http://hdl.handle.net/1842/5890 Mon, 20 May 2013 07:38:51 GMT 2013-05-20T07:38:51Z http://hdl.handle.net/1842/6584 Title: High Performance MP2 for Condensed Phase Simualations Authors: Reyes, Ruyman; Bethune, Iain Abstract: This report describes the results of a PRACE Preparatory Access Type C project to optimise the implementation of Møller-Plesset second order perturbation theory (MP2) in CP2K, to allow it to be used efficiently on the PRACE Research Infrastructure. The work consisted of three stages: firstly serial optimisation of several key computational kernels; secondly, OpenMP implementation of parallel 3D Fourier Transform to support mixedmode MPI/OpenMP use of CP2K; and thirdly - benchmarking the performance gains achieved by new code on HERMIT for a test case representative of proposed production simulations. Consistent speedups of 8% were achieved in the integration kernel routines as a result of the serial optimisation. When using 8 OpenMP threads per MPI process, speedups of up to 10x for the 3D FFT were achieved, and for some combinations of MPI processes and OpenMP threads, overall speedups of 66% for the whole code were measured. As a result of this work, a proposal for full PRACE Project Access has been submitted. Fri, 01 Feb 2013 00:00:00 GMT http://hdl.handle.net/1842/6584 2013-02-01T00:00:00Z http://hdl.handle.net/1842/6547 Title: Adding Parallel I/O to PARA-BMU Authors: Johnson, Nick; Bethune, Iain Abstract: VOX-FE is a voxel-based bone modelling suite. The solver part of the suite - PARA-BMU - currently uses only serial I/O routines which lead to poor scalability. We enhance the code by adding parallel I/O routines based on the netCDF and HDF5 libraries. We demonstrate that this gives a reduction in file sizes of up to 190x and a reduction in wall-clock time of 7x at 512 cores. This paves the way for PARA-BMU to be used effectively on HECToR and allows users to perform research with much larger and more complex models than was previously possible. Mon, 01 Oct 2012 00:00:00 GMT http://hdl.handle.net/1842/6547 2012-10-01T00:00:00Z http://hdl.handle.net/1842/6546 Title: Application Porting from External Developers Authors: Labarta, Jesus; Bethune, Iain; Garcia, Marta; Lopez, Victor; Rilley, Graham Abstract: Different developers outside the project or applications not initially considered in it have been identified and addressed in order to assess how the MPI/SMPSs model could be used in those cases. Different levels of depth have been actually been carried out, and in some of the cases only a part of the application was targeted. The study mostly focused on widening the identification of methodological issues and feedback on the suitability of the model and the functionality of its implementations. This deliverable reports on the experience with 6 apps/codes: Hydro, GROMACS, GADGET, CP2K, MRGENESIS and ShallowWaters. Very interesting feedback has been obtained addressing both methodological programming practices and the characteristics of the SMPSs/OmpSs model and implementation. Description: Deliverable from the TEXT Project describing trial implementation of SMPSs in CP2K Fri, 31 Aug 2012 00:00:00 GMT http://hdl.handle.net/1842/6546 2012-08-31T00:00:00Z http://hdl.handle.net/1842/6545 Title: CP2K – Scalable Atomistic Simulation for the PRACE Community Authors: Bethune, Iain; Carter, Adam; Stratford, Kevin; Korosoglou, Paschalis Abstract: This report describes the work undertaken under PRACE-1IP to support the European scientific communities who make use of CP2K in their research. This was done in two ways – firstly, by improving the performance of the code for a wide range of usage scenarios. The updated code was then tested and installed on the PRACE CURIE supercomputer. We believe this approach both supports existing user communities by delivering better application performance, and demonstrates to potential users the benefits of using optimized and scalable software like CP2K on the PRACE infrastructure. Sun, 01 Jan 2012 00:00:00 GMT http://hdl.handle.net/1842/6545 2012-01-01T00:00:00Z