distributed_do.py 11.6 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
import numpy as np
from .random import Random
from mpi4py import MPI

comm = MPI.COMM_WORLD
ntask = comm.Get_size()
rank = comm.Get_rank()


def shareSize(nwork, nshares, myshare):
    nbase = nwork//nshares
    return nbase if myshare>=nwork%nshares else nbase+1
Martin Reinecke's avatar
Martin Reinecke committed
13
14
15
16
17
18
def shareRange(nwork, nshares, myshare):
    nbase = nwork//nshares;
    additional = nwork%nshares;
    lo = myshare*nbase + min(myshare, additional)
    hi = lo+nbase+ (1 if myshare<additional else 0)
    return lo,hi
19
20

def get_locshape(shape, distaxis):
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
21
22
    if len(shape)==0:
        distaxis = -1
23
24
25
26
27
    if distaxis==-1:
        return shape
    shape2=list(shape)
    shape2[distaxis]=shareSize(shape[distaxis],ntask,rank)
    return tuple(shape2)
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
28
29
def local_shape(shape, distaxis):
    return get_locshape(shape,distaxis)
30
31
32
33

class data_object(object):
    def __init__(self, shape, data, distaxis):
        """Must not be called directly by users"""
Martin Reinecke's avatar
Martin Reinecke committed
34
        self._shape = tuple(shape)
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
35
36
        if len(self._shape)==0:
            distaxis = -1
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
        self._distaxis = distaxis
        lshape = get_locshape(self._shape, self._distaxis)
        self._data = data

    def sanity_checks(self):
        # check whether the distaxis is consistent
        if self._distaxis<-1 or self._distaxis>=len(self._shape):
            raise ValueError
        itmp=np.array(self._distaxis)
        otmp=np.empty(ntask,dtype=np.int)
        comm.Allgather(itmp,otmp)
        if np.any(otmp!=self._distaxis):
            raise ValueError
        # check whether the global shape is consistent
        itmp=np.array(self._shape)
Martin Reinecke's avatar
Martin Reinecke committed
52
        otmp=np.empty((ntask,len(self._shape)),dtype=np.int)
53
54
        comm.Allgather(itmp,otmp)
        for i in range(ntask):
Martin Reinecke's avatar
Martin Reinecke committed
55
            if np.any(otmp[i,:]!=self._shape):
56
57
58
59
60
61
62
                raise ValueError
        # check shape of local data
        if self._distaxis<0:
            if self._data.shape!=self._shape:
                raise ValueError
        else:
            itmp=np.array(self._shape)
Martin Reinecke's avatar
Martin Reinecke committed
63
64
            itmp[self._distaxis] = shareSize(self._shape[self._distaxis],ntask,rank)
            if np.any(self._data.shape!=itmp):
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
                raise ValueError

    @property
    def dtype(self):
        return self._data.dtype

    @property
    def shape(self):
        return self._shape

    @property
    def size(self):
        return np.prod(self._shape)

    @property
    def real(self):
Martin Reinecke's avatar
Martin Reinecke committed
81
        return data_object(self._shape, self._data.real, self._distaxis)
82
83
84

    @property
    def imag(self):
Martin Reinecke's avatar
Martin Reinecke committed
85
        return data_object(self._shape, self._data.imag, self._distaxis)
86

Martin Reinecke's avatar
Martin Reinecke committed
87
    def _contraction_helper(self, op, mpiop, axis):
88
89
90
91
        if axis is not None:
            if len(axis)==len(self._data.shape):
                axis = None
        if axis is None:
Martin Reinecke's avatar
Martin Reinecke committed
92
93
            res = np.array(getattr(self._data, op)())
            if (self._distaxis==-1):
Martin Reinecke's avatar
Martin Reinecke committed
94
                return res[0]
Martin Reinecke's avatar
Martin Reinecke committed
95
            res2 = np.empty(1,dtype=res.dtype)
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
96
            comm.Allreduce(res,res2,mpiop)
Martin Reinecke's avatar
Martin Reinecke committed
97
            return res2[0]
98
99

