forked from alibaba/libgrape-lite
-
Notifications
You must be signed in to change notification settings - Fork 0
/
append_only_edgecut_fragment.h
1012 lines (844 loc) Β· 28.6 KB
/
append_only_edgecut_fragment.h
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
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
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
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
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
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
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/** Copyright 2020 Alibaba Group Holding Limited.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
#ifndef EXAMPLES_GNN_SAMPLER_APPEND_ONLY_EDGECUT_FRAGMENT_H_
#define EXAMPLES_GNN_SAMPLER_APPEND_ONLY_EDGECUT_FRAGMENT_H_
#include <algorithm>
#include <map>
#include <memory>
#include <set>
#include <string>
#include <tuple>
#include <utility>
#include <vector>
#include <grape/config.h>
#include <grape/fragment/edgecut_fragment_base.h>
#include <grape/fragment/ev_fragment_loader.h>
#include <grape/fragment/fragment_base.h>
#include <grape/fragment/partitioner.h>
#include <grape/graph/adj_list.h>
#include <grape/graph/edge.h>
#include <grape/graph/vertex.h>
#include <grape/io/io_adaptor_base.h>
#include <grape/io/tsv_line_parser.h>
#include <grape/serialization/in_archive.h>
#include <grape/serialization/out_archive.h>
#include <grape/types.h>
#include <grape/utils/gcontainer.h>
#include <grape/utils/iterator_pair.h>
#include <grape/utils/vertex_array.h>
#include <grape/vertex_map/global_vertex_map.h>
#include <grape/worker/comm_spec.h>
#include "flat_hash_map/flat_hash_map.hpp"
#include "fragment_indices.h"
namespace grape {
/** InsertOnly Edgecut Fragment
*
* It is a kind of strategy to load fragment by a way of insert only.
* Fragments can be extended with extra vertices and edges in this way. All
* data needed for the running of an App are stored in here following its
* strategy.
*/
template <typename VID_T, typename EDATA_T>
class NbrIterImpl {
public:
NbrIterImpl() {}
virtual ~NbrIterImpl() {}
virtual NbrIterImpl* clone() = 0;
virtual void Inc() = 0;
virtual void Add(int right) = 0;
virtual const Nbr<VID_T, EDATA_T>& deref() = 0;
virtual const Nbr<VID_T, EDATA_T>* pointer() = 0;
};
template <typename VID_T, typename EDATA_T>
class NbrVertex {
public:
Nbr<VID_T, EDATA_T> vertex;
NbrVertex* Next() { return next_ ? (this + next_) : nullptr; }
void SetNext(NbrVertex* next) { next_ = next - this; }
private:
int next_;
};
template <typename VID_T, typename EDATA_T>
class NbrIterator {
public:
NbrIterator() : impl_(NULL) {}
explicit NbrIterator(NbrIterImpl<VID_T, EDATA_T>* impl) : impl_(impl) {}
NbrIterator(NbrIterator const& right) : impl_(right.impl_->clone()) {}
~NbrIterator() { delete impl_; }
NbrIterator& operator=(NbrIterator const& right) {
delete impl_;
impl_ = right.impl_->clone();
return *this;
}
NbrIterator& operator+=(int right) {
impl_->Add(right);
return *this;
}
NbrIterator& operator++() {
impl_->Inc();
return *this;
}
const NbrIterator operator++(int) {
NbrIterator it(*this);
impl_->Inc();
return it;
}
Nbr<VID_T, EDATA_T>& operator*() { return impl_->deref(); }
Nbr<VID_T, EDATA_T>* operator->() { return impl_->pointer(); }
Nbr<VID_T, EDATA_T>& nbr() { return impl_->deref(); }
Nbr<VID_T, EDATA_T>* nbr_pointer() { return impl_->pointer(); }
bool equals(const NbrIterator& it) {
return (impl_->pointer() == it.impl_->pointer());
}
bool equals(const NbrIterator* it) { return this->equals(*it); }
bool operator==(const NbrIterator& it) { return this->equals(it); }
bool operator!=(const NbrIterator& it) { return !this->equals(it); }
private:
NbrIterImpl<VID_T, EDATA_T>* impl_;
};
template <typename VID_T, typename EDATA_T>
class NbrSpaceIterImpl : public NbrIterImpl<VID_T, EDATA_T> {
using nbr_t = Nbr<VID_T, EDATA_T>;
using nbr_vertex_t = NbrVertex<VID_T, EDATA_T>;
public:
NbrSpaceIterImpl() : linked_list_ptr_(NULL) {}
explicit NbrSpaceIterImpl(nbr_vertex_t* linked_list_ptr)
: linked_list_ptr_(linked_list_ptr) {}
NbrSpaceIterImpl* clone() { return new NbrSpaceIterImpl(linked_list_ptr_); }
void Inc() {
if (linked_list_ptr_) {
linked_list_ptr_ = linked_list_ptr_->Next();
}
}
void Add(int right) {
for (int i = 0; i < right; ++i) {
Inc();
}
}
nbr_t& deref() {
// may trigger an NPE (segmentation fault) as error
return linked_list_ptr_->vertex;
}
nbr_t* pointer() {
if (linked_list_ptr_) {
return &linked_list_ptr_->vertex;
}
return nullptr; // return null pointer
}
private:
nbr_vertex_t* linked_list_ptr_;
};
template <typename VID_T, typename EDATA_T>
class NbrMapSpaceIterImpl : public NbrIterImpl<VID_T, EDATA_T> {
using nbr_t = Nbr<VID_T, EDATA_T>;
using nbr_vertex_t = NbrVertex<VID_T, EDATA_T>;
public:
NbrMapSpaceIterImpl() {}
explicit NbrMapSpaceIterImpl(
typename std::map<VID_T, Nbr<VID_T, EDATA_T>>::const_iterator map_iter)
: map_iter_(map_iter) {}
inline NbrMapSpaceIterImpl* clone() {
return new NbrMapSpaceIterImpl(map_iter_);
}
inline void Inc() { map_iter_++; }
inline void Add(int right) {
for (int i = 0; i < right; ++i) {
Inc();
}
}
inline const nbr_t& deref() {
// may trigger an NPE (segmentation fault) as error
return map_iter_->second;
}
inline const nbr_t* pointer() {
// may trigger an NPE (segmentation fault) as error
return &map_iter_->second;
}
private:
typename std::map<VID_T, Nbr<VID_T, EDATA_T>>::const_iterator map_iter_;
};
template <typename VID_T, typename EDATA_T>
class NbrMapSpace {
public:
NbrMapSpace() : index_(0) {}
// Create a new linked list
inline size_t emplace(VID_T vid, EDATA_T edata) {
buffer_.resize(index_ + 1);
buffer_[index_] = new std::map<VID_T, Nbr<VID_T, EDATA_T>>();
buffer_[index_]->operator[](vid) = Nbr<VID_T, EDATA_T>(vid, edata);
return index_++;
}
// Insert the value to an existing linked list, or update the existing value
inline size_t emplace(size_t loc, VID_T vid, EDATA_T edata, bool& created) {
if (buffer_[loc]->find(vid) != buffer_[loc]->end()) {
buffer_[loc]->operator[](vid) = Nbr<VID_T, EDATA_T>(vid, edata);
created = false;
return loc;
} else {
buffer_[loc]->operator[](vid) = Nbr<VID_T, EDATA_T>(vid, edata);
created = true;
return loc;
}
}
inline void update(size_t loc, VID_T vid, EDATA_T edata) {
