Use this URL to cite or link to this record in EThOS:
Title: An active-library based investigation into the performance optimisation of linear algebra and the finite element method
Author: Russell, Francis Prem
ISNI:       0000 0004 2705 0606
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2011
Availability of Full Text:
Access from EThOS:
Access from Institution:
In this thesis, I explore an approach called "active libraries". These are libraries that take part in their own optimisation, enabling both high-performance code and the presentation of intuitive abstractions. I investigate the use of active libraries in two domains. Firstly, dense and sparse linear algebra, particularly, the solution of linear systems of equations. Secondly, the specification and solution of finite element problems. Extending my earlier (MEng) thesis work, I describe the modifications to my linear algebra library "Desola" required to perform sparse-matrix code generation. I show that optimisations easily applied in the dense case using code-transformation must be applied at a higher level of abstraction in the sparse case. I present performance results for sparse linear system solvers generated using Desola and compare against an implementation using the Intel Math Kernel Library. I also present improved dense linear-algebra performance results. Next, I explore the active-library approach by developing a finite element library that captures runtime representations of basis functions, variational forms and sequences of operations between discretised operators and fields. Using captured representations of variational forms and basis functions, I demonstrate optimisations to cell-local integral assembly that this approach enables, and compare against the state of the art. As part of my work on optimising local assembly, I extend the work of Hosangadi et al. on common sub-expression elimination and factorisation of polynomials. I improve the weight function presented by Hosangadi et al., increasing the number of factorisations found. I present an implementation of an optimised branch-and-bound algorithm inspired by reformulating the original matrix-covering problem as a maximal graph biclique search problem. I evaluate the algorithm's effectiveness on the expressions generated by our finite element solver.
Supervisor: Kelly, Paul ; Field, Tony Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral