src/gromacs/analysisdata/framelocaldata.h

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2014,2015,2017,2019, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /*! \internal \file
  36  * \brief
  37  * Defines gmx::AnalysisDataFrameLocalData and supporting types.
  38  *
  39  * \author Teemu Murtola <teemu.murtola@gmail.com>
  40  * \ingroup module_analysisdata
  41  */
  42 #ifndef GMX_ANALYSISDATA_FRAMELOCALDATA_H
  43 #define GMX_ANALYSISDATA_FRAMELOCALDATA_H
  44
  45 #include <algorithm>
  46 #include <numeric>
  47 #include <vector>
  48
  49 #include "gromacs/analysisdata/paralleloptions.h"
  50 #include "gromacs/utility/arrayref.h"
  51 #include "gromacs/utility/gmxassert.h"
  52
  53 namespace gmx
  54 {
  55
  56 //! \addtogroup module_analysisdata
  57 //! \{
  58
  59 /*! \internal
  60  * \brief
  61  * Handle to a single data set within frame-local data array.
  62  *
  63  * Methods in this class do not throw.
  64  *
  65  * \see AnalysisDataFrameLocalData
  66  */
  67 template<typename ValueType>
  68 class AnalysisDataFrameLocalDataSetHandle
  69 {
  70     public:
  71         //! Constructs a handle from an array of values.
  72         explicit AnalysisDataFrameLocalDataSetHandle(ArrayRef<ValueType> values)
  73             : values_(values)
  74         {
  75         }
  76
  77         //! Clears all values in the data set.
  78         void clear()
  79         {
  80             std::fill(values_.begin(), values_.end(), ValueType());
  81         }
  82
  83         //! Accesses a single value in the data set.
  84         ValueType &value(int column)
  85         {
  86             GMX_ASSERT(column >= 0 && column < ssize(values_),
  87                        "Invalid column index");
  88             return values_[column];
  89         }
  90
  91     private:
  92         ArrayRef<ValueType>  values_;
  93 };
  94
  95 /*! \internal
  96  * \brief
  97  * Handle to a single frame data within frame-local data array.
  98  *
  99  * Methods in this class do not throw.
 100  *
 101  * \see AnalysisDataFrameLocalData
 102  */
 103 template<typename ValueType>
 104 class AnalysisDataFrameLocalDataHandle
 105 {
 106     public:
 107         //! Shorthand for the internal array of values.
 108         typedef std::vector<ValueType> ValueArray;
 109         //! Shorthand for a handle to a single data set.
 110         typedef AnalysisDataFrameLocalDataSetHandle<ValueType> DataSetHandle;
 111
 112         //! Constructs a handle from specified frame data.
 113         AnalysisDataFrameLocalDataHandle(const std::vector<int> *dataSetIndices,
 114                                          ValueArray             *values)
 115             : dataSetIndices_(dataSetIndices), values_(values)
 116         {
 117         }
 118
 119         //! Returns the number of data sets in the array.
 120         int dataSetCount() const
 121         {
 122             return dataSetIndices_->size() - 1;
 123         }
 124         //! Clears all values in the frame.
 125         void clear()
 126         {
 127             std::fill(values_->begin(), values_->end(), ValueType());
 128         }
 129
 130         //! Returns a handle for a single data set.
 131         DataSetHandle dataSet(int dataSet)
 132         {
 133             GMX_ASSERT(dataSet >= 0 && dataSet < dataSetCount(),
 134                        "Invalid data set index");
 135             const int firstIndex = (*dataSetIndices_)[dataSet];
 136             const int lastIndex  = (*dataSetIndices_)[dataSet + 1];
 137             return DataSetHandle(makeArrayRef(*values_).
 138                                      subArray(firstIndex, lastIndex-firstIndex));
 139         }
 140         //! Accesses a single value in the frame.
 141         ValueType &value(int dataSet, int column)
 142         {
 143             GMX_ASSERT(dataSet >= 0 && dataSet < dataSetCount(),
 144                        "Invalid data set index");
 145             const int firstIndex = (*dataSetIndices_)[dataSet];
 146             GMX_ASSERT(column >= 0
 147                        && column < (*dataSetIndices_)[dataSet+1] - firstIndex,
 148                        "Invalid column index");
 149             return (*values_)[firstIndex + column];
 150         }
 151
 152     private:
 153         const std::vector<int> *dataSetIndices_;
 154         ValueArray             *values_;
 155 };
 156
 157 /*! \internal \brief
 158  * Container for an array of frame-local values that supports parallel data
 159  * processing.
 160  *
 161  * \tparam ValueType Type of values to store.
 162  *
 163  * This class provides a convenient interface to create an array of frame-local
 164  * data for use in analysis data modules that support parallel processing.
 165  * The object is initialized by setting the desired dimensionality with
 166  * setDataSetCount() and setColumnCount(), followed by a call to init(),
 167  * typically in IAnalysisDataModule::parallelDataStarted(),
 168  *
 169  * After initialization, frameData() can be used to access the data for a given
 170  * frame, independently from other frames.  This works if the assumptions about
 171  * parallelism hold: if `N` is the parallelization factor given for init() with
 172  * AnalysisDataParallelOptions::parallelizationFactor(), then frame `i+N` must
 173  * not be accessed before all processing for frame `i` is finished.
