Package org.apache.hadoop.hive.ql.stats

Examples of org.apache.hadoop.hive.ql.stats.StatsPublisher

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
  private void publishStats() throws HiveException {
    boolean isStatsReliable = conf.isStatsReliable();

    // Initializing a stats publisher
    StatsPublisher statsPublisher = Utilities.getStatsPublisher(jc);
    if (!statsPublisher.connect(jc)) {
      // just return, stats gathering should not block the main query.
      LOG.info("StatsPublishing error: cannot connect to database.");
      if (isStatsReliable) {
        throw new HiveException(ErrorMsg.STATSPUBLISHER_CONNECTION_ERROR.getErrorCodedMsg());
      }
      return;
    }

    String key;
    String taskID = Utilities.getTaskIdFromFilename(Utilities.getTaskId(hconf));
    Map<String, String> statsToPublish = new HashMap<String, String>();

    for (String pspecs : stats.keySet()) {
      statsToPublish.clear();
      if (pspecs.isEmpty()) {
        // In case of a non-partitioned table, the key for temp storage is just
        // "tableName + taskID"
        key = conf.getStatsAggPrefix() + taskID;
      } else {
        // In case of a partition, the key for temp storage is
        // "tableName + partitionSpecs + taskID"
        key = conf.getStatsAggPrefix() + pspecs + Path.SEPARATOR + taskID;
      }
      for(String statType : stats.get(pspecs).getStoredStats()) {
        statsToPublish.put(statType, Long.toString(stats.get(pspecs).getStat(statType)));
      }
      if (!statsPublisher.publishStat(key, statsToPublish)) {
        if (isStatsReliable) {
          throw new HiveException(ErrorMsg.STATSPUBLISHER_PUBLISHING_ERROR.getErrorCodedMsg());
        }
      }
      LOG.info("publishing : " + key + " : " + statsToPublish.toString());
    }
    if (!statsPublisher.closeConnection()) {
      if (isStatsReliable) {
        throw new HiveException(ErrorMsg.STATSPUBLISHER_CLOSING_ERROR.getErrorCodedMsg());
      }
    }
  }
View Full Code Here
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
  private void publishStats() throws HiveException {
    boolean isStatsReliable = conf.isStatsReliable();

    // Initializing a stats publisher
    StatsPublisher statsPublisher = Utilities.getStatsPublisher(jc);

    if (statsPublisher == null) {
      // just return, stats gathering should not block the main query
      LOG.error("StatsPublishing error: StatsPublisher is not initialized.");
      if (isStatsReliable) {
        throw new HiveException(ErrorMsg.STATSPUBLISHER_NOT_OBTAINED.getErrorCodedMsg());
      }
      return;
    }

    if (!statsPublisher.connect(hconf)) {
      // just return, stats gathering should not block the main query
      LOG.error("StatsPublishing error: cannot connect to database");
      if (isStatsReliable) {
        throw new HiveException(ErrorMsg.STATSPUBLISHER_CONNECTION_ERROR.getErrorCodedMsg());
      }
      return;
    }

    String taskID = Utilities.getTaskIdFromFilename(Utilities.getTaskId(hconf));
    String spSpec = conf.getStaticSpec() != null ? conf.getStaticSpec() : "";

    for (String fspKey : valToPaths.keySet()) {
      FSPaths fspValue = valToPaths.get(fspKey);
      String key;

