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
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
|
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You 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.
*/
package org.apache.spark.storage
import java.io._
import java.nio.{ByteBuffer, MappedByteBuffer}
import scala.collection.mutable.{ArrayBuffer, HashMap}
import scala.concurrent.{Await, ExecutionContext, Future}
import scala.concurrent.duration._
import scala.util.Random
import scala.util.control.NonFatal
import sun.nio.ch.DirectBuffer
import org.apache.spark._
import org.apache.spark.executor.{DataReadMethod, ShuffleWriteMetrics}
import org.apache.spark.io.CompressionCodec
import org.apache.spark.memory.MemoryManager
import org.apache.spark.network._
import org.apache.spark.network.buffer.{ManagedBuffer, NioManagedBuffer}
import org.apache.spark.network.netty.SparkTransportConf
import org.apache.spark.network.shuffle.ExternalShuffleClient
import org.apache.spark.network.shuffle.protocol.ExecutorShuffleInfo
import org.apache.spark.rpc.RpcEnv
import org.apache.spark.serializer.{Serializer, SerializerInstance}
import org.apache.spark.shuffle.ShuffleManager
import org.apache.spark.storage.memory._
import org.apache.spark.util._
/* Class for returning a fetched block and associated metrics. */
private[spark] class BlockResult(
val data: Iterator[Any],
val readMethod: DataReadMethod.Value,
val bytes: Long)
/**
* Manager running on every node (driver and executors) which provides interfaces for putting and
* retrieving blocks both locally and remotely into various stores (memory, disk, and off-heap).
*
* Note that [[initialize()]] must be called before the BlockManager is usable.
*/
private[spark] class BlockManager(
executorId: String,
rpcEnv: RpcEnv,
val master: BlockManagerMaster,
defaultSerializer: Serializer,
val conf: SparkConf,
memoryManager: MemoryManager,
mapOutputTracker: MapOutputTracker,
shuffleManager: ShuffleManager,
blockTransferService: BlockTransferService,
securityManager: SecurityManager,
numUsableCores: Int)
extends BlockDataManager with Logging {
private[spark] val externalShuffleServiceEnabled =
conf.getBoolean("spark.shuffle.service.enabled", false)
val diskBlockManager = {
// Only perform cleanup if an external service is not serving our shuffle files.
val deleteFilesOnStop =
!externalShuffleServiceEnabled || executorId == SparkContext.DRIVER_IDENTIFIER
new DiskBlockManager(conf, deleteFilesOnStop)
}
private[storage] val blockInfoManager = new BlockInfoManager
private val futureExecutionContext = ExecutionContext.fromExecutorService(
ThreadUtils.newDaemonCachedThreadPool("block-manager-future", 128))
// Actual storage of where blocks are kept
private[spark] val memoryStore = new MemoryStore(conf, this, memoryManager)
private[spark] val diskStore = new DiskStore(conf, diskBlockManager)
memoryManager.setMemoryStore(memoryStore)
// Note: depending on the memory manager, `maxStorageMemory` may actually vary over time.
// However, since we use this only for reporting and logging, what we actually want here is
// the absolute maximum value that `maxStorageMemory` can ever possibly reach. We may need
// to revisit whether reporting this value as the "max" is intuitive to the user.
private val maxMemory = memoryManager.maxStorageMemory
// Port used by the external shuffle service. In Yarn mode, this may be already be
// set through the Hadoop configuration as the server is launched in the Yarn NM.
private val externalShuffleServicePort = {
val tmpPort = Utils.getSparkOrYarnConfig(conf, "spark.shuffle.service.port", "7337").toInt
if (tmpPort == 0) {
// for testing, we set "spark.shuffle.service.port" to 0 in the yarn config, so yarn finds
// an open port. But we still need to tell our spark apps the right port to use. So
// only if the yarn config has the port set to 0, we prefer the value in the spark config
conf.get("spark.shuffle.service.port").toInt
} else {
tmpPort
}
}
var blockManagerId: BlockManagerId = _
// Address of the server that serves this executor's shuffle files. This is either an external
// service, or just our own Executor's BlockManager.
private[spark] var shuffleServerId: BlockManagerId = _
// Client to read other executors' shuffle files. This is either an external service, or just the
// standard BlockTransferService to directly connect to other Executors.
private[spark] val shuffleClient = if (externalShuffleServiceEnabled) {
val transConf = SparkTransportConf.fromSparkConf(conf, "shuffle", numUsableCores)
new ExternalShuffleClient(transConf, securityManager, securityManager.isAuthenticationEnabled(),
securityManager.isSaslEncryptionEnabled())
} else {
blockTransferService
}
// Whether to compress broadcast variables that are stored
private val compressBroadcast = conf.getBoolean("spark.broadcast.compress", true)
// Whether to compress shuffle output that are stored
private val compressShuffle = conf.getBoolean("spark.shuffle.compress", true)
// Whether to compress RDD partitions that are stored serialized
private val compressRdds = conf.getBoolean("spark.rdd.compress", false)
// Whether to compress shuffle output temporarily spilled to disk
private val compressShuffleSpill = conf.getBoolean("spark.shuffle.spill.compress", true)
// Max number of failures before this block manager refreshes the block locations from the driver
private val maxFailuresBeforeLocationRefresh =
conf.getInt("spark.block.failures.beforeLocationRefresh", 5)
private val slaveEndpoint = rpcEnv.setupEndpoint(
"BlockManagerEndpoint" + BlockManager.ID_GENERATOR.next,
new BlockManagerSlaveEndpoint(rpcEnv, this, mapOutputTracker))
// Pending re-registration action being executed asynchronously or null if none is pending.
// Accesses should synchronize on asyncReregisterLock.
private var asyncReregisterTask: Future[Unit] = null
private val asyncReregisterLock = new Object
// Field related to peer block managers that are necessary for block replication
@volatile private var cachedPeers: Seq[BlockManagerId] = _
private val peerFetchLock = new Object
private var lastPeerFetchTime = 0L
/* The compression codec to use. Note that the "lazy" val is necessary because we want to delay
* the initialization of the compression codec until it is first used. The reason is that a Spark
* program could be using a user-defined codec in a third party jar, which is loaded in
* Executor.updateDependencies. When the BlockManager is initialized, user level jars hasn't been
* loaded yet. */
private lazy val compressionCodec: CompressionCodec = CompressionCodec.createCodec(conf)
/**
* Initializes the BlockManager with the given appId. This is not performed in the constructor as
* the appId may not be known at BlockManager instantiation time (in particular for the driver,
* where it is only learned after registration with the TaskScheduler).