        if self._distaxis in axis:
Martin Reinecke's avatar
Martin Reinecke committed
100
101
102
103
            res = getattr(self._data, op)(axis=axis)
            res2 = np.empty_like(res)
            comm.Allreduce(res,res2,mpiop)
            return from_global_data(res2, distaxis=0)
104
        else:
Martin Reinecke's avatar
Martin Reinecke committed
105
            # perform the contraction on the local data
Martin Reinecke's avatar
Martin Reinecke committed
106
107
108
109
110
111
112
113
114
115
            res = getattr(self._data, op)(axis=axis)
            if self._distaxis == -1:
                return from_global_data(res,distaxis=0)
            shp = list(res.shape)
            shift=0
            for ax in axis:
                if ax<self._distaxis:
                    shift+=1
            shp[self._distaxis-shift] = self.shape[self._distaxis]
            return from_local_data(shp, res, self._distaxis-shift)
116
117
118
119
120
121
122
123
124
125
126

        # check if the result is scalar or if a result_field must be constr.
        if np.isscalar(data):
            return data
        else:
            return data_object(data)

    def sum(self, axis=None):
        return self._contraction_helper("sum", MPI.SUM, axis)

    def _binary_helper(self, other, op):
Martin Reinecke's avatar
Martin Reinecke committed
127
        a = self
128
        if isinstance(other, data_object):
Martin Reinecke's avatar
Martin Reinecke committed
129
            b = other
130
131
132
133
            if a._shape != b._shape:
                raise ValueError("shapes are incompatible.")
            if a._distaxis != b._distaxis:
                raise ValueError("distributions are incompatible.")
Martin Reinecke's avatar
Martin Reinecke committed
134
135
            a = a._data
            b = b._data
136
        else:
Martin Reinecke's avatar
Martin Reinecke committed
137
            a = a._data
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
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
            b = other

        tval = getattr(a, op)(b)
        return self if tval is a else data_object(self._shape, tval, self._distaxis)

    def __add__(self, other):
        return self._binary_helper(other, op='__add__')

    def __radd__(self, other):
        return self._binary_helper(other, op='__radd__')

    def __iadd__(self, other):
        return self._binary_helper(other, op='__iadd__')

    def __sub__(self, other):
        return self._binary_helper(other, op='__sub__')

    def __rsub__(self, other):
        return self._binary_helper(other, op='__rsub__')

    def __isub__(self, other):
        return self._binary_helper(other, op='__isub__')

    def __mul__(self, other):
        return self._binary_helper(other, op='__mul__')

    def __rmul__(self, other):
        return self._binary_helper(other, op='__rmul__')

    def __imul__(self, other):
        return self._binary_helper(other, op='__imul__')

    def __div__(self, other):
        return self._binary_helper(other, op='__div__')

    def __rdiv__(self, other):
        return self._binary_helper(other, op='__rdiv__')

    def __truediv__(self, other):
        return self._binary_helper(other, op='__truediv__')

    def __rtruediv__(self, other):
        return self._binary_helper(other, op='__rtruediv__')

    def __pow__(self, other):
        return self._binary_helper(other, op='__pow__')

    def __rpow__(self, other):
        return self._binary_helper(other, op='__rpow__')

    def __ipow__(self, other):
        return self._binary_helper(other, op='__ipow__')

    def __eq__(self, other):
        return self._binary_helper(other, op='__eq__')

    def __ne__(self, other):
        return self._binary_helper(other, op='__ne__')

    def __neg__(self):
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
198
        return data_object(self._shape,-self._data,self._distaxis)
199
200

    def __abs__(self):
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
201
        return data_object(self._shape,np.abs(self._data),self._distaxis)
202

Martin Reinecke's avatar
fixes    
Martin Reinecke committed
203
204
    #def ravel(self):
    #    return data_object(self._data.ravel())
205

Martin Reinecke's avatar
fixes    
Martin Reinecke committed
206
207
    #def reshape(self, shape):
    #    return data_object(self._data.reshape(shape))
208
209
210
211
212
213
214
215

    def all(self):
        return self._data.all()

    def any(self):
        return self._data.any()


