1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
|
/****************************************************************/
/* Parallel Combinatorial BLAS Library (for Graph Computations) */
/* version 1.6 -------------------------------------------------*/
/* date: 6/15/2017 ---------------------------------------------*/
/* authors: Ariful Azad, Aydin Buluc --------------------------*/
/****************************************************************/
/*
Copyright (c) 2010-2017, The Regents of the University of California
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
*/
#include <numeric>
#include "DenseParMat.h"
#include "MPIType.h"
#include "Operations.h"
namespace combblas {
template <class IT, class NT>
template <typename _BinaryOperation>
FullyDistVec< IT,NT > DenseParMat<IT,NT>::Reduce(Dim dim, _BinaryOperation __binary_op, NT identity) const
{
switch(dim)
{
case Column: // pack along the columns, result is a vector of size (global) n
{
// we can use parvec's grid as long as the process grid is square (currently a CombBLAS requirement)
int colneighs = commGrid->GetGridRows(); // including oneself
int colrank = commGrid->GetRankInProcCol();
IT * loclens = new IT[colneighs];
IT * lensums = new IT[colneighs+1](); // begin/end points of local lengths
IT n_perproc = n / colneighs; // length on a typical processor
if(colrank == colneighs-1)
loclens[colrank] = n - (n_perproc*colrank);
else
loclens[colrank] = n_perproc;
MPI_Allgather(MPI_IN_PLACE, 0, MPIType<IT>(), loclens, 1, MPIType<IT>(), commGrid->GetColWorld());
std::partial_sum(loclens, loclens+colneighs, lensums+1); // loclens and lensums are different, but both would fit in 32-bits
std::vector<NT> trarr(loclens[colrank]);
NT * sendbuf = new NT[n];
for(int j=0; j < n; ++j)
{
sendbuf[j] = identity;
for(int i=0; i < m; ++i)
{
sendbuf[j] = __binary_op(array[i][j], sendbuf[j]);
}
}
// The MPI_REDUCE_SCATTER routine is functionally equivalent to:
// an MPI_REDUCE collective operation with count equal to the sum of loclens[i]
// followed by MPI_SCATTERV with sendcounts equal to loclens as well
MPI_Reduce_scatter(sendbuf, trarr.data(), loclens, MPIType<NT>(), MPIOp<_BinaryOperation, NT>::op(), commGrid->GetColWorld());
DeleteAll(sendbuf, loclens, lensums);
IT reallen; // Now we have to transpose the vector
IT trlen = trarr.size();
int diagneigh = commGrid->GetComplementRank();
MPI_Status status;
MPI_Sendrecv(&trlen, 1, MPIType<IT>(), diagneigh, TRNNZ, &reallen, 1, MPIType<IT>(), diagneigh, TRNNZ, commGrid->GetWorld(), &status);
IT glncols = gcols();
FullyDistVec<IT,NT> parvec(commGrid, glncols, identity);
assert((parvec.arr.size() == reallen));
MPI_Sendrecv(trarr.data(), trlen, MPIType<NT>(), diagneigh, TRX, parvec.arr.data(), reallen, MPIType<NT>(), diagneigh, TRX, commGrid->GetWorld(), &status);
return parvec;
break;
}
case Row: // pack along the rows, result is a vector of size m
{
IT glnrows = grows();
FullyDistVec<IT,NT> parvec(commGrid, glnrows, identity);
NT * sendbuf = new NT[m];
for(int i=0; i < m; ++i)
{
sendbuf[i] = std::accumulate( array[i], array[i]+n, identity, __binary_op);
}
NT * recvbuf = parvec.arr.data();
int rowneighs = commGrid->GetGridCols();
int rowrank = commGrid->GetRankInProcRow();
IT * recvcounts = new IT[rowneighs];
recvcounts[rowrank] = parvec.MyLocLength(); // local vector lengths are the ultimate receive counts
MPI_Allgather(MPI_IN_PLACE, 0, MPIType<IT>(), recvcounts, 1, MPIType<IT>(), commGrid->GetRowWorld());
// The MPI_REDUCE_SCATTER routine is functionally equivalent to:
// an MPI_REDUCE collective operation with count equal to the sum of recvcounts[i]
// followed by MPI_SCATTERV with sendcounts equal to recvcounts.
MPI_Reduce_scatter(sendbuf, recvbuf, recvcounts, MPIType<NT>(), MPIOp<_BinaryOperation, NT>::op(), commGrid->GetRowWorld());
delete [] sendbuf;
delete [] recvcounts;
return parvec;
break;
}
default:
{
std::cout << "Unknown reduction dimension, returning empty vector" << std::endl;
return FullyDistVec<IT,NT>(commGrid);
break;
}
}
}
template <class IT, class NT>
template <typename DER>
DenseParMat< IT,NT > & DenseParMat<IT,NT>::operator+=(const SpParMat< IT,NT,DER > & rhs) // add a sparse matrix
{
if(*commGrid == *rhs.commGrid)
{
(rhs.spSeq)->UpdateDense(array, std::plus<double>());
}
else
{
std::cout << "Grids are not comparable elementwise addition" << std::endl;
MPI_Abort(MPI_COMM_WORLD,GRIDMISMATCH);
}
return *this;
}
template <class IT, class NT>
DenseParMat< IT,NT > & DenseParMat<IT,NT>::operator=(const DenseParMat< IT,NT > & rhs) // assignment operator
{
if(this != &rhs)
{
if(array != NULL)
SpHelper::deallocate2D(array, m);
m = rhs.m;
n = rhs.n;
if(rhs.array != NULL)
{
array = SpHelper::allocate2D<NT>(m, n);
for(int i=0; i< m; ++i)
std::copy(array[i], array[i]+n, rhs.array[i]);
}
commGrid.reset(new CommGrid(*(rhs.commGrid)));
}
return *this;
}
}
|
