Related papers: Evaluation of disconnected contributions using GPU…
In this thesis we develop techniques to efficiently solve numerical Partial Differential Equations (PDEs) using Graphical Processing Units (GPUs). Focus is put on both performance and re--usability of the methods developed, to this end a…
In this work, we have explored the advantages and drawbacks of using GPUs instead of CPUs in the calculation of a standard 2-point correlation function algorithm, which is useful for the analysis of Large Scale Structure of galaxies. Taking…
The covariant parton model is generalized to describe quark correlators in a systematic way. Previous results are reproduced for the T-even leading-twist transverse momentum dependent parton distribution functions (TMDs), and for the first…
Disconnected diagrams give crucial contributions to the physics of flavor singlet hadrons and to scalar form factors of non-singlet hadrons. Naive lattice calculation of the disconnected diagrams, however, requires a huge number of fermion…
We describe a method for parallelizing the lexicographic enumeration algorithm for the factorization set of an element in a numerical semigroup via bounds. This enables the use of GPU and distributed computing methods. We provide a CUDA…
As graphs grow larger, full-batch GNN training becomes hard for single GPU memory. Therefore, to enhance the scalability of GNN training, some studies have proposed sampling-based mini-batch training and distributed graph learning. However,…
We evaluate the strange nucleon electromagnetic form factors using an ensemble of gauge configurations generated with two degenerate maximally twisted mass clover-improved fermions with mass tuned to approximately reproduce the physical…
While Graph Neural Networks (GNNs) have recently become the de facto standard for modeling relational data, they impose a strong assumption on the availability of the node or edge features of the graph. In many real-world applications,…
Lossy compression, widely used by scientists to reduce data from simulations, experiments, and observations, can distort features of interest even under bounded error. Such distortions may compromise downstream analyses and lead to…
Partition of unity methods (PUMs) on graphs are simple and highly adaptive auxiliary tools for graph signal processing. Based on a greedy-type metric clustering and augmentation scheme, we show how a partition of unity can be generated in…
We present the design and optimization of a linear solver on General Purpose GPUs for the efficient and high-throughput evaluation of the marginalized graph kernel between pairs of labeled graphs. The solver implements a preconditioned…
We present distributed algorithms for training dynamic Graph Neural Networks (GNN) on large scale graphs spanning multi-node, multi-GPU systems. To the best of our knowledge, this is the first scaling study on dynamic GNN. We devise…
Nucleon form factors at high momentum transfer |t| are treated in the framework of generalized parton distributions (GPD's). The possibility of obtaining information about parton high transverse momentum components by application of GPD's…
In this paper we present a methodology for data accesses when solving batches of Tridiagonal and Pentadiagonal matrices that all share the same left-hand-side (LHS) matrix. The intended application is to the numerical solution of Partial…
We report on our exploratory study for the evaluation of the parton distribution functions from lattice QCD, based on a new method proposed in Ref.~arXiv:1305.1539. Using the example of the nucleon, we compare two different methods to…
We present detailed numerical analysis of the unintegrated double gluon distribution which includes the dependence on the transverse momenta of partons. The unintegrated double gluon distribution was obtained following the…
We describe the GPU implementation of shifted or multimass iterative solvers for sparse linear systems of the sort encountered in lattice gauge theory. We provide a generic tool that can be used by those without GPU programming experience…
We present results for the nucleon electromagnetic form factors using an ensemble of maximally twisted mass clover-improved fermions with pion mass of about 130 MeV. We use multiple sink-source separations and three analysis methods to…
We provide a graphical calculus for computing averages of tensor network diagrams with respect to the distribution of random vectors containing independent uniform complex phases. Our method exploits the order structure of the partially…
The identification, and subsequent discovery, of fast radio transients through blind-search surveys requires a large amount of processing power, in worst cases scaling as $\mathcal{O}(N^3)$. For this reason, survey data are generally…