performance (full) in projects: libCEED

Searched full:performance (Results 1 – 25 of 35) sorted by relevance

/libCEED/doc/sphinx/source/
H A D	intro.md	`8 This metric for computational efficiency made sense historically, when the performance was mostly l… 9 A more relevant performance plot for current state-of-the-art high-performance machines (for which … 21 Furthermore, software packages that provide high-performance implementations have often been specia… 22 … can unobtrusively be integrated in new and legacy software to provide performance portable interf…`
H A D	releasenotes.md	117 - Various performance enhancements, analytic matrix-free and assembled Jacobian, and PETSc solver c… 188 …s to `CeedQFunctionContext` data as an optional feature to improve GPU performance. By default, ca… 214 ### Performance improvements 218 - Solid mechanics mini-app updated to explore the performance impacts of various formulations in th… 240 ### Performance improvements 243 - New HIP backends for improved tensor basis performance: `/gpu/hip/shared` and `/gpu/hip/gen`. 276 ### Performance improvements 278 - OCCA backend rebuilt to facilitate future performance enhancements. 319 ### Performance Improvements 321 - MAGMA backend performance optimization and non-tensor bases. [all …]
H A D	libCEEDdev.md	18 If there are no performance specific considerations, it is generally recommended to include a basic… 78 These backends use shared memory to improve performance for the {ref}`CeedBasis` kernels. 83 …o apply the action of the {ref}`CeedOperator`, significantly improving performance by eliminating … 87 These backends provide better performance for {ref}`CeedBasis` kernels but do not have the improvem…
H A D	references.bib	`104 title = {Roofline: an insightful visual performance model for multicore architectures}, 128 …title = {On the Order of Accuracy and Numerical Performance of Two Classes of Finite Volume WE…`
/libCEED/doc/papers/joss/
H A D	paper.md	4 - high-performance computing 76 `libCEED` provides portable performance via run-time selection of implementations optimized for CPU… 81 …ations and discretization libraries, `libCEED` provides a platform for performance engineering and… 147 # Performance benchmarks 149 …performance of high-order finite element implementations [@Fischer2020scalability; @CEED-ECP-paper… 151 …Performance for BP3 using the \texttt{xsmm/blocked} backend on a 2-socket AMD EPYC 7452 (32-core, …
H A D	paper.bib	50 journal = {International Journal of High Performance Computing Applications}, 111 title = {{CEED ECP Milestone Report: Improve performance and 132 title={Scalability of high-performance PDE solvers}, 134 journal={The International Journal of High Performance Computing Applications}, 189 …title = {{H}igh-performance operator evaluations with ease of use: lib{C}{E}{E}{D}'s {P}ython … 240 …performance of the interpreter is often a barrier when scaling to larger data sets. This paper pre… 397 title={Roofline: an insightful visual performance model for multicore architectures},
/libCEED/julia/LibCEED.jl/docs/src/
H A D	Misc.md	`5 performance, it is important to use specialized versions of these operations for 18 result in a type instability, and give poor performance.`
H A D	index.md	`111 The macro version can provide better performance if a closure is required, and`
/libCEED/
H A D	README.md	13 libCEED provides fast algebra for element-based discretizations, designed for performance portabili… 192 … `/cpu/self/opt/` backends are written in pure C and use partial e-vectors to improve performance. 194 The `/cpu/self/avx/` backends rely upon AVX instructions to provide vectorized CPU performance. 201 …on the [LIBXSMM](https://github.com/libxsmm/libxsmm) package to provide vectorized CPU performance. 205 The `/gpu/cuda/` backends provide GPU performance strictly using CUDA. 207 The `/gpu/hip/` backends provide GPU performance strictly using HIP. 211 The `/gpu/sycl/*` backends provide GPU performance strictly using SYCL. 227 …e libCEED backends use non-deterministic operations, such as `atomicAdd` for increased performance. 474 …title = {{H}igh-performance operator evaluations with ease of use: {libCEED}'s {P}ython interf…
