1# libCEED: the CEED API Library 2 3[](https://travis-ci.org/CEED/libCEED) 4[](https://codecov.io/gh/CEED/libCEED/) 5[](https://opensource.org/licenses/BSD-2-Clause) 6[](https://codedocs.xyz/CEED/libCEED/) 7 8## Code for Efficient Extensible Discretization 9 10This repository contains an initial low-level API library for the efficient 11high-order discretization methods developed by the ECP co-design [Center for 12Efficient Exascale Discretizations (CEED)](http://ceed.exascaleproject.org). 13While our focus is on high-order finite elements, the approach is mostly 14algebraic and thus applicable to other discretizations in factored form, as 15explained in the API documentation portion of the [Doxygen documentation](https://codedocs.xyz/CEED/libCEED/md_doc_libCEEDapi.html). 16 17One of the challenges with high-order methods is that a global sparse matrix is 18no longer a good representation of a high-order linear operator, both with 19respect to the FLOPs needed for its evaluation, as well as the memory transfer 20needed for a matvec. Thus, high-order methods require a new "format" that still 21represents a linear (or more generally non-linear) operator, but not through a 22sparse matrix. 23 24The goal of libCEED is to propose such a format, as well as supporting 25implementations and data structures, that enable efficient operator evaluation 26on a variety of computational device types (CPUs, GPUs, etc.). This new operator 27description is based on algebraically [factored form](https://codedocs.xyz/CEED/libCEED/md_doc_libCEEDapi.html), 28which is easy to incorporate in a wide variety of applications, without significant 29refactoring of their own discretization infrastructure. 30 31The repository is part of the [CEED software suite][ceed-soft], a collection of 32software benchmarks, miniapps, libraries and APIs for efficient exascale 33discretizations based on high-order finite element and spectral element methods. 34See http://github.com/ceed for more information and source code availability. 35 36The CEED research is supported by the [Exascale Computing Project][ecp] 37(17-SC-20-SC), a collaborative effort of two U.S. Department of Energy 38organizations (Office of Science and the National Nuclear Security 39Administration) responsible for the planning and preparation of a [capable 40exascale ecosystem](https://exascaleproject.org/what-is-exascale), including 41software, applications, hardware, advanced system engineering and early testbed 42platforms, in support of the nation’s exascale computing imperative. 43 44For more details on the CEED API see http://ceed.exascaleproject.org/ceed-code/. 45 46## Building 47 48The CEED library, `libceed`, is a C99 library with no external dependencies. It 49can be built using 50 51 make 52 53or, with optimization flags 54 55 make OPT='-O3 -march=skylake-avx512 -ffp-contract=fast' 56 57These optimization flags are used by all languages (C, C++, Fortran) and this 58makefile variable can also be set for testing and examples (below). 59 60The library attempts to automatically detect support for the AVX 61instruction set using gcc-style compiler options for the host. 62Support may need to be manually specified via 63 64 make AVX=1 65 66or 67 68 make AVX=0 69 70if your compiler does not support gcc-style options, if you are cross 71compiling, etc. 72 73## Testing 74 75The test suite produces [TAP](https://testanything.org) output and is run by: 76 77 make test 78 79or, using the `prove` tool distributed with Perl (recommended) 80 81 make prove 82 83## Backends 84 85There are multiple supported backends, which can be selected at runtime in the examples: 86 87| CEED resource | Backend | 88| :----------------------- | :------------------------------------------------ | 89| `/cpu/self/ref/serial` | Serial reference implementation | 90| `/cpu/self/ref/blocked` | Blocked refrence implementation | 91| `/cpu/self/tmpl` | Backend template, dispatches to /cpu/self/blocked | 92| `/cpu/self/avx` | Blocked AVX implementation | 93| `/cpu/self/xsmm/serial` | Serial LIBXSMM implementation | 94| `/cpu/self/xsmm/blocked` | Blocked LIBXSMM implementation | 95| `/cpu/occa` | Serial OCCA kernels | 96| `/gpu/occa` | CUDA OCCA kernels | 97| `/omp/occa` | OpenMP OCCA kernels | 98| `/ocl/occa` | OpenCL OCCA kernels | 99| `/gpu/cuda` | Pure CUDA kernels | 100| `/gpu/magma` | CUDA MAGMA kernels | 101 102 103The `/cpu/self/*/serial` backends process one element at a time and are intended for meshes 104with a smaller number of high order elements. The `/cpu/self/*/blocked` backends process 105blocked batches of eight interlaced elements and are intended for meshes with higher numbers 106of elements. 107 108The `/cpu/self/ref/*` backends are written in pure C and provide basic functionality. 109 110The `/cpu/self/avx` backend relies upon AVX instructions to provide vectorized CPU performance. 