View Javadoc

1   /*
2    * Licensed to the Apache Software Foundation (ASF) under one or more
3    * contributor license agreements. See the NOTICE file distributed with this
4    * work for additional information regarding copyright ownership. The ASF
5    * licenses this file to you under the Apache License, Version 2.0 (the
6    * "License"); you may not use this file except in compliance with the License.
7    * You may obtain a copy of the License at
8    *
9    * http://www.apache.org/licenses/LICENSE-2.0
10   *
11   * Unless required by applicable law or agreed to in writing, software
12   * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
13   * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
14   * License for the specific language governing permissions and limitations
15   * under the License.
16   */
17  package org.apache.hadoop.hbase.util.test;
18  
19  import java.util.Random;
20  
21  import org.apache.hadoop.hbase.classification.InterfaceAudience;
22  import org.apache.hadoop.hbase.util.Bytes;
23  import org.apache.hadoop.hbase.util.MD5Hash;
24  
25  /**
26   * A generator of random keys and values for load testing. Keys are generated
27   * by converting numeric indexes to strings and prefixing them with an MD5
28   * hash. Values are generated by selecting value size in the configured range
29   * and generating a pseudo-random sequence of bytes seeded by key, column
30   * qualifier, and value size.
31   */
32  @InterfaceAudience.Private
33  public class LoadTestKVGenerator {
34  
35    /** A random number generator for determining value size */
36    private Random randomForValueSize = new Random();
37  
38    private final int minValueSize;
39    private final int maxValueSize;
40  
41    public LoadTestKVGenerator(int minValueSize, int maxValueSize) {
42      if (minValueSize <= 0 || maxValueSize <= 0) {
43        throw new IllegalArgumentException("Invalid min/max value sizes: " +
44            minValueSize + ", " + maxValueSize);
45      }
46      this.minValueSize = minValueSize;
47      this.maxValueSize = maxValueSize;
48    }
49  
50    /**
51     * Verifies that the given byte array is the same as what would be generated
52     * for the given seed strings (row/cf/column/...). We are assuming that the
53     * value size is correct, and only verify the actual bytes. However, if the
54     * min/max value sizes are set sufficiently high, an accidental match should be
55     * extremely improbable.
56     */
57    public static boolean verify(byte[] value, byte[]... seedStrings) {
58      byte[] expectedData = getValueForRowColumn(value.length, seedStrings);
59      return Bytes.equals(expectedData, value);
60    }
61  
62    /**
63     * Converts the given key to string, and prefixes it with the MD5 hash of
64     * the index's string representation.
65     */
66    public static String md5PrefixedKey(long key) {
67      String stringKey = Long.toString(key);
68      String md5hash = MD5Hash.getMD5AsHex(Bytes.toBytes(stringKey));
69  
70      // flip the key to randomize
71      return md5hash + "-" + stringKey;
72    }
73  
74    /**
75     * Generates a value for the given key index and column qualifier. Size is
76     * selected randomly in the configured range. The generated value depends
77     * only on the combination of the strings passed (key/cf/column/...) and the selected
78     * value size. This allows to verify the actual value bytes when reading, as done
79     * in {#verify(byte[], byte[]...)}
80     * This method is as thread-safe as Random class. It appears that the worst bug ever
81     * found with the latter is that multiple threads will get some duplicate values, which
82     * we don't care about.
83     */
84    public byte[] generateRandomSizeValue(byte[]... seedStrings) {
85      int dataSize = minValueSize;
86      if(minValueSize != maxValueSize) {
87        dataSize = minValueSize + randomForValueSize.nextInt(Math.abs(maxValueSize - minValueSize));
88      }
89      return getValueForRowColumn(dataSize, seedStrings);
90    }
91  
92    /**
93     * Generates random bytes of the given size for the given row and column
94     * qualifier. The random seed is fully determined by these parameters.
95     */
96    private static byte[] getValueForRowColumn(int dataSize, byte[]... seedStrings) {
97      long seed = dataSize;
98      for (byte[] str : seedStrings) {
99        seed += Bytes.toString(str).hashCode();
100     }
101     Random seededRandom = new Random(seed);
102     byte[] randomBytes = new byte[dataSize];
103     seededRandom.nextBytes(randomBytes);
104     return randomBytes;
105   }
106 
107 }