Martin Reinecke's avatar
Martin Reinecke committed
216
def full(shape, fill_value, dtype=None, distaxis=0):
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
217
    return data_object(shape, np.full(get_locshape(shape, distaxis), fill_value, dtype), distaxis)
218
219


Martin Reinecke's avatar
fixes    
Martin Reinecke committed
220
221
def empty(shape, dtype=None, distaxis=0):
    return data_object(shape, np.empty(get_locshape(shape, distaxis), dtype), distaxis)
222
223


Martin Reinecke's avatar
fixes    
Martin Reinecke committed
224
225
def zeros(shape, dtype=None, distaxis=0):
    return data_object(shape, np.zeros(get_locshape(shape, distaxis), dtype), distaxis)
226
227


Martin Reinecke's avatar
fixes    
Martin Reinecke committed
228
229
def ones(shape, dtype=None, distaxis=0):
    return data_object(shape, np.ones(get_locshape(shape, distaxis), dtype), distaxis)
230
231
232
233
234
235
236


def empty_like(a, dtype=None):
    return data_object(np.empty_like(a._data, dtype))


def vdot(a, b):
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
237
238
239
    tmp = np.vdot(a._data.ravel(), b._data.ravel())
    res = np.empty(1,dtype=type(tmp))
    comm.Allreduce(tmp,res,MPI.SUM)
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
240
    return res[0]
241
242
243
244
245
246
247


def _math_helper(x, function, out):
    if out is not None:
        function(x._data, out=out._data)
        return out
    else:
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
248
        return data_object(x.shape,function(x._data),x._distaxis)
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274


def abs(a, out=None):
    return _math_helper(a, np.abs, out)


def exp(a, out=None):
    return _math_helper(a, np.exp, out)


def log(a, out=None):
    return _math_helper(a, np.log, out)


def sqrt(a, out=None):
    return _math_helper(a, np.sqrt, out)


def bincount(x, weights=None, minlength=None):
    if weights is not None:
        weights = weights._data
    res = np.bincount(x._data, weights, minlength)
    return data_object(res)


def from_object(object, dtype=None, copy=True):
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
275
    return data_object(object._shape, np.array(object._data, dtype=dtype, copy=copy), distaxis=object._distaxis)
276
277


Martin Reinecke's avatar
fixes    
Martin Reinecke committed
278
def from_random(random_type, shape, dtype=np.float64, distaxis=0, **kwargs):
279
    generator_function = getattr(Random, random_type)
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
280
281
    lshape = get_locshape(shape, distaxis)
    return data_object(shape, generator_function(dtype=dtype, shape=lshape, **kwargs), distaxis=distaxis)
282
283


Martin Reinecke's avatar
Martin Reinecke committed
284
285
286
287
def local_data(arr):
    return arr._data


Martin Reinecke's avatar
fixes    
Martin Reinecke committed
288
289
290
291
def ibegin(arr):
    res = [0] * arr._data.ndim
    res[arr._distaxis] = shareRange(arr._shape[arr._distaxis],ntask,rank)[0]
    return tuple(res)
Martin Reinecke's avatar
Martin Reinecke committed
292
293


Martin Reinecke's avatar
fixes    
Martin Reinecke committed
294
295
296
297
def np_allreduce_sum(arr):
    res = np.empty_like(arr)
    comm.Allreduce(arr,res,MPI.SUM)
    return res
Martin Reinecke's avatar
Martin Reinecke committed
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316


def distaxis(arr):
    return arr._distaxis


def from_local_data (shape, arr, distaxis):
    return data_object(shape, arr, distaxis)


def from_global_data (arr, distaxis=0):
    if distaxis==-1:
        return data_object(arr.shape, arr, distaxis)
    lo, hi = shareRange(arr.shape[distaxis],ntask,rank)
    sl = [slice(None)]*len(arr.shape)
    sl[distaxis]=slice(lo,hi)
    return data_object(arr.shape, arr[sl], distaxis)