 174  * Technically, the data for different frames is kept in a ring buffer of size
 175  * `N`.
 176  *
 177  * The data for a frame is not cleared after it is reused for a new frame (but
 178  * is initially cleared).  This allows using the data for accumulating values
 179  * over all frames in a lock-free manner.
 180  *
 181  * frameDataSet() is provided for convenience when only a single data set
 182  * needs to be accessed (typically in IAnalysisDataModule::pointsAdded()).
 183  *
 184  * Methods in this class do not throw except where indicated.
 185  *
 186  * \see AnalysisDataFrameLocalData
 187  */
 188 template<typename ValueType>
 189 class AnalysisDataFrameLocalData
 190 {
 191     public:
 192         //! Shorthand for the internal array of values for a frame.
 193         typedef std::vector<ValueType> ValueArray;
 194         //! Shorthand for a handle to a single frame.
 195         typedef AnalysisDataFrameLocalDataHandle<ValueType> FrameHandle;
 196         //! Shorthand for a handle to a single data set.
 197         typedef AnalysisDataFrameLocalDataSetHandle<ValueType> DataSetHandle;
 198
 199         //! Constructs an empty container with a single data set.
 200         AnalysisDataFrameLocalData()
 201         {
 202             dataSetColumns_.resize(2);
 203         }
 204
 205         //! Whether init() has been called.
 206         bool isInitialized() const { return !values_.empty(); }
 207         /*! \brief
 208          * Returns number of independent data frames in this object.
 209          *
 210          * This supports looping over all the frame arrays to, e.g., sum them
 211          * up at the end in accumulation scenarios.
 212          */
 213         int frameCount() const { return values_.size(); }
 214
 215         /*! \brief
 216          * Sets the number of data sets stored for each frame.
 217          *
 218          * \throws std::bad_alloc if out of memory.
 219          *
 220          * If not called, there is a single data set in the object.
 221          * Cannot be called after init().
 222          */
 223         void setDataSetCount(int dataSetCount)
 224         {
 225             GMX_RELEASE_ASSERT(!isInitialized(),
 226                                "Cannot change value count after init()");
 227             GMX_RELEASE_ASSERT(dataSetCount >= 0,
 228                                "Invalid data set count");
 229             dataSetColumns_.resize(dataSetCount + 1);
 230         }
 231         /*! \brief
 232          * Sets the number of columns stored for a data set.
 233          *
 234          * Must be called for each data set that needs to have values,
 235          * otherwise there will be zero columns for that data set.
 236          * Cannot be called after init().
 237          */
 238         void setColumnCount(int dataSet, int columnCount)
 239         {
 240             GMX_RELEASE_ASSERT(!isInitialized(),
 241                                "Cannot change value count after init()");
 242             GMX_RELEASE_ASSERT(dataSet >= 0 && dataSet < ssize(dataSetColumns_) - 1,
 243                                "Invalid data set index");
 244             GMX_RELEASE_ASSERT(columnCount >= 0,
 245                                "Invalid column count");
 246             dataSetColumns_[dataSet + 1] = columnCount;
 247         }
 248
 249         /*! \brief
 250          * Initializes the storage to support specified parallelism.
 251          *
 252          * \throws std::bad_alloc if out of memory.
 253          */
 254         void init(const AnalysisDataParallelOptions &opt)
 255         {
 256             GMX_RELEASE_ASSERT(!isInitialized(), "init() called multiple times");
 257             std::partial_sum(dataSetColumns_.begin(), dataSetColumns_.end(),
 258                              dataSetColumns_.begin());
 259             values_.resize(opt.parallelizationFactor());
 260             typename std::vector<ValueArray>::iterator i;
 261             for (i = values_.begin(); i != values_.end(); ++i)
 262             {
 263                 i->resize(dataSetColumns_.back());
 264             }
 265         }
 266
 267         //! Returns a handle to access data for a frame.
 268         FrameHandle frameData(int frameIndex)
 269         {
 270             GMX_ASSERT(frameIndex >= 0, "Invalid frame index");
 271             GMX_ASSERT(isInitialized(), "Cannot access data before init()");
 272             return FrameHandle(&dataSetColumns_,
 273                                &values_[frameIndex % values_.size()]);
 274         }
 275         //! Returns a handle to access a single data set within a frame.
 276         DataSetHandle frameDataSet(int frameIndex, int dataSet)
 277         {
 278             return frameData(frameIndex).dataSet(dataSet);
 279         }
 280
 281     private:
 282         /*! \brief
 283          * Index to find data sets within a per-frame array in `values_`.
 284          *
 285          * The first entry is always zero, followed by one entry for each data
 286          * set.  Before init(), the data set entries hold the numbers set with
 287          * setColumnCount().  After init(), the data set entries hold the
 288          * indices of the first column for that data set in the per-frame
 289          * arrays in `values_`.
 290          */
 291         std::vector<int>         dataSetColumns_;
 292         /*! \brief
 293          * Data array for each frame.
 294          *
 295          * This is a ring buffer whose size is specified by the desired
 296          * parallelism level.  For each frame, there is a single array of
 297          * values, where the individual data sets are indexed with
 298          * `dataSetColumns_`.
 299          */
 300         std::vector<ValueArray>  values_;
 301 };
 302
 303 //! \}
 304
 305 } // namespace gmx
 306
 307 #endif