if (buffer_[loc]->find(vid) != buffer_[loc]->end()) {
buffer_[loc]->operator[](vid) = Nbr<VID_T, EDATA_T>(vid, edata);
}
}
inline std::map<VID_T, Nbr<VID_T, EDATA_T>>& operator[](size_t loc) {
return *buffer_[loc];
}
inline const std::map<VID_T, Nbr<VID_T, EDATA_T>>& operator[](
size_t loc) const {
return *buffer_[loc];
}
void Clear() {
for (size_t i = 0; i < buffer_.size(); ++i) {
delete buffer_[i];
buffer_[i] = nullptr;
}
buffer_.clear();
index_ = 0;
}
private:
std::vector<std::map<VID_T, Nbr<VID_T, EDATA_T>>*> buffer_;
size_t index_;
};
template <typename OID_T, typename VID_T, typename VDATA_T, typename EDATA_T,
LoadStrategy _load_strategy = LoadStrategy::kBothOutIn>
class AppendOnlyEdgecutFragment
: public EdgecutFragmentBase<OID_T, VID_T, VDATA_T, EDATA_T> {
public:
using internal_vertex_t = internal::Vertex<VID_T, VDATA_T>;
using edge_t = Edge<VID_T, EDATA_T>;
using nbr_t = Nbr<VID_T, EDATA_T>;
using nbr_vertex_t = NbrVertex<VID_T, EDATA_T>;
using nbr_iterator_t = NbrIterator<VID_T, EDATA_T>;
using vertex_t = Vertex<VID_T>;
using vertex_range_t = VertexRange<VID_T>;
using adj_list_t = AdjList<VID_T, EDATA_T>;
using const_adj_list_t = ConstAdjList<VID_T, EDATA_T>;
using vid_t = VID_T;
using oid_t = OID_T;
using vdata_t = VDATA_T;
using edata_t = EDATA_T;
template <typename DATA_T>
using vertex_array_t = VertexArray<DATA_T, vid_t>;
using nbr_space_iter_impl = NbrSpaceIterImpl<VID_T, EDATA_T>;
using nbr_mapspace_iter_impl = NbrMapSpaceIterImpl<VID_T, EDATA_T>;
using vertex_map_t = GlobalVertexMap<oid_t, vid_t>;
using IsEdgeCut = std::true_type;
using IsVertexCut = std::false_type;
static constexpr LoadStrategy load_strategy = LoadStrategy::kOnlyOut;
AppendOnlyEdgecutFragment() {}
/** Constructor.
* @param vm_ptr the vertex map.
*/
explicit AppendOnlyEdgecutFragment(std::shared_ptr<vertex_map_t> vm_ptr)
: vm_ptr_(vm_ptr) {}
virtual ~AppendOnlyEdgecutFragment() {}
void Init(fid_t fid, std::vector<internal_vertex_t>& vertices,
std::vector<edge_t>& edges) override {
fid_ = fid;
fnum_ = vm_ptr_->GetFragmentNum();
calcFidBitWidth(fnum_, id_mask_, fid_offset_);
ivnum_ = vm_ptr_->GetInnerVertexSize(fid);
ovnum_ = 0;
oenum_ = 0;
oe_.clear();
oeoffset_.clear();
ovg2i_.clear();
vid_t invalid_vid = std::numeric_limits<vid_t>::max();
auto is_iv_gid = [this](vid_t id) { return (id >> fid_offset_) == fid_; };
{
std::vector<vid_t> outer_vertices;
for (auto& e : edges) {
if (is_iv_gid(e.src())) {
if (!is_iv_gid(e.dst())) {
outer_vertices.push_back(e.dst());
}
} else {
e.set_src(invalid_vid);
}
}
DistinctSort(outer_vertices);
ovgid_.resize(outer_vertices.size());
memcpy(&ovgid_[0], &outer_vertices[0],
outer_vertices.size() * sizeof(vid_t));
}
ovnum_ = 0;
for (auto gid : ovgid_) {
ovg2i_.emplace(gid, ovnum_);
++ovnum_;
}
tvnum_ = ivnum_ + ovnum_;
max_old_ilid_ = ivnum_;
min_old_olid_ = id_mask_ - ovnum_;
{
std::vector<int> odegree(ivnum_, 0);
oenum_ = 0;
auto gid_to_lid = [this](vid_t gid) {
return ((gid >> fid_offset_) == fid_) ? (gid & id_mask_)
: (id_mask_ - ovg2i_.at(gid));
};
auto iv_gid_to_lid = [this](vid_t gid) { return gid & id_mask_; };