      // construct the key(fileID) to insert into the intermediate stats table
      if (fspKey == "") {
        // for non-partitioned/static partitioned table, the key for temp storage is
        // common key prefix + static partition spec + taskID
        key = conf.getStatsAggPrefix() + spSpec + taskID;
      } else {
        // for partitioned table, the key is
        // common key prefix + static partition spec + DynamicPartSpec + taskID
        key = conf.getStatsAggPrefix() + spSpec + fspKey + Path.SEPARATOR + taskID;
      }
      Map<String, String> statsToPublish = new HashMap<String, String>();
      for (String statType : fspValue.stat.getStoredStats()) {
        statsToPublish.put(statType, Long.toString(fspValue.stat.getStat(statType)));
      }
      if (!statsPublisher.publishStat(key, statsToPublish)) {
        // The original exception is lost.
        // Not changing the interface to maintain backward compatibility
        if (isStatsReliable) {
          throw new HiveException(ErrorMsg.STATSPUBLISHER_PUBLISHING_ERROR.getErrorCodedMsg());
        }
      }
    }
    if (!statsPublisher.closeConnection()) {
      // The original exception is lost.
      // Not changing the interface to maintain backward compatibility
      if (isStatsReliable) {
        throw new HiveException(ErrorMsg.STATSPUBLISHER_CLOSING_ERROR.getErrorCodedMsg());
      }
View Full Code Here
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
      // make this client wait if job trcker is not behaving well.
      Throttle.checkJobTracker(job, LOG);

      if (work.isGatheringStats()) {
        // initialize stats publishing table
        StatsPublisher statsPublisher;
        String statsImplementationClass = HiveConf.getVar(job, HiveConf.ConfVars.HIVESTATSDBCLASS);
        if (StatsFactory.setImplementation(statsImplementationClass, job)) {
          statsPublisher = StatsFactory.getStatsPublisher();
          if (!statsPublisher.init(job)) { // creating stats table if not exists
            if (HiveConf.getBoolVar(job, HiveConf.ConfVars.HIVE_STATS_RELIABLE)) {
              throw
                new HiveException(ErrorMsg.STATSPUBLISHER_INITIALIZATION_ERROR.getErrorCodedMsg());
            }
          }
View Full Code Here
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
    return OperatorType.TABLESCAN;
  }

  private void publishStats() {
    // Initializing a stats publisher
    StatsPublisher statsPublisher = Utilities.getStatsPublisher(jc);
    if (!statsPublisher.connect(jc)) {
      // just return, stats gathering should not block the main query.
      LOG.info("StatsPublishing error: cannot connect to database.");
      return;
    }

    String key;
    String taskID = Utilities.getTaskIdFromFilename(Utilities.getTaskId(hconf));
    Map<String, String> statsToPublish = new HashMap<String, String>();

    for (String pspecs : stats.keySet()) {
      statsToPublish.clear();
      if (pspecs.isEmpty()) {
        // In case of a non-partitioned table, the key for temp storage is just
        // "tableName + taskID"
        key = conf.getStatsAggPrefix() + taskID;
      } else {
        // In case of a partition, the key for temp storage is
        // "tableName + partitionSpecs + taskID"
        key = conf.getStatsAggPrefix() + pspecs + Path.SEPARATOR + taskID;
      }
      for(String statType : stats.get(pspecs).getStoredStats()) {
        statsToPublish.put(statType, Long.toString(stats.get(pspecs).getStat(statType)));
      }
      statsPublisher.publishStat(key, statsToPublish);
      LOG.info("publishing : " + key + " : " + statsToPublish.toString());
    }
    statsPublisher.closeConnection();
  }
View Full Code Here
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
    }
  }

  private void publishStats() {
    // Initializing a stats publisher
    StatsPublisher statsPublisher = Utilities.getStatsPublisher(jc);

    if (statsPublisher == null) {
      // just return, stats gathering should not block the main query
      LOG.error("StatsPublishing error: StatsPublisher is not initialized.");
      return;
    }
    if (!statsPublisher.connect(hconf)) {
      // just return, stats gathering should not block the main query
      LOG.error("StatsPublishing error: cannot connect to database");
      return;
    }

    String taskID = Utilities.getTaskIdFromFilename(Utilities.getTaskId(hconf));
    String spSpec = conf.getStaticSpec() != null ? conf.getStaticSpec() : "";

    for (String fspKey : valToPaths.keySet()) {
      FSPaths fspValue = valToPaths.get(fspKey);
      String key;