*
* This method initializes the BlockTransferService and ShuffleClient, registers with the
* BlockManagerMaster, starts the BlockManagerWorker endpoint, and registers with a local shuffle
* service if configured.
*/
def initialize(appId: String): Unit = {
blockTransferService.init(this)
shuffleClient.init(appId)
blockManagerId = BlockManagerId(
executorId, blockTransferService.hostName, blockTransferService.port)
shuffleServerId = if (externalShuffleServiceEnabled) {
logInfo(s"external shuffle service port = $externalShuffleServicePort")
BlockManagerId(executorId, blockTransferService.hostName, externalShuffleServicePort)
} else {
blockManagerId
}
master.registerBlockManager(blockManagerId, maxMemory, slaveEndpoint)
// Register Executors' configuration with the local shuffle service, if one should exist.
if (externalShuffleServiceEnabled && !blockManagerId.isDriver) {
registerWithExternalShuffleServer()
}
}
private def registerWithExternalShuffleServer() {
logInfo("Registering executor with local external shuffle service.")
val shuffleConfig = new ExecutorShuffleInfo(
diskBlockManager.localDirs.map(_.toString),
diskBlockManager.subDirsPerLocalDir,
shuffleManager.shortName)
val MAX_ATTEMPTS = 3
val SLEEP_TIME_SECS = 5
for (i <- 1 to MAX_ATTEMPTS) {
try {
// Synchronous and will throw an exception if we cannot connect.
shuffleClient.asInstanceOf[ExternalShuffleClient].registerWithShuffleServer(
shuffleServerId.host, shuffleServerId.port, shuffleServerId.executorId, shuffleConfig)
return
} catch {
case e: Exception if i < MAX_ATTEMPTS =>
logError(s"Failed to connect to external shuffle server, will retry ${MAX_ATTEMPTS - i}"
+ s" more times after waiting $SLEEP_TIME_SECS seconds...", e)
Thread.sleep(SLEEP_TIME_SECS * 1000)
}
}
}
/**
* Report all blocks to the BlockManager again. This may be necessary if we are dropped
* by the BlockManager and come back or if we become capable of recovering blocks on disk after
* an executor crash.
*
* This function deliberately fails silently if the master returns false (indicating that
* the slave needs to re-register). The error condition will be detected again by the next
* heart beat attempt or new block registration and another try to re-register all blocks
* will be made then.
*/
private def reportAllBlocks(): Unit = {
logInfo(s"Reporting ${blockInfoManager.size} blocks to the master.")
for ((blockId, info) <- blockInfoManager.entries) {
val status = getCurrentBlockStatus(blockId, info)
if (!tryToReportBlockStatus(blockId, info, status)) {
logError(s"Failed to report $blockId to master; giving up.")
return
}
}
}
/**
* Re-register with the master and report all blocks to it. This will be called by the heart beat
* thread if our heartbeat to the block manager indicates that we were not registered.
*
* Note that this method must be called without any BlockInfo locks held.
*/
def reregister(): Unit = {
// TODO: We might need to rate limit re-registering.
logInfo("BlockManager re-registering with master")
master.registerBlockManager(blockManagerId, maxMemory, slaveEndpoint)
reportAllBlocks()
}
/**
* Re-register with the master sometime soon.
*/
private def asyncReregister(): Unit = {
asyncReregisterLock.synchronized {
if (asyncReregisterTask == null) {
asyncReregisterTask = Future[Unit] {
// This is a blocking action and should run in futureExecutionContext which is a cached
// thread pool
reregister()
asyncReregisterLock.synchronized {
asyncReregisterTask = null
}
}(futureExecutionContext)
}
}
}
/**
* For testing. Wait for any pending asynchronous re-registration; otherwise, do nothing.
*/
def waitForAsyncReregister(): Unit = {
val task = asyncReregisterTask
if (task != null) {
Await.ready(task, Duration.Inf)
}
}
/**
* Interface to get local block data. Throws an exception if the block cannot be found or
* cannot be read successfully.
*/
override def getBlockData(blockId: BlockId): ManagedBuffer = {
if (blockId.isShuffle) {
shuffleManager.shuffleBlockResolver.getBlockData(blockId.asInstanceOf[ShuffleBlockId])
} else {
getLocalBytes(blockId) match {
case Some(buffer) => new BlockManagerManagedBuffer(this, blockId, buffer)
case None => throw new BlockNotFoundException(blockId.toString)
}
}
}
/**
* Put the block locally, using the given storage level.
*/
override def putBlockData(blockId: BlockId, data: ManagedBuffer, level: StorageLevel): Boolean = {
putBytes(blockId, data.nioByteBuffer(), level)
}
/**
* Get the BlockStatus for the block identified by the given ID, if it exists.
* NOTE: This is mainly for testing, and it doesn't fetch information from external block store.
*/
def getStatus(blockId: BlockId): Option[BlockStatus] = {
blockInfoManager.get(blockId).map { info =>
val memSize = if (memoryStore.contains(blockId)) memoryStore.getSize(blockId) else 0L
val diskSize = if (diskStore.contains(blockId)) diskStore.getSize(blockId) else 0L
BlockStatus(info.level, memSize = memSize, diskSize = diskSize)
}
}
/**
* Get the ids of existing blocks that match the given filter. Note that this will
* query the blocks stored in the disk block manager (that the block manager
* may not know of).
*/
def getMatchingBlockIds(filter: BlockId => Boolean): Seq[BlockId] = {
// The `toArray` is necessary here in order to force the list to be materialized so that we
// don't try to serialize a lazy iterator when responding to client requests.
(blockInfoManager.entries.map(_._1) ++ diskBlockManager.getAllBlocks())
.filter(filter)
.toArray
.toSeq
}
/**
* Tell the master about the current storage status of a block. This will send a block update
* message reflecting the current status, *not* the desired storage level in its block info.