for (auto& e : edges) {
if (e.src() != invalid_vid) {
e.set_src(iv_gid_to_lid(e.src()));
e.set_dst(gid_to_lid(e.dst()));
++odegree[e.src()];
++oenum_;
}
}
oe_.resize(oenum_);
oeoffset_.resize(ivnum_ + 1);
oeoffset_[0] = &oe_[0];
for (vid_t i = 0; i < ivnum_; ++i) {
oeoffset_[i + 1] = oeoffset_[i] + odegree[i];
}
}
{
Array<nbr_t*, Allocator<nbr_t*>> oeiter(oeoffset_);
for (auto& e : edges) {
if (e.src() != invalid_vid) {
oeiter[e.src()]->GetEdgeDst(e);
++oeiter[e.src()];
}
}
}
for (vid_t i = 0; i < ivnum_; ++i) {
std::sort(oeoffset_[i], oeoffset_[i + 1],
[](const nbr_t& lhs, const nbr_t& rhs) {
return lhs.neighbor.GetValue() < rhs.neighbor.GetValue();
});
}
extra_oenum_ = 0;
extra_oe_.clear();
extra_oe_.resize(ivnum_, -1);
ivdata_.clear();
ivdata_.resize(ivnum_);
if (sizeof(internal_vertex_t) > sizeof(vid_t)) {
for (auto& v : vertices) {
vid_t gid = v.vid();
if (gid >> fid_offset_ == fid_) {
ivdata_[(gid & id_mask_)] = v.vdata();
}
}
}
}
void InitIndices() {
fragment_indices_ =
FragmentIndicesBase<oid_t, vid_t, vdata_t, edata_t>::Create();
fragment_indices_->Init(this);
}
void ExtendFragment(std::vector<std::string>& edge_messages,
const CommSpec& comm_spec, const LoadGraphSpec& spec) {
if (comm_spec.worker_id() == kCoordinatorRank) {
BcastSend(edge_messages, comm_spec.comm());
} else {
BcastRecv(edge_messages, comm_spec.comm(), kCoordinatorRank);
}
if (edge_messages.empty()) {
return;
}
std::vector<edge_t> edges;
edges.reserve(edge_messages.size());
std::vector<oid_t> empty_id_list;
HashPartitioner<oid_t> partitioner(fnum_, empty_id_list);
{
edata_t e_data;
oid_t src, dst, src_gid, dst_gid;
fid_t src_fid, dst_fid;
auto line_parser_ptr =
std::make_shared<TSVLineParser<oid_t, vdata_t, edata_t>>();
for (auto& msg : edge_messages) {
if (msg.empty() || msg[0] == '#') {
continue;
}
try {
line_parser_ptr->LineParserForEFile(msg, src, dst, e_data);
} catch (std::exception& e) {
LOG(ERROR) << e.what();
continue;
}
src_fid = partitioner.GetPartitionId(src);
dst_fid = partitioner.GetPartitionId(dst);
vm_ptr_->AddVertex(src_fid, src, src_gid);
vm_ptr_->AddVertex(dst_fid, dst, dst_gid);
if (src_fid == fid_ || dst_fid == fid_) {
edges.emplace_back(src_gid, dst_gid, e_data);
}
}
}
ivnum_ = vm_ptr_->GetInnerVertexSize(fid_);
extra_oe_.resize(ivnum_, -1);
fragment_indices_->Resize(ivnum_);
{
vid_t src_lid, dst_lid;
vid_t old_ovnum = ovnum_;
std::vector<vid_t> ov_to_extend;
auto is_iv_gid = [this](vid_t id) { return (id >> fid_offset_) == fid_; };
for (auto& e : edges) {
if (is_iv_gid(e.src())) {
src_lid = e.src() & id_mask_;
if (is_iv_gid(e.dst())) {
dst_lid = e.dst() & id_mask_;
if (!spec.directed) {
addOutgoingEdge(dst_lid, src_lid, e.edata());
fragment_indices_->Insert(dst_lid, e.src(), e.edata());
}
} else {
auto iter = ovg2i_.find(e.dst());
if (iter != ovg2i_.end()) {
dst_lid = id_mask_ - iter->second;
} else {
ovg2i_.emplace(e.dst(), ovnum_);
ov_to_extend.emplace_back(e.dst());
dst_lid = id_mask_ - ovnum_;
++ovnum_;
}
}
addOutgoingEdge(src_lid, dst_lid, e.edata());
fragment_indices_->Insert(src_lid, e.dst(), e.edata());
} else if (!spec.directed && is_iv_gid(e.dst())) {