H A D	.gitlab-ci.yml	45 …desiredSize":"larger"},{"name":"Requests","value":4,"desiredSize":"smaller"}]}]' > performance.json 59 performance: performance.json 95 …desiredSize":"larger"},{"name":"Requests","value":4,"desiredSize":"smaller"}]}]' > performance.json 136 performance: performance.json 343 …desiredSize":"larger"},{"name":"Requests","value":4,"desiredSize":"smaller"}]}]' > performance.json 372 # performance: performance.json 452 …desiredSize":"larger"},{"name":"Requests","value":4,"desiredSize":"smaller"}]}]' > performance.json 476 performance: performance.json
H A D	setup.py	`68 sparse matrices, and can achieve very high performance on modern CPU and GPU`
H A D	CITATION.cff	`173 title: "High-performance operator evaluations with ease of use: libCEED's Python interface"`
/libCEED/examples/petsc/
H A D	bps.c	`53 // Main body of program, called in a loop for performance benchmarking purposes 213 // First run's performance log is not considered for benchmarking purposes in RunWithDM() 231 // -- Performance logging in RunWithDM() 239 // -- Performance logging in RunWithDM() 262 PetscCall(PetscPrintf(rp->comm, " Performance:\n")); in RunWithDM()`
H A D	bpssphere.c	`243 // -- Performance logging in main() 252 // -- Performance logging in main() 275 PetscCall(PetscPrintf(comm, " Performance:\n")); in main()`
H A D	bpsswarm.c	`335 // -- Performance logging in main() 344 // -- Performance logging in main() 367 PetscCall(PetscPrintf(comm, " Performance:\n")); in main()`
H A D	multigrid.c	`463 // -- Performance logging in main() 472 // -- Performance logging in main() 502 PetscCall(PetscPrintf(comm, " Performance:\n")); in main()`
H A D	bpsraw.c	`703 // First run's performance log is not considered for benchmarking purposes in main() 721 // -- Performance logging in main() 730 // -- Performance logging in main() 753 PetscCall(PetscPrintf(comm, " Performance:\n")); in main()`
/libCEED/examples/solids/
H A D	elasticity.c	`115 // Performance logging in main() 209 // Performance logging in main() 215 // Performance logging in main() 283 // Performance logging in main() 374 // Performance logging in main() 573 // Performance logging in main() 593 // Performance logging in main() 652 // Performance logging in main() 727 " Performance:\n" in main()`
/libCEED/julia/LibCEED.jl/src/
H A D	CeedVector.jl	266 Because of performance issues involving closures, if `f` is a complex operation, it may be 267 more efficient to use the macro version `@witharray` (cf. the section on "Performance of 269 documentation](https://docs.julialang.org/en/v1/manual/performance-tips) and related [GitHub
/libCEED/benchmarks/
H A D	README.md	`3 This directory contains benchmark problems for performance evaluation of libCEED`
/libCEED/doc/bib/
H A D	references.bib	`48 …title = {{H}igh-performance operator evaluations with ease of use: lib{C}{E}{E}{D}'s {P}ython …`
/libCEED/rust/libceed/
H A D	README.md	`6 This crate provides an interface to [libCEED](https://libceed.org), which is a performance-portable…`
/libCEED/include/ceed/jit-source/magma/
H A D	magma-basis-interp-deriv-nontensor.h	`34 …// unrolling this loop yields dramatic performance drop using hipcc, so let the compiler decide (n… in magma_basis_nontensor_device_n() 74 …// unrolling this loop yields dramatic performance drop using hipcc, so let the compiler decide (n… in magma_basis_nontensor_device_t() 120 …// unrolling this loop yields dramatic performance drop using hipcc, so let the compiler decide (n… in magma_basis_nontensor_device_ta()`
/libCEED/examples/
H A D	README.md	`14 …retizations (CEED) uses Bakeoff Problems (BPs) to test and compare the performance of high-order f…`
/libCEED/backends/sycl-ref/
H A D	ceed-sycl-ref-basis.sycl.cpp	`108 // Use older version of sycl workgroup barrier for performance reasons in CeedBasisApplyInterp_Sycl() 109 … // Can be updated in future to align with SYCL2020 spec if performance bottleneck is removed in CeedBasisApplyInterp_Sycl() 209 // Use older version of sycl workgroup barrier for performance reasons in CeedBasisApplyGrad_Sycl() 210 … // Can be updated in future to align with SYCL2020 spec if performance bottleneck is removed in CeedBasisApplyGrad_Sycl()`

Project(s)

Full Search
Definition
Symbol
File Path
History
Type