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/ref/memcheck` | Memcheck backend, undefined value checks | 92| `/cpu/self/opt/serial` | Serial optimized C implementation | 93| `/cpu/self/opt/blocked` | Blocked optimized C implementation | 94| `/cpu/self/avx/serial` | Serial AVX implementation | 95| `/cpu/self/avx/blocked` | Blocked AVX implementation | 96| `/cpu/self/xsmm/serial` | Serial LIBXSMM implementation | 97| `/cpu/self/xsmm/blocked` | Blocked LIBXSMM implementation | 98| `/cpu/occa` | Serial OCCA kernels | 99| `/gpu/occa` | CUDA OCCA kernels | 100| `/omp/occa` | OpenMP OCCA kernels | 101| `/ocl/occa` | OpenCL OCCA kernels | 102| `/gpu/cuda/ref` | Reference pure CUDA kernels | 103| `/gpu/cuda/reg` | Pure CUDA kernels using one thread per element | 104| `/gpu/cuda/shared` | Optimized pure CUDA kernels using shared memory | 105| `/gpu/magma` | CUDA MAGMA kernels | 106 107The `/cpu/self/*/serial` backends process one element at a time and are intended for meshes 108with a smaller number of high order elements. The `/cpu/self/*/blocked` backends process 109blocked batches of eight interlaced elements and are intended for meshes with higher numbers 110of elements. 111 112The `/cpu/self/ref/*` backends are written in pure C and provide basic functionality. 113 114The `/cpu/self/opt/*` backends are written in pure C and use partial e-vectors to improve performance. 115 116The `/cpu/self/avx/*` backends rely upon AVX instructions to provide vectorized CPU performance. 117 118The `/cpu/self/xsmm/*` backends rely upon the [LIBXSMM](http://github.com/hfp/libxsmm) package 119to provide vectorized CPU performance. The LIBXSMM backend does not use BLAS or MKL; however, 120if LIBXSMM was linked to MKL, this can be specified with the compilation flag `MKL=1`. 121 122The `/cpu/self/ref/memcheck` backend relies upon the [Valgrind](http://valgrind.org/) Memcheck tool 123to help verify that user QFunctions have no undefined values. To use, run your code with 124Valgrind and the Memcheck backend, e.g. `valgrind ./build/ex1 -ceed /cpu/self/ref/memcheck`. A 125'development' or 'debugging' version of Valgrind with headers is required to use this backend. 126 127The `/*/occa` backends rely upon the [OCCA](http://github.com/libocca/occa) package to provide 128cross platform performance. 129 130The `/gpu/cuda/*` backends provide GPU performance strictly using CUDA. 131 132The `/gpu/magma` backend relies upon the [MAGMA](https://bitbucket.org/icl/magma) package. 133 134## Examples 135 136libCEED comes with several examples of its usage, ranging from standalone C 137codes in the `/examples/ceed` directory to examples based on external packages, 138such as MFEM, PETSc, and Nek5000. Nek5000 v18.0 or greater is required. 139 140To build the examples, set the `MFEM_DIR`, `PETSC_DIR` and `NEK5K_DIR` variables 141and run: 142 143```console 144# libCEED examples on CPU and GPU 145cd examples/ceed 146make 147./ex1 -ceed /cpu/self 148./ex1 -ceed /gpu/occa 149cd ../.. 150 151# MFEM+libCEED examples on CPU and GPU 152cd examples/mfem 153make 154./bp1 -ceed /cpu/self -no-vis 155./bp1 -ceed /gpu/occa -no-vis 156cd ../.. 157 158# PETSc+libCEED examples on CPU and GPU 159cd examples/petsc 160make 161./bp1 -ceed /cpu/self 162./bp1 -ceed /gpu/occa 163cd ../.. 164 165cd navier-stokes 166make 167./navierstokes -ceed /cpu/self 168./navierstokes -ceed /gpu/occa 169cd ../.. 170 171# Nek+libCEED examples on CPU and GPU 172cd examples/nek5000 173./make-nek-examples.sh 174./run-nek-example.sh -ceed /cpu/self -b 3 175./run-nek-example.sh -ceed /gpu/occa -b 3 176cd ../.. 177``` 178 179The above code assumes a GPU-capable machine with the OCCA backend 180enabled. Depending on the available backends, other Ceed resource specifiers can 181be provided with the `-ceed` option. 182 183## Benchmarks 184 185A sequence of benchmarks for all enabled backends can be run using 186 187```console 188make benchmarks 189``` 190 191The results from the benchmarks are stored inside the `benchmarks/` directory 192and they can be viewed using the commands (requires python with matplotlib): 193 194```console 195cd benchmarks 196python postprocess-plot.py petsc-bp1-*-output.txt 197python postprocess-plot.py petsc-bp3-*-output.txt 198``` 199 200Using the `benchmarks` target runs a comprehensive set of benchmarks which may 201take some time to run. Subsets of the benchmarks can be run using targets such 202as `make bench-petsc-bp1`, or `make bench-petsc-bp3`. 203 204For more details about the benchmarks, see 205[`benchmarks/README.md`](benchmarks/README.md) 206 207 208## Install 209 210To install libCEED, run 211 212 make install prefix=/usr/local 213 214or (e.g., if creating packages), 215 216 make install prefix=/usr DESTDIR=/packaging/path 217 218Note that along with the library, libCEED installs kernel sources, e.g. OCCA 219kernels are installed in `$prefix/lib/okl`. This allows the OCCA backend to 220build specialized kernels at run-time. In a normal setting, the kernel sources 221will be found automatically (relative to the library file `libceed.so`). 222However, if that fails (e.g. if `libceed.so` is moved), one can copy (cache) the 223kernel sources inside the user OCCA directory, `~/.occa` using 224 225 $(OCCA_DIR)/bin/occa cache ceed $(CEED_DIR)/lib/okl/*.okl 226 227This will allow OCCA to find the sources regardless of the location of the CEED 228library. One may occasionally need to clear the OCCA cache, which can be accomplished 229by removing the `~/.occa` directory or by calling `$(OCCA_DIR)/bin/occa clear -a`. 230 231### pkg-config 232 233In addition to library and header, libCEED provides a [pkg-config][pkg-config1] 234file that can be used to easily compile and link. [For example][pkg-config2], if 235`$prefix` is a standard location or you set the environment variable 236`PKG_CONFIG_PATH`, 237 238 cc `pkg-config --cflags --libs ceed` -o myapp myapp.c 239 240will build `myapp` with libCEED. This can be used with the source or 241installed directories. Most build systems have support for pkg-config. 242 243## Contact 244 245You can reach the libCEED team by emailing [ceed-users@llnl.gov](mailto:ceed-users@llnl.gov) 246or by leaving a comment in the [issue tracker](https://github.com/CEED/libCEED/issues). 247 248## Copyright 249 250The following copyright applies to each file in the CEED software suite, unless 251otherwise stated in the file: 252 253> Copyright (c) 2017, Lawrence Livermore National Security, LLC. Produced at the 254> Lawrence Livermore National Laboratory. LLNL-CODE-734707. All Rights reserved. 255 256See files LICENSE and NOTICE for details. 257 258[ceed-soft]: http://ceed.exascaleproject.org/software/ 259[ecp]: https://exascaleproject.org/exascale-computing-project 260[pkg-config1]: https://en.wikipedia.org/wiki/Pkg-config 261[pkg-config2]: https://people.freedesktop.org/~dbn/pkg-config-guide.html#faq 262
){kind=link}
){kind=link}
){kind=link}
){kind=link}