dst_lid = e.dst() & id_mask_;
auto iter = ovg2i_.find(e.src());
if (iter != ovg2i_.end()) {
src_lid = id_mask_ - iter->second;
} else {
ovg2i_.emplace(e.src(), ovnum_);
ov_to_extend.emplace_back(e.src());
src_lid = id_mask_ - ovnum_;
++ovnum_;
}
addOutgoingEdge(dst_lid, src_lid, e.edata());
fragment_indices_->Insert(dst_lid, e.src(), e.edata());
}
}
tvnum_ = ivnum_ + ovnum_;
ovgid_.resize(ovnum_);
memcpy(&ovgid_[old_ovnum], &ov_to_extend[0],
sizeof(vid_t) * ov_to_extend.size());
}
fragment_indices_->Rebuild();
}
template <typename IOADAPTOR_T>
void Serialize(const std::string prefix) {
char fbuf[1024];
snprintf(fbuf, sizeof(fbuf), kSerializationFilenameFormat, prefix.c_str(),
fid_);
VLOG(1) << "Serialize to " << fbuf;
auto io_adaptor =
std::unique_ptr<IOADAPTOR_T>(new IOADAPTOR_T(std::string(fbuf)));
io_adaptor->Open("wb");
InArchive ia;
vid_t xivnum = max_old_ilid_;
vid_t xovnum = id_mask_ - min_old_olid_;
ia << xivnum << xovnum << oenum_ << fid_offset_ << fid_ << id_mask_
<< fnum_;
io_adaptor->WriteArchive(ia);
ia.Clear();
std::map<vid_t, vid_t> xovg2i;
for (auto& pair : ovg2i_) {
if (pair.second < xovnum) {
xovg2i[pair.first] = pair.second;
}
}
ia << xovg2i;
io_adaptor->WriteArchive(ia);
ia.Clear();
io_adaptor->Write(&ovgid_[0], xovnum * sizeof(vid_t));
{
if (std::is_pod<EDATA_T>::value || (sizeof(nbr_t) == sizeof(VID_T))) {
if (oenum_ > 0) {
CHECK(io_adaptor->Write(&oe_[0], oenum_ * sizeof(nbr_t)));
}
} else {
ia << oe_;
CHECK(io_adaptor->WriteArchive(ia));
ia.Clear();
}
std::vector<int> odegree(xivnum);
for (VID_T i = 0; i < xivnum; ++i) {
odegree[i] = oeoffset_[i + 1] - oeoffset_[i];
}
CHECK(io_adaptor->Write(&odegree[0], sizeof(int) * xivnum));
}
io_adaptor->Close();
}
template <typename IOADAPTOR_T>
void Deserialize(const std::string prefix, const fid_t fid) {
char fbuf[1024];
snprintf(fbuf, sizeof(fbuf), kSerializationFilenameFormat, prefix.c_str(),
fid);
VLOG(1) << "Deserialize from " << fbuf;
auto io_adaptor =
std::unique_ptr<IOADAPTOR_T>(new IOADAPTOR_T(std::string(fbuf)));
io_adaptor->Open();
OutArchive oa;
CHECK(io_adaptor->ReadArchive(oa));
oa >> ivnum_ >> ovnum_ >> oenum_ >> fid_offset_ >> fid_ >> id_mask_ >>
fnum_;
tvnum_ = ivnum_ + ovnum_;
oa.Clear();
CHECK(io_adaptor->ReadArchive(oa));
oa >> ovg2i_;
oa.Clear();
ovgid_.clear();
ovgid_.resize(ovnum_);
CHECK(io_adaptor->Read(&ovgid_[0], ovnum_ * sizeof(vid_t)));
oe_.clear();
if (std::is_pod<EDATA_T>::value || (sizeof(nbr_t) == sizeof(vid_t))) {
oe_.resize(oenum_);
if (oenum_ > 0) {
CHECK(io_adaptor->Read(&oe_[0], oenum_ * sizeof(nbr_t)));
}
} else {
CHECK(io_adaptor->ReadArchive(oa));
oa >> oe_;
oa.Clear();
CHECK_EQ(oe_.size(), oenum_);
}
oeoffset_.clear();
oeoffset_.clear();
oeoffset_.resize(ivnum_ + 1);
oeoffset_[0] = &oe_[0];
{
std::vector<int> odegree(tvnum_);
CHECK(io_adaptor->Read(&odegree[0], sizeof(int) * tvnum_));
for (VID_T i = 0; i < ivnum_; ++i) {
oeoffset_[i + 1] = oeoffset_[i] + odegree[i];
}
}
ivdata_.clear();
ivdata_.resize(ivnum_);
io_adaptor->Close();
max_old_ilid_ = ivnum_;
min_old_olid_ = id_mask_ - ovnum_;
extra_oenum_ = 0;
extra_oe_.clear();
extra_oe_.resize(ivnum_, -1);
}