111 112The `/cpu/self/xsmm/*` backends relies upon the [LIBXSMM](http://github.com/hfp/libxsmm) package 113to provide vectorized CPU performance. 114 115The `/*/occa` backends rely upon the [OCCA](http://github.com/libocca/occa) package to provide 116cross platform performance. 117 118The `/gpu/cuda` backend provides GPU performance strictly using CUDA. 119 120The `/gpu/magma` backend relies upon the [MAGMA](https://bitbucket.org/icl/magma) package. 121 122## Examples 123 124libCEED comes with several examples of its usage, ranging from standalone C 125codes in the `/examples/ceed` directory to examples based on external packages, 126such as MFEM, PETSc, and Nek5000. Nek5000 v18.0 or greater is required. 127 128To build the examples, set the `MFEM_DIR`, `PETSC_DIR` and `NEK5K_DIR` variables 129and run: 130 131```console 132# libCEED examples on CPU and GPU 133cd examples/ceed 134make 135./ex1 -ceed /cpu/self 136./ex1 -ceed /gpu/occa 137cd ../.. 138 139# MFEM+libCEED examples on CPU and GPU 140cd examples/mfem 141make 142./bp1 -ceed /cpu/self -no-vis 143./bp1 -ceed /gpu/occa -no-vis 144cd ../.. 145 146# PETSc+libCEED examples on CPU and GPU 147cd examples/petsc 148make 149./bp1 -ceed /cpu/self 150./bp1 -ceed /gpu/occa 151cd ../.. 152 153# Nek+libCEED examples on CPU and GPU 154cd examples/nek5000 155./make-nek-examples.sh 156./run-nek-example.sh -ceed /cpu/self -b 3 157./run-nek-example.sh -ceed /gpu/occa -b 3 158cd ../.. 159``` 160 161The above code assumes a GPU-capable machine with the OCCA backend 162enabled. Depending on the available backends, other Ceed resource specifiers can 163be provided with the `-ceed` option. 164 165## Benchmarks 166 167A sequence of benchmarks for all enabled backends can be run using 168 169```console 170make benchmarks 171``` 172 173The results from the benchmarks are stored inside the `benchmarks/` directory 174and they can be viewed using the commands (requires python with matplotlib): 175 176```console 177cd benchmarks 178python postprocess-plot.py petsc-bp1-*-output.txt 179python postprocess-plot.py petsc-bp3-*-output.txt 180``` 181 182Using the `benchmarks` target runs a comprehensive set of benchmarks which may 183take some time to run. Subsets of the benchmarks can be run using targets such 184as `make bench-petsc-bp1`, or `make bench-petsc-bp3`. 185 186For more details about the benchmarks, see 187[`benchmarks/README.md`](benchmarks/README.md) 188 189 190## Install 191 192To install libCEED, run 193 194 make install prefix=/usr/local 195 196or (e.g., if creating packages), 197 198 make install prefix=/usr DESTDIR=/packaging/path 199 200Note that along with the library, libCEED installs kernel sources, e.g. OCCA 201kernels are installed in `$prefix/lib/okl`. This allows the OCCA backend to 202build specialized kernels at run-time. In a normal setting, the kernel sources 203will be found automatically (relative to the library file `libceed.so`). 204However, if that fails (e.g. if `libceed.so` is moved), one can copy (cache) the 205kernel sources inside the user OCCA directory, `~/.occa` using 206 207 $(OCCA_DIR)/bin/occa cache ceed $(CEED_DIR)/lib/okl/*.okl 208 209This will allow OCCA to find the sources regardless of the location of the CEED 210library. One may occasionally need to clear the OCCA cache, which can be accomplished 211by removing the `~/.occa` directory or by calling `$(OCCA_DIR)/bin/occa clear -a`. 212 213### pkg-config 214 215In addition to library and header, libCEED provides a [pkg-config][pkg-config1] 216file that can be used to easily compile and link. [For example][pkg-config2], if 217`$prefix` is a standard location or you set the environment variable 218`PKG_CONFIG_PATH`, 219 220 cc `pkg-config --cflags --libs ceed` -o myapp myapp.c 221 222will build `myapp` with libCEED. This can be used with the source or 223installed directories. Most build systems have support for pkg-config. 224 225## Contact 226 227You can reach the libCEED team by emailing [ceed-users@llnl.gov](mailto:ceed-users@llnl.gov) 228or by leaving a comment in the [issue tracker](https://github.com/CEED/libCEED/issues). 229 230## Copyright 231 232The following copyright applies to each file in the CEED software suite, unless 233otherwise stated in the file: 234 235> Copyright (c) 2017, Lawrence Livermore National Security, LLC. Produced at the 236> Lawrence Livermore National Laboratory. LLNL-CODE-734707. All Rights reserved. 237 238See files LICENSE and NOTICE for details. 239 240[ceed-soft]: http://ceed.exascaleproject.org/software/ 241[ecp]: https://exascaleproject.org/exascale-computing-project 242[pkg-config1]: https://en.wikipedia.org/wiki/Pkg-config 243[pkg-config2]: https://people.freedesktop.org/~dbn/pkg-config-guide.html#faq 244
){kind=link}
){kind=link}
){kind=link}
){kind=link}