      // construct the key(fileID) to insert into the intermediate stats table
      if (fspKey == "") {
        // for non-partitioned/static partitioned table, the key for temp storage is
        // common key prefix + static partition spec + taskID
        key = conf.getStatsAggPrefix() + spSpec + taskID;
      } else {
        // for partitioned table, the key is
        // common key prefix + static partition spec + DynamicPartSpec + taskID
        key = conf.getStatsAggPrefix() + spSpec + fspKey + Path.SEPARATOR + taskID;
      }
      Map<String, String> statsToPublish = new HashMap<String, String>();
      for (String statType : fspValue.stat.getStoredStats()) {
        statsToPublish.put(statType, Long.toString(fspValue.stat.getStat(statType)));
      }
      statsPublisher.publishStat(key, statsToPublish);
    }
    statsPublisher.closeConnection();
  }
View Full Code Here
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
  public void testStatsPublisherOneStat() throws Throwable {
    try {
      System.out.println("StatsPublisher - one stat published per key - aggregating matching key");

      // instantiate stats publisher
      StatsPublisher statsPublisher = Utilities.getStatsPublisher((JobConf) conf);
      assertNotNull(statsPublisher);
      assertTrue(statsPublisher.init(conf));
      assertTrue(statsPublisher.connect(conf));

      // instantiate stats aggregator
      StatsAggregator statsAggregator = StatsFactory.getStatsAggregator();
      assertNotNull(statsAggregator);
      assertTrue(statsAggregator.connect(conf));

      // publish stats
      fillStatMap("200", "1000");
      assertTrue(statsPublisher.publishStat("file_00000", stats));
      fillStatMap("400", "3000");
      assertTrue(statsPublisher.publishStat("file_00001", stats));


      // aggregate existing stats
      String rows0 = statsAggregator.aggregateStats("file_00000", StatsSetupConst.ROW_COUNT);
      assertEquals("200", rows0);
      String usize0 = statsAggregator.aggregateStats("file_00000",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("1000", usize0);

      String rows1 = statsAggregator.aggregateStats("file_00001", StatsSetupConst.ROW_COUNT);
      assertEquals("400", rows1);
      String usize1 = statsAggregator.aggregateStats("file_00001",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("3000", usize1);

      // close connections
      assertTrue(statsPublisher.closeConnection());
      assertTrue(statsAggregator.closeConnection());

      System.out
          .println("StatsPublisher - one stat published per key - aggregating matching key - OK");
    } catch (Throwable e) {
View Full Code Here
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
  public void testStatsPublisher() throws Throwable {
    try {
      System.out.println("StatsPublisher - basic functionality");

      // instantiate stats publisher
      StatsPublisher statsPublisher = Utilities.getStatsPublisher(
          (JobConf) conf);
      assertNotNull(statsPublisher);
      assertTrue(statsPublisher.init(conf));
      assertTrue(statsPublisher.connect(conf));

      // instantiate stats aggregator
      StatsAggregator statsAggregator = StatsFactory.getStatsAggregator();
      assertNotNull(statsAggregator);
      assertTrue(statsAggregator.connect(conf));
      // statsAggregator.cleanUp("file_0000");
      // assertTrue(statsAggregator.connect(conf));

      // publish stats
      fillStatMap("200", "1000");
      assertTrue(statsPublisher.publishStat("file_00000_a", stats));
      fillStatMap("300", "2000");
      assertTrue(statsPublisher.publishStat("file_00000_b", stats));

      fillStatMap("400", "3000");
      assertTrue(statsPublisher.publishStat("file_00001_a", stats));
      fillStatMap("500", "4000");
      assertTrue(statsPublisher.publishStat("file_00001_b", stats));

      // aggregate existing stats
      String rows0 = statsAggregator.aggregateStats("file_00000", StatsSetupConst.ROW_COUNT);
      assertEquals("500", rows0);
      String usize0 = statsAggregator.aggregateStats("file_00000",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("3000", usize0);

      String rows1 = statsAggregator.aggregateStats("file_00001", StatsSetupConst.ROW_COUNT);
      assertEquals("900", rows1);
      String usize1 = statsAggregator.aggregateStats("file_00001",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("7000", usize1);

      // aggregate non-existent stats
      String rowsX = statsAggregator.aggregateStats("file_00002", StatsSetupConst.ROW_COUNT);
      assertEquals("0", rowsX);
      String usizeX = statsAggregator.aggregateStats("file_00002",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("0", usizeX);

      assertTrue(statsAggregator.cleanUp("file_0000"));

