Jonas Thies
YOU?
Author Swipe
View article: Performance of linear solvers in tensor-train format on current multicore architectures
Performance of linear solvers in tensor-train format on current multicore architectures Open
Tensor networks are a class of algorithms aimed at reducing the computational complexity of high-dimensional problems. They are used in an increasing number of applications, from quantum simulations to machine learning. Exploiting data par…
View article: Algebraic temporal blocking for sparse iterative solvers on multi-core CPUs
Algebraic temporal blocking for sparse iterative solvers on multi-core CPUs Open
Sparse linear iterative solvers are essential for many large-scale simulations. Much of the runtime of these solvers is often spent in the implicit evaluation of matrix polynomials via a sequence of sparse matrix-vector products. A variety…
View article: Performance of linear solvers in tensor-train format on current multicore architectures
Performance of linear solvers in tensor-train format on current multicore architectures Open
Tensor networks are a class of algorithms aimed at reducing the computational complexity of high-dimensional problems. They are used in an increasing number of applications, from quantum simulations to machine learning. Exploiting data par…
View article: Algebraic Temporal Blocking for Sparse Iterative Solvers on Multi-Core CPUs
Algebraic Temporal Blocking for Sparse Iterative Solvers on Multi-Core CPUs Open
Sparse linear iterative solvers are essential for many large-scale simulations. Much of the runtime of these solvers is often spent in the implicit evaluation of matrix polynomials via a sequence of sparse matrix-vector products. A variety…
View article: SIMD vectorization for simultaneous solution of locally varying linear systems with multiple right-hand sides
SIMD vectorization for simultaneous solution of locally varying linear systems with multiple right-hand sides Open
Developments in numerical simulation of flows and high-performance computing influence one another. More detailed simulation methods create a permanent need for more computational power, while new hardware developments often require change…
View article: Tensor product scheme for computing bound states of the quantum mechanical three-body problem
Tensor product scheme for computing bound states of the quantum mechanical three-body problem Open
We develop a computationally and numerically efficient method to calculate binding energies and corresponding wave functions of quantum mechanical three-body problems in low dimensions. Our approach exploits the tensor structure of the mul…
View article: SIMD vectorization for simultaneous solution of locally varying linear systems with multiple right hand sides
SIMD vectorization for simultaneous solution of locally varying linear systems with multiple right hand sides Open
Developments in numerical simulation of flows and high performance computing influence one another. More detailed simulation methods create a permanent need for more computational power, while new hardware developments often require change…
View article: (R)SE challenges in HPC
(R)SE challenges in HPC Open
We discuss some specific software engineering challenges in the field of high-performance computing, and argue that the slow adoption of SE tools and techniques is at least in part caused by the fact that these do not address the HPC chall…
View article: Tensor Product Scheme for Computing Bound States of the Quantum Mechanical Three-Body Problem
Tensor Product Scheme for Computing Bound States of the Quantum Mechanical Three-Body Problem Open
We develop a computationally and numerically efficient method to calculate binding energies and corresponding wave functions of quantum mechanical three-body problems in low dimensions. Our approach exploits the tensor structure of the mul…
View article: Exploiting Tensor Structure for Computing Bound States of the Quantum Mechanical Three-Body Problem.
Exploiting Tensor Structure for Computing Bound States of the Quantum Mechanical Three-Body Problem. Open
We develop a computationally and numerically efficient approach to determine binding energies and corresponding wave functions of a quantum-mechanical three-body problem in low dimensions. Our approach exploits the tensor structure intrins…
View article: Performance of the low-rank tensor-train SVD (TT-SVD) for large dense tensors on modern multi-core CPUs
Performance of the low-rank tensor-train SVD (TT-SVD) for large dense tensors on modern multi-core CPUs Open
There are several factorizations of multi-dimensional tensors into lower-dimensional components, known as `tensor networks'. We consider the popular `tensor-train' (TT) format and ask: How efficiently can we compute a low-rank approximatio…
View article: Performance of low-rank approximations in tensor train format (TT-SVD) for large dense tensors.