void PrepareToRunApp(MessageStrategy strategy,
bool need_split_edge) override {}
fid_t fid() const override { return fid_; }
fid_t fnum() const override { return fnum_; }
vid_t id_mask() const { return id_mask_; }
int fid_offfset() const { return fid_offset_; }
fid_t GetFragId(const vertex_t& u) const override {
return IsInnerVertex(u)
? fid_
: (fid_t)(ovgid_[id_mask_ - u.GetValue()] >> fid_offset_);
}
fid_t GetFragIdByGid(const vid_t& gid) const {
return (fid_t)(gid >> fid_offset_);
}
size_t GetEdgeNum() const override { return oenum_ + extra_oenum_; }
vid_t GetVerticesNum() const override { return tvnum_; }
size_t GetTotalVerticesNum() const override {
return vm_ptr_->GetTotalVertexSize();
}
bool GetVertex(const oid_t& oid, vertex_t& v) const override {
vid_t gid;
if (vm_ptr_->GetGid(oid, gid)) {
return ((gid >> fid_offset_) == fid_) ? InnerVertexGid2Vertex(gid, v)
: OuterVertexGid2Vertex(gid, v);
} else {
return false;
}
}
oid_t GetId(const vertex_t& v) const override {
return (IsInnerVertex(v)) ? GetInnerVertexId(v) : GetOuterVertexId(v);
}
const vdata_t& GetData(const vertex_t& v) const override {
return ivdata_[v.GetValue()];
}
void SetData(const vertex_t& v, const vdata_t& val) override {
if (IsInnerVertex(v)) {
ivdata_[v.GetValue()] = val;
}
}
bool HasChild(const vertex_t& v) const override {
auto extra_oes = GetExtraOutgoingAdjList(v);
return GetOutgoingAdjList(v).NotEmpty() ||
(extra_oes.begin() != extra_oes.end());
}
bool HasParent(const vertex_t& v) const override { return false; }
int GetLocalOutDegree(const vertex_t& v) const override {
int od = GetOutgoingAdjList(v).Size();
auto es = GetExtraOutgoingAdjList(v);
for (auto e = es.begin(); e != es.end(); ++e) {
od += 1;
}
return od;
}
int GetLocalInDegree(const vertex_t& v) const override { return 0; }
const std::vector<vid_t>& GetOuterVerticesGid() const { return ovgid_; }
bool Gid2Vertex(const vid_t& gid, vertex_t& v) const override {
return ((gid >> fid_offset_) == fid_) ? InnerVertexGid2Vertex(gid, v)
: OuterVertexGid2Vertex(gid, v);
}
vid_t Vertex2Gid(const vertex_t& v) const override {
return IsInnerVertex(v) ? GetInnerVertexGid(v) : GetOuterVertexGid(v);
}
bool Oid2Gid(const oid_t& oid, vid_t& gid) const {
return vm_ptr_->GetGid(oid, gid);
}
bool Oid2Gid(fid_t fid, const oid_t& oid, vid_t& gid) const {
return vm_ptr_->GetGid(fid, oid, gid);
}
oid_t Gid2Oid(const vid_t& gid) const {
OID_T oid;
vm_ptr_->GetOid(gid, oid);
return oid;
}
vid_t GetInnerVerticesNum() const override { return ivnum_; }
vid_t GetOuterVerticesNum() const override { return ovnum_; }
vertex_range_t InnerVertices() const override {
return vertex_range_t(0, ivnum_);
}
vertex_range_t OuterVertices() const override {
return vertex_range_t(id_mask_ - ovnum_ + 1, id_mask_ + 1);
}
vertex_range_t OuterVertices(fid_t fid) const { return vertex_range_t(0, 0); }
vertex_range_t Vertices() const override { return vertex_range_t(0, 0); }
bool IsInnerVertex(const vertex_t& v) const override {
return (v.GetValue() < ivnum_);
}
bool IsInnerVertex(const vid_t& v) const { return (v < ivnum_); }
bool IsOuterVertex(const vertex_t& v) const override {
