1 /**
2 * Licensed to the Apache Software Foundation (ASF) under one
3 * or more contributor license agreements. See the NOTICE file
4 * distributed with this work for additional information
5 * regarding copyright ownership. The ASF licenses this file
6 * to you under the Apache License, Version 2.0 (the
7 * "License"); you may not use this file except in compliance
8 * with the License. You may obtain a copy of the License at
9 *
10 * http://www.apache.org/licenses/LICENSE-2.0
11 *
12 * Unless required by applicable law or agreed to in writing, software
13 * distributed under the License is distributed on an "AS IS" BASIS,
14 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
15 * See the License for the specific language governing permissions and
16 * limitations under the License.
17 */
18 package org.apache.hadoop.hbase.regionserver;
19
20 import java.io.IOException;
21 import java.util.Collection;
22 import java.util.List;
23 import java.util.NavigableSet;
24
25 import org.apache.hadoop.classification.InterfaceAudience;
26 import org.apache.hadoop.classification.InterfaceStability;
27 import org.apache.hadoop.fs.FileSystem;
28 import org.apache.hadoop.fs.Path;
29 import org.apache.hadoop.hbase.Cell;
30 import org.apache.hadoop.hbase.HColumnDescriptor;
31 import org.apache.hadoop.hbase.HRegionInfo;
32 import org.apache.hadoop.hbase.KeyValue;
33 import org.apache.hadoop.hbase.client.Scan;
34 import org.apache.hadoop.hbase.io.HeapSize;
35 import org.apache.hadoop.hbase.io.compress.Compression;
36 import org.apache.hadoop.hbase.io.hfile.CacheConfig;
37 import org.apache.hadoop.hbase.io.hfile.HFileDataBlockEncoder;
38 import org.apache.hadoop.hbase.protobuf.generated.WALProtos.CompactionDescriptor;
39 import org.apache.hadoop.hbase.regionserver.compactions.CompactionContext;
40 import org.apache.hadoop.hbase.regionserver.compactions.CompactionProgress;
41 import org.apache.hadoop.hbase.regionserver.compactions.CompactionRequest;
42
43 /**
44 * Interface for objects that hold a column family in a Region. Its a memstore and a set of zero or
45 * more StoreFiles, which stretch backwards over time.
46 */
47 @InterfaceAudience.Private
48 @InterfaceStability.Evolving
49 public interface Store extends HeapSize, StoreConfigInformation {
50
51 /* The default priority for user-specified compaction requests.
52 * The user gets top priority unless we have blocking compactions. (Pri <= 0)
53 */
54 public static final int PRIORITY_USER = 1;
55 public static final int NO_PRIORITY = Integer.MIN_VALUE;
56
57 // General Accessors
58 public KeyValue.KVComparator getComparator();
59
60 public Collection<StoreFile> getStorefiles();
61
62 /**
63 * Close all the readers We don't need to worry about subsequent requests because the HRegion
64 * holds a write lock that will prevent any more reads or writes.
65 * @return the {@link StoreFile StoreFiles} that were previously being used.
66 * @throws IOException on failure
67 */
68 public Collection<StoreFile> close() throws IOException;
69
70 /**
71 * Return a scanner for both the memstore and the HStore files. Assumes we are not in a
72 * compaction.
73 * @param scan Scan to apply when scanning the stores
74 * @param targetCols columns to scan
75 * @return a scanner over the current key values
76 * @throws IOException on failure
77 */
78 public KeyValueScanner getScanner(Scan scan, final NavigableSet<byte[]> targetCols)
79 throws IOException;
80
81 /**
82 * Get all scanners with no filtering based on TTL (that happens further down
83 * the line).
84 * @param cacheBlocks
85 * @param isGet
86 * @param isCompaction
87 * @param matcher
88 * @param startRow
89 * @param stopRow
90 * @return all scanners for this store
91 */
92 public List<KeyValueScanner> getScanners(boolean cacheBlocks,
93 boolean isGet, boolean isCompaction, ScanQueryMatcher matcher, byte[] startRow,
94 byte[] stopRow) throws IOException;
95
96 public ScanInfo getScanInfo();
97
98 /**
99 * Adds or replaces the specified KeyValues.