* For example, a block with MEMORY_AND_DISK set might have fallen out to be only on disk.
*
* droppedMemorySize exists to account for when the block is dropped from memory to disk (so
* it is still valid). This ensures that update in master will compensate for the increase in
* memory on slave.
*/
private def reportBlockStatus(
blockId: BlockId,
info: BlockInfo,
status: BlockStatus,
droppedMemorySize: Long = 0L): Unit = {
val needReregister = !tryToReportBlockStatus(blockId, info, status, droppedMemorySize)
if (needReregister) {
logInfo(s"Got told to re-register updating block $blockId")
// Re-registering will report our new block for free.
asyncReregister()
}
logDebug(s"Told master about block $blockId")
}
/**
* Actually send a UpdateBlockInfo message. Returns the master's response,
* which will be true if the block was successfully recorded and false if
* the slave needs to re-register.
*/
private def tryToReportBlockStatus(
blockId: BlockId,
info: BlockInfo,
status: BlockStatus,
droppedMemorySize: Long = 0L): Boolean = {
if (info.tellMaster) {
val storageLevel = status.storageLevel
val inMemSize = Math.max(status.memSize, droppedMemorySize)
val onDiskSize = status.diskSize
master.updateBlockInfo(blockManagerId, blockId, storageLevel, inMemSize, onDiskSize)
} else {
true
}
}
/**
* Return the updated storage status of the block with the given ID. More specifically, if
* the block is dropped from memory and possibly added to disk, return the new storage level
* and the updated in-memory and on-disk sizes.
*/
private def getCurrentBlockStatus(blockId: BlockId, info: BlockInfo): BlockStatus = {
info.synchronized {
info.level match {
case null =>
BlockStatus(StorageLevel.NONE, memSize = 0L, diskSize = 0L)
case level =>
val inMem = level.useMemory && memoryStore.contains(blockId)
val onDisk = level.useDisk && diskStore.contains(blockId)
val deserialized = if (inMem) level.deserialized else false
val replication = if (inMem || onDisk) level.replication else 1
val storageLevel =
StorageLevel(onDisk, inMem, deserialized, replication)
val memSize = if (inMem) memoryStore.getSize(blockId) else 0L
val diskSize = if (onDisk) diskStore.getSize(blockId) else 0L
BlockStatus(storageLevel, memSize, diskSize)
}
}
}
/**
* Get locations of an array of blocks.
*/
private def getLocationBlockIds(blockIds: Array[BlockId]): Array[Seq[BlockManagerId]] = {
val startTimeMs = System.currentTimeMillis
val locations = master.getLocations(blockIds).toArray
logDebug("Got multiple block location in %s".format(Utils.getUsedTimeMs(startTimeMs)))
locations
}
/**
* Get block from local block manager as an iterator of Java objects.
*/
def getLocalValues(blockId: BlockId): Option[BlockResult] = {
logDebug(s"Getting local block $blockId")
blockInfoManager.lockForReading(blockId) match {
case None =>
logDebug(s"Block $blockId was not found")
None
case Some(info) =>
val level = info.level
logDebug(s"Level for block $blockId is $level")
if (level.useMemory && memoryStore.contains(blockId)) {
val iter: Iterator[Any] = if (level.deserialized) {
memoryStore.getValues(blockId).get
} else {
dataDeserialize(blockId, memoryStore.getBytes(blockId).get)
}
val ci = CompletionIterator[Any, Iterator[Any]](iter, releaseLock(blockId))
Some(new BlockResult(ci, DataReadMethod.Memory, info.size))
} else if (level.useDisk && diskStore.contains(blockId)) {
val iterToReturn: Iterator[Any] = {
val diskBytes = diskStore.getBytes(blockId)
if (level.deserialized) {
val diskValues = dataDeserialize(blockId, diskBytes)
maybeCacheDiskValuesInMemory(info, blockId, level, diskValues)
} else {
dataDeserialize(blockId, maybeCacheDiskBytesInMemory(info, blockId, level, diskBytes))
}
}
val ci = CompletionIterator[Any, Iterator[Any]](iterToReturn, releaseLock(blockId))
Some(new BlockResult(ci, DataReadMethod.Disk, info.size))
} else {
releaseLock(blockId)
throw new SparkException(s"Block $blockId was not found even though it's read-locked")
}
}
}
/**
* Get block from the local block manager as serialized bytes.
*/
def getLocalBytes(blockId: BlockId): Option[ByteBuffer] = {
logDebug(s"Getting local block $blockId as bytes")
// As an optimization for map output fetches, if the block is for a shuffle, return it
// without acquiring a lock; the disk store never deletes (recent) items so this should work
if (blockId.isShuffle) {
val shuffleBlockResolver = shuffleManager.shuffleBlockResolver
// TODO: This should gracefully handle case where local block is not available. Currently
// downstream code will throw an exception.
Option(
shuffleBlockResolver.getBlockData(blockId.asInstanceOf[ShuffleBlockId]).nioByteBuffer())
} else {
blockInfoManager.lockForReading(blockId).map { info => doGetLocalBytes(blockId, info) }
}
}
/**
* Get block from the local block manager as serialized bytes.
*
* Must be called while holding a read lock on the block.
* Releases the read lock upon exception; keeps the read lock upon successful return.