      // close connections
      assertTrue(statsPublisher.closeConnection());
      assertTrue(statsAggregator.closeConnection());

      System.out.println("StatsPublisher - basic functionality - OK");
    } catch (Throwable e) {
      e.printStackTrace();
View Full Code Here
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
  public void testStatsPublisherMultipleUpdates() throws Throwable {
    try {
      System.out.println("StatsPublisher - multiple updates");

      // instantiate stats publisher
      StatsPublisher statsPublisher = Utilities.getStatsPublisher((JobConf) conf);
      assertNotNull(statsPublisher);
      assertTrue(statsPublisher.init(conf));
      assertTrue(statsPublisher.connect(conf));

      // instantiate stats aggregator
      StatsAggregator statsAggregator = StatsFactory.getStatsAggregator();
      assertNotNull(statsAggregator);
      assertTrue(statsAggregator.connect(conf));

      // publish stats
      fillStatMap("200", "1000");
      assertTrue(statsPublisher.publishStat("file_00000_a", stats));
      fillStatMap("300", "2000");
      assertTrue(statsPublisher.publishStat("file_00000_b", stats));

      fillStatMap("400", "3000");
      assertTrue(statsPublisher.publishStat("file_00001_a", stats));
      fillStatMap("500", "4000");
      assertTrue(statsPublisher.publishStat("file_00001_b", stats));

      // update which should not take any effect
      fillStatMap("190", "1000");
      assertTrue(statsPublisher.publishStat("file_00000_a", stats));
      fillStatMap("290", "2000");
      assertTrue(statsPublisher.publishStat("file_00000_b", stats));

      // update that should take effect
      fillStatMap("500", "5000");
      assertTrue(statsPublisher.publishStat("file_00001_a", stats));
      fillStatMap("600", "6000");
      assertTrue(statsPublisher.publishStat("file_00001_b", stats));

      // aggregate existing stats
      String rows0 = statsAggregator.aggregateStats("file_00000", StatsSetupConst.ROW_COUNT);
      assertEquals("500", rows0);
      String usize0 = statsAggregator.aggregateStats("file_00000",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("3000", usize0);

      String rows1 = statsAggregator.aggregateStats("file_00001", StatsSetupConst.ROW_COUNT);
      assertEquals("1100", rows1);
      String usize1 = statsAggregator.aggregateStats("file_00001",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("11000", usize1);

      assertTrue(statsAggregator.cleanUp("file_0000"));

      // close connections
      assertTrue(statsPublisher.closeConnection());
      assertTrue(statsAggregator.closeConnection());

      System.out.println("StatsPublisher - multiple updates - OK");
    } catch (Throwable e) {
      e.printStackTrace();
View Full Code Here
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
    try {
      System.out
          .println("StatsPublisher - (multiple updates + publishing subset of supported statistics)");

      // instantiate stats publisher
      StatsPublisher statsPublisher = Utilities.getStatsPublisher((JobConf) conf);
      assertNotNull(statsPublisher);
      assertTrue(statsPublisher.init(conf));
      assertTrue(statsPublisher.connect(conf));

      // instantiate stats aggregator
      StatsAggregator statsAggregator = StatsFactory.getStatsAggregator();
      assertNotNull(statsAggregator);
      assertTrue(statsAggregator.connect(conf));

      // publish stats
      fillStatMap("200", "");
      assertTrue(statsPublisher.publishStat("file_00000_a", stats));
      fillStatMap("300", "2000");
      assertTrue(statsPublisher.publishStat("file_00000_b", stats));


      // aggregate existing stats
      String rows0 = statsAggregator.aggregateStats("file_00000", StatsSetupConst.ROW_COUNT);
      assertEquals("500", rows0);
      String usize0 = statsAggregator.aggregateStats("file_00000",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("2000", usize0);

      // update which should not take any effect - plus the map published is a supset of supported
      // stats
      fillStatMap("190", "");
      assertTrue(statsPublisher.publishStat("file_00000_a", stats));
      fillStatMap("290", "");
      assertTrue(statsPublisher.publishStat("file_00000_b", stats));