100 * <p>
101 * For each KeyValue specified, if a cell with the same row, family, and qualifier exists in
102 * MemStore, it will be replaced. Otherwise, it will just be inserted to MemStore.
103 * <p>
104 * This operation is atomic on each KeyValue (row/family/qualifier) but not necessarily atomic
105 * across all of them.
106 * @param cells
107 * @param readpoint readpoint below which we can safely remove duplicate KVs
108 * @return memstore size delta
109 * @throws IOException
110 */
111 public long upsert(Iterable<? extends Cell> cells, long readpoint) throws IOException;
112
113 /**
114 * Adds a value to the memstore
115 * @param kv
116 * @return memstore size delta
117 */
118 public long add(KeyValue kv);
119
120 /**
121 * When was the last edit done in the memstore
122 */
123 long timeOfOldestEdit();
124
125 /**
126 * Removes a kv from the memstore. The KeyValue is removed only if its key & memstoreTS match the
127 * key & memstoreTS value of the kv parameter.
128 * @param kv
129 */
130 public void rollback(final KeyValue kv);
131
132 /**
133 * Find the key that matches <i>row</i> exactly, or the one that immediately precedes it. WARNING:
134 * Only use this method on a table where writes occur with strictly increasing timestamps. This
135 * method assumes this pattern of writes in order to make it reasonably performant. Also our
136 * search is dependent on the axiom that deletes are for cells that are in the container that
137 * follows whether a memstore snapshot or a storefile, not for the current container: i.e. we'll
138 * see deletes before we come across cells we are to delete. Presumption is that the
139 * memstore#kvset is processed before memstore#snapshot and so on.
140 * @param row The row key of the targeted row.
141 * @return Found keyvalue or null if none found.
142 * @throws IOException
143 */
144 public KeyValue getRowKeyAtOrBefore(final byte[] row) throws IOException;
145
146 public FileSystem getFileSystem();
147
148 /*
149 * @param maxKeyCount
150 * @param compression Compression algorithm to use
151 * @param isCompaction whether we are creating a new file in a compaction
152 * @param includeMVCCReadpoint whether we should out the MVCC readpoint
153 * @return Writer for a new StoreFile in the tmp dir.
154 */
155 public StoreFile.Writer createWriterInTmp(long maxKeyCount, Compression.Algorithm compression,
156 boolean isCompaction, boolean includeMVCCReadpoint) throws IOException;
157
158 // Compaction oriented methods
159
160 public boolean throttleCompaction(long compactionSize);
161
162 /**
163 * getter for CompactionProgress object
164 * @return CompactionProgress object; can be null
165 */
166 public CompactionProgress getCompactionProgress();
167
168 public CompactionContext requestCompaction() throws IOException;
169
170 public CompactionContext requestCompaction(int priority, CompactionRequest baseRequest)
171 throws IOException;
172
173 public void cancelRequestedCompaction(CompactionContext compaction);
174
175 public List<StoreFile> compact(CompactionContext compaction) throws IOException;
176
177 /**
178 * @return true if we should run a major compaction.
179 */
180 public boolean isMajorCompaction() throws IOException;
181
182 public void triggerMajorCompaction();
183
184 /**
185 * See if there's too much store files in this store
186 * @return true if number of store files is greater than the number defined in minFilesToCompact
187 */
188 public boolean needsCompaction();
189
190 public int getCompactPriority();
191
192 public StoreFlushContext createFlushContext(long cacheFlushId);
193
194 /**
195 * Call to complete a compaction. Its for the case where we find in the WAL a compaction
196 * that was not finished. We could find one recovering a WAL after a regionserver crash.
197 * See HBASE-2331.