return (v.GetValue() >= ivnum_);
}
bool GetInnerVertex(const oid_t& oid, vertex_t& v) const override {
vid_t gid;
bool res = vm_ptr_->GetGid(oid, gid);
if (res && ((gid >> fid_offset_) == fid_)) {
return InnerVertexGid2Vertex(gid, v);
} else {
return false;
}
}
bool GetOuterVertex(const oid_t& oid, vertex_t& v) const override {
vid_t gid;
if (vm_ptr_->GetGid(oid, gid)) {
return OuterVertexGid2Vertex(gid, v);
} else {
return false;
}
}
oid_t GetInnerVertexId(const vertex_t& v) const override {
oid_t oid;
vm_ptr_->GetOid(fid_, v.GetValue(), oid);
return oid;
}
oid_t GetOuterVertexId(const vertex_t& v) const override {
vid_t gid = ovgid_[id_mask_ - v.GetValue()];
oid_t oid;
vm_ptr_->GetOid(gid, oid);
return oid;
}
bool InnerVertexGid2Vertex(const vid_t& gid, vertex_t& v) const override {
v.SetValue(gid & id_mask_);
return true;
}
bool OuterVertexGid2Vertex(const vid_t& gid, vertex_t& v) const override {
auto iter = ovg2i_.find(gid);
if (iter != ovg2i_.end()) {
v.SetValue(id_mask_ - iter->second);
return true;
} else {
return false;
}
}
vid_t GetOuterVertexGid(const vertex_t& v) const override {
return ovgid_[id_mask_ - v.GetValue()];
}
vid_t GetInnerVertexGid(const vertex_t& v) const override {
return (v.GetValue() | ((vid_t) fid_ << fid_offset_));
}
// AppendOnlyEdgecutFragment doesn't support along edge message strategy
DestList IEDests(const vertex_t& v) const override {
LOG(FATAL) << "Not implemented.";
return DestList(0, 0);
}
DestList OEDests(const vertex_t& v) const override {
LOG(FATAL) << "Not implemented.";
return DestList(0, 0);
}
DestList IOEDests(const vertex_t& v) const override {
LOG(FATAL) << "Not implemented.";
return DestList(0, 0);
}
adj_list_t GetIncomingAdjList(const vertex_t& v) override {
LOG(FATAL) << "Not implemented.";
return adj_list_t(NULL, NULL);
}
const_adj_list_t GetIncomingAdjList(const vertex_t& v) const override {
LOG(FATAL) << "Not implemented.";
return const_adj_list_t(NULL, NULL);
}
adj_list_t GetOutgoingAdjList(const vertex_t& v) override {
return (v.GetValue() < max_old_ilid_)
? adj_list_t(oeoffset_[v.GetValue()],
oeoffset_[v.GetValue() + 1])
: adj_list_t(NULL, NULL);
}
const_adj_list_t GetOutgoingAdjList(const vertex_t& v) const override {
return (v.GetValue() < max_old_ilid_)
? const_adj_list_t(oeoffset_[v.GetValue()],
oeoffset_[v.GetValue() + 1])
: const_adj_list_t(NULL, NULL);
}
adj_list_t GetIncomingInnerVertexAdjList(const vertex_t& v) override {
LOG(FATAL) << "Not implemented.";
return adj_list_t(NULL, NULL);
}
const_adj_list_t GetIncomingInnerVertexAdjList(
const vertex_t& v) const override {
LOG(FATAL) << "Not implemented.";
return const_adj_list_t(NULL, NULL);
}
adj_list_t GetOutgoingInnerVertexAdjList(const vertex_t& v) override {
return (v.GetValue() < max_old_ilid_)
? adj_list_t(oeoffset_[v.GetValue()],
oeoffset_[v.GetValue() + 1])
: adj_list_t(NULL, NULL);
}
const_adj_list_t GetOutgoingInnerVertexAdjList(
const vertex_t& v) const override {
return (v.GetValue() < max_old_ilid_)
? const_adj_list_t(oeoffset_[v.GetValue()],
oeoffset_[v.GetValue() + 1])
: const_adj_list_t(NULL, NULL);
}