      // nothing changed
      rows0 = statsAggregator.aggregateStats("file_00000", StatsSetupConst.ROW_COUNT);
      assertEquals("500", rows0);
      usize0 = statsAggregator.aggregateStats("file_00000",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("2000", usize0);

      fillStatMap("500", "");
      assertTrue(statsPublisher.publishStat("file_00000_a", stats));
      fillStatMap("500", "");
      assertTrue(statsPublisher.publishStat("file_00000_b", stats));

      // changed + the rawDataSize size was overwriten !!!
      rows0 = statsAggregator.aggregateStats("file_00000", StatsSetupConst.ROW_COUNT);
      assertEquals("1000", rows0);
      usize0 = statsAggregator.aggregateStats("file_00000",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("0", usize0);

      assertTrue(statsAggregator.cleanUp("file_0000"));

      // close connections
      assertTrue(statsPublisher.closeConnection());
      assertTrue(statsAggregator.closeConnection());

      System.out
          .println("StatsPublisher - (multiple updates + publishing subset of supported statistics) - OK");
    } catch (Throwable e) {
View Full Code Here
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
  public void testStatsAggregatorCleanUp() throws Throwable {
    try {
      System.out.println("StatsAggregator - clean-up");

      // instantiate stats publisher
      StatsPublisher statsPublisher = Utilities.getStatsPublisher((JobConf) conf);
      assertNotNull(statsPublisher);
      assertTrue(statsPublisher.init(conf));
      assertTrue(statsPublisher.connect(conf));

      // instantiate stats aggregator
      StatsAggregator statsAggregator = StatsFactory.getStatsAggregator();
      assertNotNull(statsAggregator);
      assertTrue(statsAggregator.connect(conf));

      // publish stats
      fillStatMap("200", "1000");
      assertTrue(statsPublisher.publishStat("file_00000_a", stats));
      fillStatMap("300", "2000");
      assertTrue(statsPublisher.publishStat("file_00000_b", stats));

      fillStatMap("400", "3000");
      assertTrue(statsPublisher.publishStat("file_00001_a", stats));
      fillStatMap("500", "4000");
      assertTrue(statsPublisher.publishStat("file_00001_b", stats));

      // cleanUp
      assertTrue(statsAggregator.cleanUp("file_00000"));

      // now clean-up just for one key
      String rows0 = statsAggregator.aggregateStats("file_00000", StatsSetupConst.ROW_COUNT);
      assertEquals("0", rows0);
      String usize0 = statsAggregator.aggregateStats("file_00000",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("0", usize0);

      // this should still be in the table
      String rows1 = statsAggregator.aggregateStats("file_00001", StatsSetupConst.ROW_COUNT);
      assertEquals("900", rows1);
      String usize1 = statsAggregator.aggregateStats("file_00001",
          StatsSetupConst.RAW_DATA_SIZE);
      assertEquals("7000", usize1);

      assertTrue(statsAggregator.cleanUp("file_0000"));

      // close connections
      assertTrue(statsPublisher.closeConnection());
      assertTrue(statsAggregator.closeConnection());

      System.out.println("StatsAggregator - clean-up - OK");
    } catch (Throwable e) {
      e.printStackTrace();
View Full Code Here
TOP

Related Classes of org.apache.hadoop.hive.ql.stats.StatsPublisher

  • org.apache.hadoop.hive.ql.exec.ExecDriver
  • org.apache.hadoop.hive.ql.exec.FileSinkOperator
  • org.apache.hadoop.hive.ql.exec.mr.ExecDriver
  • org.apache.hadoop.hive.ql.exec.StatsTask
  • org.apache.hadoop.hive.ql.exec.TableScanOperator
  • org.apache.hadoop.hive.ql.exec.TestStatsPublisherEnhanced
  • org.apache.hadoop.hive.ql.exec.tez.DagUtils
  • org.apache.hadoop.hive.ql.io.rcfile.stats.PartialScanMapper
  • org.apache.hadoop.hive.ql.io.rcfile.stats.PartialScanTask
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.