198 * @param compaction
199 */
200 public void completeCompactionMarker(CompactionDescriptor compaction)
201 throws IOException;
202
203 // Split oriented methods
204
205 public boolean canSplit();
206
207 /**
208 * Determines if Store should be split
209 * @return byte[] if store should be split, null otherwise.
210 */
211 public byte[] getSplitPoint();
212
213 // Bulk Load methods
214
215 /**
216 * This throws a WrongRegionException if the HFile does not fit in this region, or an
217 * InvalidHFileException if the HFile is not valid.
218 */
219 public void assertBulkLoadHFileOk(Path srcPath) throws IOException;
220
221 /**
222 * This method should only be called from HRegion. It is assumed that the ranges of values in the
223 * HFile fit within the stores assigned region. (assertBulkLoadHFileOk checks this)
224 *
225 * @param srcPathStr
226 * @param sequenceId sequence Id associated with the HFile
227 */
228 public void bulkLoadHFile(String srcPathStr, long sequenceId) throws IOException;
229
230 // General accessors into the state of the store
231 // TODO abstract some of this out into a metrics class
232
233 /**
234 * @return <tt>true</tt> if the store has any underlying reference files to older HFiles
235 */
236 public boolean hasReferences();
237
238 /**
239 * @return The size of this store's memstore, in bytes
240 */
241 public long getMemStoreSize();
242
243 public HColumnDescriptor getFamily();
244
245 /**
246 * @return The maximum memstoreTS in all store files.
247 */
248 public long getMaxMemstoreTS();
249
250 /**
251 * @return the data block encoder
252 */
253 public HFileDataBlockEncoder getDataBlockEncoder();
254
255 /** @return aggregate size of all HStores used in the last compaction */
256 public long getLastCompactSize();
257
258 /** @return aggregate size of HStore */
259 public long getSize();
260
261 /**
262 * @return Count of store files
263 */
264 public int getStorefilesCount();
265
266 /**
267 * @return The size of the store files, in bytes, uncompressed.
268 */
269 public long getStoreSizeUncompressed();
270
271 /**
272 * @return The size of the store files, in bytes.
273 */
274 public long getStorefilesSize();
275
276 /**
277 * @return The size of the store file indexes, in bytes.
278 */
279 public long getStorefilesIndexSize();
280
281 /**
282 * Returns the total size of all index blocks in the data block indexes, including the root level,
283 * intermediate levels, and the leaf level for multi-level indexes, or just the root level for
284 * single-level indexes.
285 * @return the total size of block indexes in the store
286 */
287 public long getTotalStaticIndexSize();
288
289 /**
290 * Returns the total byte size of all Bloom filter bit arrays. For compound Bloom filters even the
291 * Bloom blocks currently not loaded into the block cache are counted.
292 * @return the total size of all Bloom filters in the store
293 */
294 public long getTotalStaticBloomSize();
295
296 // Test-helper methods
297
298 /**
299 * Used for tests.
300 * @return cache configuration for this Store.
301 */
302 public CacheConfig getCacheConfig();
303
304 /**
305 * @return the parent region info hosting this store
306 */
307 public HRegionInfo getRegionInfo();
308
309 public RegionCoprocessorHost getCoprocessorHost();
310
311 public boolean areWritesEnabled();
312
313 /**
314 * @return The smallest mvcc readPoint across all the scanners in this
315 * region. Writes older than this readPoint, are included in every
316 * read operation.
317 */
318 public long getSmallestReadPoint();
319
320 public String getColumnFamilyName();
321
322 public String getTableName();
323
324 /*
325 * @param o Observer who wants to know about changes in set of Readers
326 */
327 public void addChangedReaderObserver(ChangedReadersObserver o);
328
329 /*
330 * @param o Observer no longer interested in changes in set of Readers.
331 */
332 public void deleteChangedReaderObserver(ChangedReadersObserver o);
333
334 /**
335 * @return Whether this store has too many store files.
336 */
337 public boolean hasTooManyStoreFiles();
338 }