Publications

Specialized hardware accelerators have gained popularity in high-performance computing as a way to increase the performance and power efficiency characteristics of the systems. However, a large number of scientific applications running on these systems are written in Fortran, which does not provide a performance-portable vendor-agnostic API to target heterogeneous architectures. In this paper, we use the PSyclone source-to-source code generation and transformation system to automatically translate a subset of the Fortran language to OpenCL for weather and climate applications conforming to the PSyKAl kernel-based parallelism model. This allows Fortran applications to take advantage of the mature accelerator portability and runtime compilation capabilities of OpenCL. However, a direct translation does not always produce optimal code for each device. To improve the performance portability of the generated code, the presented approach also allows application developers to provide a list of additional code transformations needed to make the generated OpenCL kernels appropriate for each target architecture. We tested the system with the NEMOLite2D Fortran application, which produced competitive performance portability results for CPU and GPU platforms from different vendors and provided an initial port of the application to an FPGA system.

The need for compilers to generate highly vectorized code is at an all-time high with the increasing vectorization capabilities of modern processors. To this end, the information that compilers have at their disposal, either through code analysis or via user annotations, is instrumental for auto-vectorization, and hence for the overall performance. However, the information that is available to compilers at compile time and its accuracy varies greatly, as does the resulting performance of vectorizing compilers. Benchmarks like the Test Suite for Vectorizing Compilers (TSVC) have been developed to evaluate the vectorization capability of such compilers. The overarching approach of TSVC and similar benchmarks is to evaluate the compilers under the best possible scenario (i.e., assuming that compilers have access to all useful contextual information at compile time). Although this idealistic view is useful to observe the capability of compilers for auto-vectorization, it is not a true reflection of the conditions found in real-world applications. In this article, we propose a novel method for evaluating the auto-vectorization capability of compilers. Instead of assuming that compilers have access to a wealth of information at compile time, we formulate a method to objectively supply or withdraw information that would otherwise aid the compiler in the auto-vectorization process. This method is orthogonal to the approach adopted by TSVC, and as such, it provides the means of assessing the capabilities of modern vectorizing compilers in a more detailed way.

The Hartree Centre has been established as a UK focus for industrial engagement. STFC has acquired a new IBM system based on the OpenPOWER architecture, comprising 32 nodes with POWER8 CPUs and NVIDIA Kepler K80 GPUs. We report early evaluation of the system using some real applications based on the Lattice Boltzmann Method, Direct Numerical Simulation of Turbulence and using FFTs. No optimisation has been carried out yet, but results are encouraging with performance comparable or better on a per core basis to Intel IvyBridge CPUs. Use of the GPUs for suitable algorithms such as Lattice Boltzmann kernels and for FFTs provides further performance enhancements.

The Lattice Boltzmann Equations of DL_MESO have been re-implemented using a Two-Grid algorithm, threaded programming and vectorization in order to effectively utilize novel highly parallel architectures such as the one offered by the Intel Xeon Phi.

In this work the details of a novel highly-modular and parallel Smoothed Particle Hydrodynamics (SPH) code are presented. The aim of this new code is to introduce an Application Program Interface (API) developed using the object-oriented programming approach available within modern Fortran language , the motivation is to maximize the efficiency when variable resolution in SPH is used. Hence, a novel type of data structure has been introduced, making an extensive use of pointers and liked-list algorithms. In the paper details on the different objects introduced and most relevant algorithm to create and delete particles at minimal computational cost are reported. Moreover most relevant algorithms are described and efficiency of the parallelization is discussed considering different architectures, in particular a first attempt has been made to run the code on Intel Xeon Phi.

The Coupled Coherent States is a quantum molecular dynamics method which solves the time-dependent Schrödinger equation in a basis of semi-classical trajectories in an attempt to avoid exponential scaling. This makes it able to simulate and study the time evolution of photochemical reactions. This current project studies the performance of the Coupled Coherent States method and explores its potential parallelism. It proposes a parallel version of the simulation in order run the software faster and more accurately. The simulation is divisible in two parts where different parallelisation approaches are used. The generation of the trajectories represents an embarrassingly parallel problem and can be solved with a task distribution solution. The propagation of the trajectories needs to solve a ODE, which represent a highly coupled problem. To solve it, the parallisation of the Runge-Kutta Cash-Karp method, the implementation of a parallel derivation routine and the use of parallel lineal algebra routines (ScaLAPACK) are proposed. The final implementation achieves a speedup of x10 in the first part of the simulation, the generation of trajectories, and a x4.5 in the second part, the propagation of the trajectories, when they are run on 12 processors. Suggestions for further development of the parallel CCS code are also presented.

Examine and analyse how we can use the summarized information of parallel applications generated by the Extrae instrumentation tool in order to model the behavior of the application in a feasible way to predict its impact and the performance in different architectures.

This document is part of a wider project which presents an e-Voting platform based on elliptic curve cryptography. It uses an hybrid combination of two of the main e-Voting paradigms to guarantee privacy and security in the counting phase, these are precisely, the mixnets and the homomorphic protocols. This document is focused in the description of the system and the maths and programming needed to solve the homomorphic part of it. In later chapters, there is a comparison between a simple mixing system and our system proposal.