Martin Reinecke's avatar
fixes    
Martin Reinecke committed
317
318
319
320
321
322
323
def to_global_data (arr):
    if arr._distaxis==-1:
        return arr._data
    tmp = redistribute(arr, dist=-1)
    return tmp._data


Martin Reinecke's avatar
Martin Reinecke committed
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
def redistribute (arr, dist=None, nodist=None):
    if dist is not None:
        if nodist is not None:
            raise ValueError
        if dist==arr._distaxis:
            return arr
    else:
        if nodist is None:
            raise ValueError
        if arr._distaxis not in nodist:
            return arr
        dist=-1
        for i in range(len(arr.shape)):
            if i not in nodist:
                dist=i
                break
Martin Reinecke's avatar
Martin Reinecke committed
340

Martin Reinecke's avatar
Martin Reinecke committed
341
342
343
344
345
346
347
348
349
350
351
352
353
354
    if arr._distaxis==-1:  # just pick the proper subset
        return from_global_data(arr._data, dist)
    if dist==-1: # gather data
        tmp = np.moveaxis(arr._data, arr._distaxis, 0)
        slabsize=np.prod(tmp.shape[1:])*tmp.itemsize
        sz=np.empty(ntask,dtype=np.int)
        for i in range(ntask):
            sz[i]=slabsize*shareSize(arr.shape[arr._distaxis],ntask,i)
        disp=np.empty(ntask,dtype=np.int)
        disp[0]=0
        disp[1:]=np.cumsum(sz[:-1])
        tmp=tmp.flatten()
        out = np.empty(arr.size,dtype=arr.dtype)
        comm.Allgatherv(tmp,[out,sz,disp,MPI.BYTE])
Martin Reinecke's avatar
Martin Reinecke committed
355
356
357
358
        shp = np.array(arr._shape)
        shp[1:arr._distaxis+1] = shp[0:arr._distaxis]
        shp[0] = arr.shape[arr._distaxis]
        out = out.reshape(shp)
Martin Reinecke's avatar
Martin Reinecke committed
359
360
361
        out = np.moveaxis(out, 0, arr._distaxis)
        return from_global_data (out, distaxis=-1)
    # real redistribution via Alltoallv
Martin Reinecke's avatar
Martin Reinecke committed
362
363
364
    # temporary slow, but simple solution
    return redistribute(redistribute(arr,dist=-1),dist=dist)

Martin Reinecke's avatar
Martin Reinecke committed
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
    tmp = np.moveaxis(arr._data, (dist, arr._distaxis), (0, 1))
    tshape = tmp.shape
    slabsize=np.prod(tmp.shape[2:])*tmp.itemsize
    ssz=np.empty(ntask,dtype=np.int)
    rsz=np.empty(ntask,dtype=np.int)
    for i in range(ntask):
        ssz[i]=slabsize*tmp.shape[1]*shareSize(arr.shape[dist],ntask,i)
        rsz[i]=slabsize*shareSize(arr.shape[dist],ntask,rank)*shareSize(arr.shape[arr._distaxis],ntask,i)
    sdisp=np.empty(ntask,dtype=np.int)
    rdisp=np.empty(ntask,dtype=np.int)
    sdisp[0]=0
    rdisp[0]=0
    sdisp[1:]=np.cumsum(ssz[:-1])
    rdisp[1:]=np.cumsum(rsz[:-1])
    tmp=tmp.flatten()
    out = np.empty(np.prod(get_locshape(arr.shape,dist)),dtype=arr.dtype)
    s_msg = [tmp, (ssz, sdisp), MPI.BYTE]
    r_msg = [out, (rsz, rdisp), MPI.BYTE]
    comm.Alltoallv(s_msg, r_msg)
    new_shape = [shareSize(arr.shape[dist],ntask,rank), arr.shape[arr._distaxis]] +list(tshape[2:])
    out=out.reshape(new_shape)
    out = np.moveaxis(out, (0, 1), (dist, arr._distaxis))
    return from_local_data (arr.shape, out, dist)


def default_distaxis():
    return 0