*/
private def doGetLocalBytes(blockId: BlockId, info: BlockInfo): ByteBuffer = {
val level = info.level
logDebug(s"Level for block $blockId is $level")
// In order, try to read the serialized bytes from memory, then from disk, then fall back to
// serializing in-memory objects, and, finally, throw an exception if the block does not exist.
if (level.deserialized) {
// Try to avoid expensive serialization by reading a pre-serialized copy from disk:
if (level.useDisk && diskStore.contains(blockId)) {
// Note: we purposely do not try to put the block back into memory here. Since this branch
// handles deserialized blocks, this block may only be cached in memory as objects, not
// serialized bytes. Because the caller only requested bytes, it doesn't make sense to
// cache the block's deserialized objects since that caching may not have a payoff.
diskStore.getBytes(blockId)
} else if (level.useMemory && memoryStore.contains(blockId)) {
// The block was not found on disk, so serialize an in-memory copy:
dataSerialize(blockId, memoryStore.getValues(blockId).get)
} else {
releaseLock(blockId)
throw new SparkException(s"Block $blockId was not found even though it's read-locked")
}
} else { // storage level is serialized
if (level.useMemory && memoryStore.contains(blockId)) {
memoryStore.getBytes(blockId).get
} else if (level.useDisk && diskStore.contains(blockId)) {
maybeCacheDiskBytesInMemory(info, blockId, level, diskStore.getBytes(blockId))
} else {
releaseLock(blockId)
throw new SparkException(s"Block $blockId was not found even though it's read-locked")
}
}
}
/**
* Get block from remote block managers.
*
* This does not acquire a lock on this block in this JVM.
*/
def getRemoteValues(blockId: BlockId): Option[BlockResult] = {
getRemoteBytes(blockId).map { data =>
new BlockResult(dataDeserialize(blockId, data), DataReadMethod.Network, data.limit())
}
}
/**
* Return a list of locations for the given block, prioritizing the local machine since
* multiple block managers can share the same host.
*/
private def getLocations(blockId: BlockId): Seq[BlockManagerId] = {
val locs = Random.shuffle(master.getLocations(blockId))
val (preferredLocs, otherLocs) = locs.partition { loc => blockManagerId.host == loc.host }
preferredLocs ++ otherLocs
}
/**
* Get block from remote block managers as serialized bytes.
*/
def getRemoteBytes(blockId: BlockId): Option[ByteBuffer] = {
logDebug(s"Getting remote block $blockId")
require(blockId != null, "BlockId is null")
var runningFailureCount = 0
var totalFailureCount = 0
val locations = getLocations(blockId)
val maxFetchFailures = locations.size
var locationIterator = locations.iterator
while (locationIterator.hasNext) {
val loc = locationIterator.next()
logDebug(s"Getting remote block $blockId from $loc")
val data = try {
blockTransferService.fetchBlockSync(
loc.host, loc.port, loc.executorId, blockId.toString).nioByteBuffer()
} catch {
case NonFatal(e) =>
runningFailureCount += 1
totalFailureCount += 1
if (totalFailureCount >= maxFetchFailures) {
// Give up trying anymore locations. Either we've tried all of the original locations,
// or we've refreshed the list of locations from the master, and have still
// hit failures after trying locations from the refreshed list.
throw new BlockFetchException(s"Failed to fetch block after" +
s" ${totalFailureCount} fetch failures. Most recent failure cause:", e)
}
logWarning(s"Failed to fetch remote block $blockId " +
s"from $loc (failed attempt $runningFailureCount)", e)
// If there is a large number of executors then locations list can contain a
// large number of stale entries causing a large number of retries that may
// take a significant amount of time. To get rid of these stale entries
// we refresh the block locations after a certain number of fetch failures
if (runningFailureCount >= maxFailuresBeforeLocationRefresh) {
locationIterator = getLocations(blockId).iterator
logDebug(s"Refreshed locations from the driver " +
s"after ${runningFailureCount} fetch failures.")
runningFailureCount = 0
}
// This location failed, so we retry fetch from a different one by returning null here
null
}
if (data != null) {
return Some(data)
}
logDebug(s"The value of block $blockId is null")
}
logDebug(s"Block $blockId not found")
None
}
/**
* Get a block from the block manager (either local or remote).
*
* This acquires a read lock on the block if the block was stored locally and does not acquire
* any locks if the block was fetched from a remote block manager. The read lock will
* automatically be freed once the result's `data` iterator is fully consumed.
*/
def get(blockId: BlockId): Option[BlockResult] = {
val local = getLocalValues(blockId)
if (local.isDefined) {
logInfo(s"Found block $blockId locally")
return local
}
val remote = getRemoteValues(blockId)
if (remote.isDefined) {
logInfo(s"Found block $blockId remotely")
return remote
}
None
}
/**
* Downgrades an exclusive write lock to a shared read lock.
*/
def downgradeLock(blockId: BlockId): Unit = {
blockInfoManager.downgradeLock(blockId)
}
/**
* Release a lock on the given block.
*/
def releaseLock(blockId: BlockId): Unit = {
blockInfoManager.unlock(blockId)
}
/**
* Registers a task with the BlockManager in order to initialize per-task bookkeeping structures.
*/
def registerTask(taskAttemptId: Long): Unit = {
blockInfoManager.registerTask(taskAttemptId)
}
/**
* Release all locks for the given task.
*
* @return the blocks whose locks were released.
*/
def releaseAllLocksForTask(taskAttemptId: Long): Seq[BlockId] = {
blockInfoManager.releaseAllLocksForTask(taskAttemptId)
}
/**
* Retrieve the given block if it exists, otherwise call the provided `makeIterator` method
* to compute the block, persist it, and return its values.
*
* @return either a BlockResult if the block was successfully cached, or an iterator if the block
* could not be cached.
*/
def getOrElseUpdate(
blockId: BlockId,
level: StorageLevel,
makeIterator: () => Iterator[Any]): Either[BlockResult, Iterator[Any]] = {
// Initially we hold no locks on this block.
doPutIterator(blockId, makeIterator, level, keepReadLock = true) match {
case None =>
// doPut() didn't hand work back to us, so the block already existed or was successfully
// stored. Therefore, we now hold a read lock on the block.
val blockResult = getLocalValues(blockId).getOrElse {
// Since we held a read lock between the doPut() and get() calls, the block should not
// have been evicted, so get() not returning the block indicates some internal error.
releaseLock(blockId)
throw new SparkException(s"get() failed for block $blockId even though we held a lock")
}
// We already hold a read lock on the block from the doPut() call and getLocalValues()
// acquires the lock again, so we need to call releaseLock() here so that the net number
// of lock acquisitions is 1 (since the caller will only call release() once).
releaseLock(blockId)
Left(blockResult)
case Some(iter) =>
// The put failed, likely because the data was too large to fit in memory and could not be
// dropped to disk. Therefore, we need to pass the input iterator back to the caller so
// that they can decide what to do with the values (e.g. process them without caching).