Performance of low-rank approximations in tensor train format (TT-SVD) for large dense tensors. Open
There are several factorizations of multi-dimensional tensors into lower-dimensional components, known as `tensor networks'. We consider the popular `tensor-train' (TT) format and ask, how efficiently can we compute a low-rank approximatio…
View article: Performance engineering for real and complex tall & skinny matrix multiplication kernels on GPUs
Performance engineering for real and complex tall & skinny matrix multiplication kernels on GPUs Open
General matrix-matrix multiplications with double-precision real and complex entries (DGEMM and ZGEMM) in vendor-supplied BLAS libraries are best optimized for square matrices but often show bad performance for tall & skinny matrices, whic…
View article: A Recursive Algebraic Coloring Technique for Hardware-efficient Symmetric Sparse Matrix-vector Multiplication
A Recursive Algebraic Coloring Technique for Hardware-efficient Symmetric Sparse Matrix-vector Multiplication Open
The symmetric sparse matrix-vector multiplication (SymmSpMV) is an important building block for many numerical linear algebra kernel operations or graph traversal applications. Parallelizing SymmSpMV on today’s multicore platforms with up …
View article: ESSEX: Equipping Sparse Solvers For Exascale
ESSEX: Equipping Sparse Solvers For Exascale Open
The ESSEX project has investigated programming concepts, data structures, and numerical algorithms for scalable, efficient, and robust sparse eigenvalue solvers on future heterogeneous exascale systems. Starting without the burden of legac…
View article: Practical HPC Software Engineering for Research
Practical HPC Software Engineering for Research Open
The talk gives an introduction to software engineering topics like version control and unit testing. The focus in the second part is on performance engineering for numerical codes, and the two topics are brought together by presenting usef…
View article: CRAFT: A Library for Easier Application-Level Checkpoint/Restart and Automatic Fault Tolerance
CRAFT: A Library for Easier Application-Level Checkpoint/Restart and Automatic Fault Tolerance Open
In order to efficiently use the future generations of supercomputers, fault tolerance and power consumption are two of the prime challenges anticipated by the High Performance Computing (HPC) community. Checkpoint/Restart (CR) has been and…
View article: Benefits from using mixed precision computations in the ELPA-AEO and ESSEX-II eigensolver projects
Benefits from using mixed precision computations in the ELPA-AEO and ESSEX-II eigensolver projects Open
We first briefly report on the status and recent achievements of the ELPA-AEO (Eigenvalue Solvers for Petaflop Applications - Algorithmic Extensions and Optimizations) and ESSEX II (Equipping Sparse Solvers for Exascale) projects. In both …
View article: The Jacobi-Davidson Eigensolver on GPU Clusters
The Jacobi-Davidson Eigensolver on GPU Clusters Open
Compared to multi-core processors, GPUs typically offer a higher memory bandwidth, which
\nmakes them attractive for memory-bounded codes like sparse linear and eigenvalue solvers.
\nThe fundamental performance issue we encounter when impl…
View article: Algorithmic Developments and Software Engineering for Scalable Sparse Eigensolvers in the DFG Project ESSEX
Algorithmic Developments and Software Engineering for Scalable Sparse Eigensolvers in the DFG Project ESSEX Open
In the German Research Foundation (DFG) project ESSEX (Equipping Sparse Solvers for Exascale), we develop scalable sparse eigensolver libraries for large quantum physics problems. Partners in ESSEX are the Universities of Erlangen, Greifsw…
View article: Software and Performance Engineering for Iterative Eigensolvers
Software and Performance Engineering for Iterative Eigensolvers Open
The complexity of the latest HPC architectures increasingly limits the productivity of researchers in numerical
\nalgorithms and the `time to market' for parallel algorithms. Implementing a new method on a supercomputer today
\ninvolves …
View article: Employing HPC for Analyzing Nonlinear PDE Systems Beyond Simulation
Employing HPC for Analyzing Nonlinear PDE Systems Beyond Simulation Open
We review techniques of numerical bifurcation and stability analysis with examples from computational fluid dynamics and biology. The methodology allows insight into the complete dynamics of nonlinear PDE systems, where standard simulation…
View article: A Software Infrastructure for Solving Quantum Physics Problems on Extremely Parallel Systems
A Software Infrastructure for Solving Quantum Physics Problems on Extremely Parallel Systems Open
In the German Research Foundation (DFG) project ESSEX (Equipping Sparse Solvers for Exascale), we develop scalable sparse eigensolver libraries for large quantum physics problems. Partners in ESSEX are the Universities of Erlangen, Greifsw…