics code
- Octopus - ab initio code, currently being modified to include complex environment by conjoining with SETE code.
- SDTrimSP - ion implantation simulation code
- HARES - real space electronic structure code
- Lumerical package - a proprietary set of codes, including the FDTD suite, for solving Maxwell's equations in complex geometries. Users must possess their own license. However an additional ten licenses for production runs have been generously provided through a donation from Lumerical.
- VASP - Vienna Ab-initio Package Simulation, is also a proprietary code for which users must possess their own license. NNIN/C maintains a copy of the code accessible to those users with a current license.
for further information contact Michael Stopa at stopa@cns.fas.harvard.edu.
|
NNIN Computation Portal
NNIN Computation Sites
Activities and Resources
|
