Package com.netflix.aegisthus.columnar_input.splits

Examples of com.netflix.aegisthus.columnar_input.splits.AegSplit

52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
public class AegisthusInputFormat extends FileInputFormat<CompositeKey, AtomWritable> {
    private static final Logger LOG = LoggerFactory.getLogger(AegisthusInputFormat.class);

    @Override
    public RecordReader<CompositeKey, AtomWritable> createRecordReader(InputSplit inputSplit, TaskAttemptContext context) {
        AegSplit split = null;
        if (inputSplit instanceof AegCombinedSplit) {
            return new CombineSSTableReader();
        }
        split = (AegSplit) inputSplit;
        RecordReader<CompositeKey, AtomWritable> reader = null;
        switch (split.getType()) {
        case sstable:
            reader = new SSTableRecordReader();
            break;
        case commitlog:
            reader = new CommitLogRecordReader();
View Full Code Here
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
                        newIndexOffset = pair.right;

                    }
                    int blkIndex = getBlockIndex(blkLocations, splitStart + (splitSize / 2));
                    LOG.debug("split path: {}:{}:{}", path.getName(), splitStart, splitSize);
                    splits.add(new AegSplit(path, splitStart, splitSize, blkLocations[blkIndex].getHosts()));
                    indexOffset = newIndexOffset;
                    bytesRemaining -= splitSize;
                    splitStart += splitSize;
                }
                if (scanner != null) {
                    scanner.close();
                }
            }

            if (bytesRemaining != 0) {
                LOG.debug("end path: {}:{}:{}", path.getName(), length - bytesRemaining, bytesRemaining);
                if (fs.exists(compressionPath)) {
                    splits.add(new AegCompressedSplit(path, length - bytesRemaining, bytesRemaining,
                            blkLocations[blkLocations.length - 1].getHosts(), compressionPath));
                } else {
                    splits.add(new AegSplit(path, length - bytesRemaining, bytesRemaining,
                            blkLocations[blkLocations.length - 1].getHosts()));
                }
            }
        } else {
            LOG.info("skipping zero length file: {}", path.toString());
View Full Code Here
154
155
156
157
158
159
160
161
162
163
            } else if (name.startsWith("CommitLog")) {
                LOG.info(String.format("adding %s as a CommitLog split", file.getPath().toUri().toString()));
                BlockLocation[] blkLocations = file.getPath()
                        .getFileSystem(job.getConfiguration())
                        .getFileBlockLocations(file, 0, file.getLen());
                splits.add(new AegSplit(file.getPath(), 0, file.getLen(), blkLocations[0].getHosts(), Type.commitlog));
            }
        }
        return splits;
    }
View Full Code Here
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
    @SuppressWarnings("rawtypes")
    @Override
    public void initialize(InputSplit inputSplit, final TaskAttemptContext ctx) throws IOException,
            InterruptedException {
        AegSplit split = (AegSplit) inputSplit;

        start = split.getStart();
        end = split.getEnd();
        final Path file = split.getPath();

        try {
            cfId = ctx.getConfiguration().getInt("commitlog.cfid", -1000);
            if (cfId == -1000) {
                throw new IOException("commitlog.cfid must be set");
            }
            // open the file and seek to the start of the split
            FileSystem fs = file.getFileSystem(ctx.getConfiguration());
            FSDataInputStream fileIn = fs.open(split.getPath());
            InputStream dis = new BufferedInputStream(fileIn);
            scanner = new CommitLogColumnarScanner(new DataInputStream(dis),
                    Descriptor.fromFilename(split.getPath().getName()).version, cfId);
            this.pos = start;
            iterator = scanner.observable()
                    .onErrorFlatMap(new Func1<OnErrorThrowable, Observable<? extends AtomWritable>>() {
                        @Override
                        public Observable<? extends AtomWritable> call(OnErrorThrowable onErrorThrowable) {
View Full Code Here
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
    }

    @Override
    public void initialize(InputSplit inputSplit, final TaskAttemptContext ctx) throws IOException,
            InterruptedException {
        AegSplit split = (AegSplit) inputSplit;

        start = split.getStart();
        InputStream is = split.getInput(ctx.getConfiguration());
        end = split.getDataEnd();
        String filename = split.getPath().toUri().toString();

        LOG.info(String.format("File: %s", split.getPath().toUri().getPath()));
        LOG.info("Start: " + start);
        LOG.info("End: " + end);

        try {
            DataInput indexInput;
View Full Code Here
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
        int cnt = 0;
        int cntAdded = 0;

        top:
        for (InputSplit split : splits) {
            AegSplit aegSplit = (AegSplit) split;
            switch (aegSplit.getType()) {
            case sstable:
                cnt++;
                if (aegSplit.getLength() >= MAX_SPLIT_SIZE) {
                    cntAdded++;
                    combinedSplits.add(aegSplit);
                    continue;
                }
                try {
                    String lastLocation = null;
                    for (String location : aegSplit.getLocations()) {
                        lastLocation = location;
                        if (map.containsKey(location)) {
                            AegCombinedSplit temp = map.get(location);
                            cntAdded++;
                            temp.getSplits().add(aegSplit);
View Full Code Here
TOP

Related Classes of com.netflix.aegisthus.columnar_input.splits.AegSplit

  • com.netflix.aegisthus.columnar_input.AegisthusCombinedInputFormat
  • com.netflix.aegisthus.columnar_input.AegisthusInputFormat
  • com.netflix.aegisthus.columnar_input.readers.CommitLogRecordReader
  • com.netflix.aegisthus.columnar_input.readers.SSTableRecordReader
  • org.apache.hadoop.fs.FileSystem
  • org.apache.hadoop.fs.FSDataInputStream
  • org.apache.hadoop.fs.Path
  • java.io.BufferedInputStream
  • java.io.DataInputStream
    • 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.