Right(iter)
}
}
/**
* @return true if the block was stored or false if an error occurred.
*/
def putIterator(
blockId: BlockId,
values: Iterator[Any],
level: StorageLevel,
tellMaster: Boolean = true): Boolean = {
require(values != null, "Values is null")
doPutIterator(blockId, () => values, level, tellMaster) match {
case None =>
true
case Some(iter) =>
// Caller doesn't care about the iterator values, so we can close the iterator here
// to free resources earlier
iter.close()
false
}
}
/**
* A short circuited method to get a block writer that can write data directly to disk.
* The Block will be appended to the File specified by filename. Callers should handle error
* cases.
*/
def getDiskWriter(
blockId: BlockId,
file: File,
serializerInstance: SerializerInstance,
bufferSize: Int,
writeMetrics: ShuffleWriteMetrics): DiskBlockObjectWriter = {
val compressStream: OutputStream => OutputStream = wrapForCompression(blockId, _)
val syncWrites = conf.getBoolean("spark.shuffle.sync", false)
new DiskBlockObjectWriter(file, serializerInstance, bufferSize, compressStream,
syncWrites, writeMetrics, blockId)
}
/**
* Put a new block of serialized bytes to the block manager.
*
* @return true if the block was stored or false if an error occurred.
*/
def putBytes(
blockId: BlockId,
bytes: ByteBuffer,
level: StorageLevel,
tellMaster: Boolean = true): Boolean = {
require(bytes != null, "Bytes is null")
doPutBytes(blockId, bytes, level, tellMaster)
}
/**
* Put the given bytes according to the given level in one of the block stores, replicating
* the values if necessary.
*
* If the block already exists, this method will not overwrite it.
*
* @param keepReadLock if true, this method will hold the read lock when it returns (even if the
* block already exists). If false, this method will hold no locks when it
* returns.
* @return true if the block was already present or if the put succeeded, false otherwise.
*/
private def doPutBytes(
blockId: BlockId,
bytes: ByteBuffer,
level: StorageLevel,
tellMaster: Boolean = true,
keepReadLock: Boolean = false): Boolean = {
doPut(blockId, level, tellMaster = tellMaster, keepReadLock = keepReadLock) { putBlockInfo =>
val startTimeMs = System.currentTimeMillis
// Since we're storing bytes, initiate the replication before storing them locally.
// This is faster as data is already serialized and ready to send.
val replicationFuture = if (level.replication > 1) {
// Duplicate doesn't copy the bytes, but just creates a wrapper
val bufferView = bytes.duplicate()
Future {
// This is a blocking action and should run in futureExecutionContext which is a cached
// thread pool
replicate(blockId, bufferView, level)
}(futureExecutionContext)
} else {
null
}
bytes.rewind()
val size = bytes.limit()
if (level.useMemory) {
// Put it in memory first, even if it also has useDisk set to true;
// We will drop it to disk later if the memory store can't hold it.
val putSucceeded = if (level.deserialized) {
val values = dataDeserialize(blockId, bytes.duplicate())
memoryStore.putIterator(blockId, values, level) match {
case Right(_) => true
case Left(iter) =>
// If putting deserialized values in memory failed, we will put the bytes directly to
// disk, so we don't need this iterator and can close it to free resources earlier.
iter.close()
false
}
} else {
memoryStore.putBytes(blockId, size, () => bytes)
}
if (!putSucceeded && level.useDisk) {
logWarning(s"Persisting block $blockId to disk instead.")
diskStore.putBytes(blockId, bytes)
}
} else if (level.useDisk) {
diskStore.putBytes(blockId, bytes)
}
val putBlockStatus = getCurrentBlockStatus(blockId, putBlockInfo)
val blockWasSuccessfullyStored = putBlockStatus.storageLevel.isValid
if (blockWasSuccessfullyStored) {
// Now that the block is in either the memory, externalBlockStore, or disk store,
// tell the master about it.
putBlockInfo.size = size
if (tellMaster) {
reportBlockStatus(blockId, putBlockInfo, putBlockStatus)
}
Option(TaskContext.get()).foreach { c =>
c.taskMetrics().incUpdatedBlockStatuses(Seq((blockId, putBlockStatus)))
}
}
logDebug("Put block %s locally took %s".format(blockId, Utils.getUsedTimeMs(startTimeMs)))
if (level.replication > 1) {
// Wait for asynchronous replication to finish
Await.ready(replicationFuture, Duration.Inf)
}
if (blockWasSuccessfullyStored) {
None
} else {
Some(bytes)
}
}.isEmpty
}
/**
* Helper method used to abstract common code from [[doPutBytes()]] and [[doPutIterator()]].
*
* @param putBody a function which attempts the actual put() and returns None on success
* or Some on failure.
*/
private def doPut[T](
blockId: BlockId,
level: StorageLevel,
tellMaster: Boolean,
keepReadLock: Boolean)(putBody: BlockInfo => Option[T]): Option[T] = {
require(blockId != null, "BlockId is null")
require(level != null && level.isValid, "StorageLevel is null or invalid")
val putBlockInfo = {
val newInfo = new BlockInfo(level, tellMaster)
if (blockInfoManager.lockNewBlockForWriting(blockId, newInfo)) {
newInfo
} else {
logWarning(s"Block $blockId already exists on this machine; not re-adding it")
if (!keepReadLock) {
// lockNewBlockForWriting returned a read lock on the existing block, so we must free it:
releaseLock(blockId)
}
return None
}
}
val startTimeMs = System.currentTimeMillis
var blockWasSuccessfullyStored: Boolean = false
val result: Option[T] = try {
val res = putBody(putBlockInfo)
blockWasSuccessfullyStored = res.isEmpty
res
} finally {
if (blockWasSuccessfullyStored) {
if (keepReadLock) {
blockInfoManager.downgradeLock(blockId)
} else {
blockInfoManager.unlock(blockId)
}
} else {
blockInfoManager.removeBlock(blockId)
logWarning(s"Putting block $blockId failed")
}
}
if (level.replication > 1) {
logDebug("Putting block %s with replication took %s"
.format(blockId, Utils.getUsedTimeMs(startTimeMs)))
} else {
logDebug("Putting block %s without replication took %s"
.format(blockId, Utils.getUsedTimeMs(startTimeMs)))
}
result
}
/**
* Put the given block according to the given level in one of the block stores, replicating
* the values if necessary.