adj_list_t GetIncomingOuterVertexAdjList(const vertex_t& v) override {
LOG(FATAL) << "Not implemented.";
return adj_list_t(NULL, NULL);
}
const_adj_list_t GetIncomingOuterVertexAdjList(
const vertex_t& v) const override {
LOG(FATAL) << "Not implemented.";
return const_adj_list_t(NULL, NULL);
}
adj_list_t GetOutgoingOuterVertexAdjList(const vertex_t& v) override {
return adj_list_t(NULL, NULL);
}
const_adj_list_t GetOutgoingOuterVertexAdjList(
const vertex_t& v) const override {
return const_adj_list_t(NULL, NULL);
}
size_t GetOriginalEdgeNum() const { return oenum_; }
size_t GetExtraEdgeNum() const { return extra_oenum_; }
IteratorPair<nbr_iterator_t> GetExtraOutgoingAdjList(
const vertex_t& v) const {
int loc = extra_oe_[v.GetValue()];
if (loc >= 0) {
return IteratorPair<nbr_iterator_t>(
nbr_iterator_t(
new nbr_mapspace_iter_impl(extra_edge_space_[loc].begin())),
nbr_iterator_t(
new nbr_mapspace_iter_impl(extra_edge_space_[loc].end())));
} else {
return IteratorPair<nbr_iterator_t>(
nbr_iterator_t(new nbr_space_iter_impl(nullptr)),
nbr_iterator_t(new nbr_space_iter_impl(nullptr)));
}
}
IteratorPair<nbr_iterator_t> GetExtraIncomingAdjList(
const vertex_t& v) const {
LOG(FATAL) << "Not implemented.";
return IteratorPair<nbr_iterator_t>(
nbr_iterator_t(new nbr_space_iter_impl(nullptr)),
nbr_iterator_t(new nbr_space_iter_impl(nullptr)));
}
const std::unique_ptr<FragmentIndicesBase<oid_t, vid_t, vdata_t, edata_t>>&
GetFragmentIndices() const {
return fragment_indices_;
}
void SetupMirrorInfo(fid_t fid, const VertexRange<vid_t>& range,
const std::vector<vid_t>& gid_list) {}
private:
vid_t lid2Index(const vid_t& lid) const {
return (lid < ivnum_) ? lid : (id_mask_ - lid + ivnum_);
}
bool binarySearchAndUpdate(adj_list_t& edge_list, vid_t lid,
const edata_t& edata) {
if (edge_list.NotEmpty()) {
nbr_t* l = edge_list.begin_pointer();
nbr_t* r = edge_list.end_pointer() - 1;
while (l <= r) {
nbr_t* m = l + (r - l) / 2;
if (m->neighbor.GetValue() == lid) {
m->data = edata;
return true;
} else if (m->neighbor.GetValue() < lid) {
l = m + 1;
} else {
r = m - 1;
}
}
}
return false;
}
bool addOutgoingEdge(vid_t u_lid, vid_t v_lid, const edata_t& edata) {
auto oes = GetOutgoingAdjList(vertex_t(u_lid));
if (binarySearchAndUpdate(oes, v_lid, edata)) {
return false;
}
int loc = extra_oe_[u_lid];
if (loc == -1) {
++extra_oenum_;
extra_oe_[u_lid] = extra_edge_space_.emplace(v_lid, edata);
return true;
} else {
bool created = false;
extra_oe_[u_lid] = extra_edge_space_.emplace(loc, v_lid, edata, created);
if (created) {
++extra_oenum_;
}
return created;
}
}
template <typename T>
void calcFidBitWidth(fid_t fnum, T& id_mask, int& fid_offset) {
fid_t maxfid = fnum - 1;
if (maxfid == 0) {
fid_offset = (sizeof(T) * 8) - 1;
} else {
int i = 0;
while (maxfid) {
maxfid >>= 1;
++i;
}
fid_offset = (sizeof(T) * 8) - i;
}
id_mask = ((T) 1 << fid_offset) - (T) 1;
}
std::shared_ptr<vertex_map_t> vm_ptr_;
vid_t ivnum_, ovnum_, tvnum_, id_mask_;
vid_t max_old_ilid_, min_old_olid_;
size_t oenum_;
size_t extra_oenum_;
int fid_offset_;
fid_t fid_, fnum_;
ska::flat_hash_map<vid_t, vid_t> ovg2i_;
std::vector<vid_t> ovgid_;