*
* If the block already exists, this method will not overwrite it.
*
* @param keepReadLock if true, this method will hold the read lock when it returns (even if the
* block already exists). If false, this method will hold no locks when it
* returns.
* @return None if the block was already present or if the put succeeded, or Some(iterator)
* if the put failed.
*/
private def doPutIterator(
blockId: BlockId,
iterator: () => Iterator[Any],
level: StorageLevel,
tellMaster: Boolean = true,
keepReadLock: Boolean = false): Option[PartiallyUnrolledIterator] = {
doPut(blockId, level, tellMaster = tellMaster, keepReadLock = keepReadLock) { putBlockInfo =>
val startTimeMs = System.currentTimeMillis
var iteratorFromFailedMemoryStorePut: Option[PartiallyUnrolledIterator] = None
// Size of the block in bytes
var size = 0L
if (level.useMemory) {
// Put it in memory first, even if it also has useDisk set to true;
// We will drop it to disk later if the memory store can't hold it.
memoryStore.putIterator(blockId, iterator(), level) match {
case Right(s) =>
size = s
case Left(iter) =>
// Not enough space to unroll this block; drop to disk if applicable
if (level.useDisk) {
logWarning(s"Persisting block $blockId to disk instead.")
diskStore.put(blockId) { fileOutputStream =>
dataSerializeStream(blockId, fileOutputStream, iter)
}
size = diskStore.getSize(blockId)
} else {
iteratorFromFailedMemoryStorePut = Some(iter)
}
}
} else if (level.useDisk) {
diskStore.put(blockId) { fileOutputStream =>
dataSerializeStream(blockId, fileOutputStream, iterator())
}
size = diskStore.getSize(blockId)
}
val putBlockStatus = getCurrentBlockStatus(blockId, putBlockInfo)
val blockWasSuccessfullyStored = putBlockStatus.storageLevel.isValid
if (blockWasSuccessfullyStored) {
// Now that the block is in either the memory, externalBlockStore, or disk store,
// tell the master about it.
putBlockInfo.size = size
if (tellMaster) {
reportBlockStatus(blockId, putBlockInfo, putBlockStatus)
}
Option(TaskContext.get()).foreach { c =>
c.taskMetrics().incUpdatedBlockStatuses(Seq((blockId, putBlockStatus)))
}
logDebug("Put block %s locally took %s".format(blockId, Utils.getUsedTimeMs(startTimeMs)))
if (level.replication > 1) {
val remoteStartTime = System.currentTimeMillis
val bytesToReplicate = doGetLocalBytes(blockId, putBlockInfo)
try {
replicate(blockId, bytesToReplicate, level)
} finally {
BlockManager.dispose(bytesToReplicate)
}
logDebug("Put block %s remotely took %s"
.format(blockId, Utils.getUsedTimeMs(remoteStartTime)))
}
}
assert(blockWasSuccessfullyStored == iteratorFromFailedMemoryStorePut.isEmpty)
iteratorFromFailedMemoryStorePut
}
}
/**
* Attempts to cache spilled bytes read from disk into the MemoryStore in order to speed up
* subsequent reads. This method requires the caller to hold a read lock on the block.
*
* @return a copy of the bytes. The original bytes passed this method should no longer
* be used after this method returns.
*/
private def maybeCacheDiskBytesInMemory(
blockInfo: BlockInfo,
blockId: BlockId,
level: StorageLevel,
diskBytes: ByteBuffer): ByteBuffer = {
require(!level.deserialized)
if (level.useMemory) {
// Synchronize on blockInfo to guard against a race condition where two readers both try to
// put values read from disk into the MemoryStore.
blockInfo.synchronized {
if (memoryStore.contains(blockId)) {
BlockManager.dispose(diskBytes)
memoryStore.getBytes(blockId).get
} else {
val putSucceeded = memoryStore.putBytes(blockId, diskBytes.limit(), () => {
// https://issues.apache.org/jira/browse/SPARK-6076
// If the file size is bigger than the free memory, OOM will happen. So if we
// cannot put it into MemoryStore, copyForMemory should not be created. That's why
// this action is put into a `() => ByteBuffer` and created lazily.
val copyForMemory = ByteBuffer.allocate(diskBytes.limit)
copyForMemory.put(diskBytes)
})
if (putSucceeded) {
BlockManager.dispose(diskBytes)
memoryStore.getBytes(blockId).get
} else {
diskBytes.rewind()
diskBytes
}
}
}
} else {
diskBytes
}
}
/**
* Attempts to cache spilled values read from disk into the MemoryStore in order to speed up
* subsequent reads. This method requires the caller to hold a read lock on the block.
*
* @return a copy of the iterator. The original iterator passed this method should no longer
* be used after this method returns.
*/
private def maybeCacheDiskValuesInMemory(
blockInfo: BlockInfo,
blockId: BlockId,
level: StorageLevel,
diskIterator: Iterator[Any]): Iterator[Any] = {
require(level.deserialized)
if (level.useMemory) {
// Synchronize on blockInfo to guard against a race condition where two readers both try to
// put values read from disk into the MemoryStore.
blockInfo.synchronized {
if (memoryStore.contains(blockId)) {
// Note: if we had a means to discard the disk iterator, we would do that here.
memoryStore.getValues(blockId).get
} else {
memoryStore.putIterator(blockId, diskIterator, level) match {
case Left(iter) =>
// The memory store put() failed, so it returned the iterator back to us:
iter
case Right(_) =>
// The put() succeeded, so we can read the values back:
memoryStore.getValues(blockId).get
}
}
}
} else {
diskIterator
}
}
/**
* Get peer block managers in the system.
*/
private def getPeers(forceFetch: Boolean): Seq[BlockManagerId] = {
peerFetchLock.synchronized {
val cachedPeersTtl = conf.getInt("spark.storage.cachedPeersTtl", 60 * 1000) // milliseconds
val timeout = System.currentTimeMillis - lastPeerFetchTime > cachedPeersTtl
if (cachedPeers == null || forceFetch || timeout) {
cachedPeers = master.getPeers(blockManagerId).sortBy(_.hashCode)
lastPeerFetchTime = System.currentTimeMillis
logDebug("Fetched peers from master: " + cachedPeers.mkString("[", ",", "]"))
}
cachedPeers
}
}
/**
* Replicate block to another node. Not that this is a blocking call that returns after
* the block has been replicated.
*/
private def replicate(blockId: BlockId, data: ByteBuffer, level: StorageLevel): Unit = {
val maxReplicationFailures = conf.getInt("spark.storage.maxReplicationFailures", 1)
val numPeersToReplicateTo = level.replication - 1
val peersForReplication = new ArrayBuffer[BlockManagerId]
val peersReplicatedTo = new ArrayBuffer[BlockManagerId]
val peersFailedToReplicateTo = new ArrayBuffer[BlockManagerId]
val tLevel = StorageLevel(level.useDisk, level.useMemory, level.deserialized, 1)
val startTime = System.currentTimeMillis
val random = new Random(blockId.hashCode)
var replicationFailed = false
var failures = 0
var done = false
// Get cached list of peers
peersForReplication ++= getPeers(forceFetch = false)
// Get a random peer. Note that this selection of a peer is deterministic on the block id.
// So assuming the list of peers does not change and no replication failures,
// if there are multiple attempts in the same node to replicate the same block,
// the same set of peers will be selected.
def getRandomPeer(): Option[BlockManagerId] = {
// If replication had failed, then force update the cached list of peers and remove the peers
// that have been already used
if (replicationFailed) {
peersForReplication.clear()
peersForReplication ++= getPeers(forceFetch = true)
peersForReplication --= peersReplicatedTo
peersForReplication --= peersFailedToReplicateTo
}
if (!peersForReplication.isEmpty) {
Some(peersForReplication(random.nextInt(peersForReplication.size)))
} else {
None
}
}
// One by one choose a random peer and try uploading the block to it
// If replication fails (e.g., target peer is down), force the list of cached peers
// to be re-fetched from driver and then pick another random peer for replication. Also
// temporarily black list the peer for which replication failed.
//
// This selection of a peer and replication is continued in a loop until one of the
// following 3 conditions is fulfilled:
// (i) specified number of peers have been replicated to
// (ii) too many failures in replicating to peers
// (iii) no peer left to replicate to
//
while (!done) {
getRandomPeer() match {
case Some(peer) =>
try {
val onePeerStartTime = System.currentTimeMillis
data.rewind()
logTrace(s"Trying to replicate $blockId of ${data.limit()} bytes to $peer")
blockTransferService.uploadBlockSync(
peer.host, peer.port, peer.executorId, blockId, new NioManagedBuffer(data), tLevel)
logTrace(s"Replicated $blockId of ${data.limit()} bytes to $peer in %s ms"
.format(System.currentTimeMillis - onePeerStartTime))
peersReplicatedTo += peer
peersForReplication -= peer
replicationFailed = false
if (peersReplicatedTo.size == numPeersToReplicateTo) {
done = true // specified number of peers have been replicated to
}
} catch {
case e: Exception =>
logWarning(s"Failed to replicate $blockId to $peer, failure #$failures", e)
failures += 1
replicationFailed = true
peersFailedToReplicateTo += peer
if (failures > maxReplicationFailures) { // too many failures in replicating to peers
done = true
}
}
case None => // no peer left to replicate to
done = true
}
}
val timeTakeMs = (System.currentTimeMillis - startTime)
logDebug(s"Replicating $blockId of ${data.limit()} bytes to " +
s"${peersReplicatedTo.size} peer(s) took $timeTakeMs ms")
if (peersReplicatedTo.size < numPeersToReplicateTo) {
logWarning(s"Block $blockId replicated to only " +
s"${peersReplicatedTo.size} peer(s) instead of $numPeersToReplicateTo peers")
}
}
/**
* Read a block consisting of a single object.
*/
def getSingle(blockId: BlockId): Option[Any] = {
get(blockId).map(_.data.next())
}
/**
* Write a block consisting of a single object.
*
* @return true if the block was stored or false if the block was already stored or an
* error occurred.
*/
def putSingle(
blockId: BlockId,
value: Any,
level: StorageLevel,
tellMaster: Boolean = true): Boolean = {
putIterator(blockId, Iterator(value), level, tellMaster)
}
/**
* Drop a block from memory, possibly putting it on disk if applicable. Called when the memory
* store reaches its limit and needs to free up space.
*
* If `data` is not put on disk, it won't be created.
*
* The caller of this method must hold a write lock on the block before calling this method.
* This method does not release the write lock.
*
* @return the block's new effective StorageLevel.
*/
def dropFromMemory(
blockId: BlockId,
data: () => Either[Array[Any], ByteBuffer]): StorageLevel = {
logInfo(s"Dropping block $blockId from memory")
val info = blockInfoManager.assertBlockIsLockedForWriting(blockId)
var blockIsUpdated = false
val level = info.level
// Drop to disk, if storage level requires
if (level.useDisk && !diskStore.contains(blockId)) {
logInfo(s"Writing block $blockId to disk")
data() match {
case Left(elements) =>
diskStore.put(blockId) { fileOutputStream =>
dataSerializeStream(blockId, fileOutputStream, elements.toIterator)
}
case Right(bytes) =>
diskStore.putBytes(blockId, bytes)
}
blockIsUpdated = true
}
// Actually drop from memory store
val droppedMemorySize =
if (memoryStore.contains(blockId)) memoryStore.getSize(blockId) else 0L
val blockIsRemoved = memoryStore.remove(blockId)
if (blockIsRemoved) {
blockIsUpdated = true
} else {
logWarning(s"Block $blockId could not be dropped from memory as it does not exist")
}
val status = getCurrentBlockStatus(blockId, info)
if (info.tellMaster) {
reportBlockStatus(blockId, info, status, droppedMemorySize)
}
if (blockIsUpdated) {
Option(TaskContext.get()).foreach { c =>
c.taskMetrics().incUpdatedBlockStatuses(Seq((blockId, status)))
}
}
status.storageLevel
}
/**
* Remove all blocks belonging to the given RDD.
*
* @return The number of blocks removed.
*/
def removeRdd(rddId: Int): Int = {
// TODO: Avoid a linear scan by creating another mapping of RDD.id to blocks.
logInfo(s"Removing RDD $rddId")
val blocksToRemove = blockInfoManager.entries.flatMap(_._1.asRDDId).filter(_.rddId == rddId)
blocksToRemove.foreach { blockId => removeBlock(blockId, tellMaster = false) }
blocksToRemove.size
}
/**
* Remove all blocks belonging to the given broadcast.
*/
def removeBroadcast(broadcastId: Long, tellMaster: Boolean): Int = {
logDebug(s"Removing broadcast $broadcastId")
val blocksToRemove = blockInfoManager.entries.map(_._1).collect {
case bid @ BroadcastBlockId(`broadcastId`, _) => bid
}
blocksToRemove.foreach { blockId => removeBlock(blockId, tellMaster) }
blocksToRemove.size
}
/**
* Remove a block from both memory and disk.
*/
def removeBlock(blockId: BlockId, tellMaster: Boolean = true): Unit = {
logDebug(s"Removing block $blockId")
blockInfoManager.lockForWriting(blockId) match {
case None =>
// The block has already been removed; do nothing.
logWarning(s"Asked to remove block $blockId, which does not exist")
case Some(info) =>
// Removals are idempotent in disk store and memory store. At worst, we get a warning.
val removedFromMemory = memoryStore.remove(blockId)
val removedFromDisk = diskStore.remove(blockId)
if (!removedFromMemory && !removedFromDisk) {
logWarning(s"Block $blockId could not be removed as it was not found in either " +
"the disk, memory, or external block store")
}
blockInfoManager.removeBlock(blockId)
if (tellMaster && info.tellMaster) {
val status = getCurrentBlockStatus(blockId, info)
reportBlockStatus(blockId, info, status)
}
}
}
private def shouldCompress(blockId: BlockId): Boolean = {
blockId match {
case _: ShuffleBlockId => compressShuffle
case _: BroadcastBlockId => compressBroadcast
case _: RDDBlockId => compressRdds
case _: TempLocalBlockId => compressShuffleSpill
case _: TempShuffleBlockId => compressShuffle
case _ => false
}
}
/**
* Wrap an output stream for compression if block compression is enabled for its block type
*/
def wrapForCompression(blockId: BlockId, s: OutputStream): OutputStream = {
if (shouldCompress(blockId)) compressionCodec.compressedOutputStream(s) else s
}
/**
* Wrap an input stream for compression if block compression is enabled for its block type
*/
def wrapForCompression(blockId: BlockId, s: InputStream): InputStream = {
if (shouldCompress(blockId)) compressionCodec.compressedInputStream(s) else s
}
/** Serializes into a stream. */
def dataSerializeStream(
blockId: BlockId,
outputStream: OutputStream,
values: Iterator[Any]): Unit = {
val byteStream = new BufferedOutputStream(outputStream)
val ser = defaultSerializer.newInstance()
ser.serializeStream(wrapForCompression(blockId, byteStream)).writeAll(values).close()
}
/** Serializes into a byte buffer. */
def dataSerialize(blockId: BlockId, values: Iterator[Any]): ByteBuffer = {
val byteStream = new ByteBufferOutputStream(4096)
dataSerializeStream(blockId, byteStream, values)
byteStream.toByteBuffer
}
/**
* Deserializes a ByteBuffer into an iterator of values and disposes of it when the end of
* the iterator is reached.
*/
def dataDeserialize(blockId: BlockId, bytes: ByteBuffer): Iterator[Any] = {
bytes.rewind()
dataDeserializeStream(blockId, new ByteBufferInputStream(bytes, true))
}
/**
* Deserializes a InputStream into an iterator of values and disposes of it when the end of
* the iterator is reached.
*/
def dataDeserializeStream(blockId: BlockId, inputStream: InputStream): Iterator[Any] = {
val stream = new BufferedInputStream(inputStream)
defaultSerializer
.newInstance()
.deserializeStream(wrapForCompression(blockId, stream))
.asIterator
}
def stop(): Unit = {
blockTransferService.close()
if (shuffleClient ne blockTransferService) {
// Closing should be idempotent, but maybe not for the NioBlockTransferService.
shuffleClient.close()
}
diskBlockManager.stop()
rpcEnv.stop(slaveEndpoint)
blockInfoManager.clear()
memoryStore.clear()
futureExecutionContext.shutdownNow()
logInfo("BlockManager stopped")
}
}
private[spark] object BlockManager extends Logging {
private val ID_GENERATOR = new IdGenerator
/**
* Attempt to clean up a ByteBuffer if it is memory-mapped. This uses an *unsafe* Sun API that
* might cause errors if one attempts to read from the unmapped buffer, but it's better than
* waiting for the GC to find it because that could lead to huge numbers of open files. There's
* unfortunately no standard API to do this.
*/
def dispose(buffer: ByteBuffer): Unit = {
if (buffer != null && buffer.isInstanceOf[MappedByteBuffer]) {
logTrace(s"Unmapping $buffer")
if (buffer.asInstanceOf[DirectBuffer].cleaner() != null) {
buffer.asInstanceOf[DirectBuffer].cleaner().clean()
}
}
}
def blockIdsToHosts(
blockIds: Array[BlockId],
env: SparkEnv,
blockManagerMaster: BlockManagerMaster = null): Map[BlockId, Seq[String]] = {
// blockManagerMaster != null is used in tests
assert(env != null || blockManagerMaster != null)
val blockLocations: Seq[Seq[BlockManagerId]] = if (blockManagerMaster == null) {
env.blockManager.getLocationBlockIds(blockIds)
} else {
blockManagerMaster.getLocations(blockIds)
}
val blockManagers = new HashMap[BlockId, Seq[String]]
for (i <- 0 until blockIds.length) {
blockManagers(blockIds(i)) = blockLocations(i).map(_.host)
}
blockManagers.toMap
}
}
|