Apache
/
hbase
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

HBASE-11586 HFile's HDFS op latency sampling code is not used

This commit is contained in:
Andrew Purtell 2014-07-24 17:07:34 -07:00
parent 7b5a309697
commit 531eee0031
4 changed files with 0 additions and 923 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

89
hbase-server/src/main/java/org/apache/hadoop/hbase/io/hfile/HFile.java
View File

@ -182,101 +182,12 @@ public class HFile {
public static final int DEFAULT_BYTES_PER_CHECKSUM = 16 * 1024;
public static final ChecksumType DEFAULT_CHECKSUM_TYPE = ChecksumType.CRC32;
// For measuring latency of "sequential" reads and writes
private static final AtomicInteger readOps = new AtomicInteger();
private static final AtomicLong readTimeNano = new AtomicLong();
private static final AtomicInteger writeOps = new AtomicInteger();
private static final AtomicLong writeTimeNano = new AtomicLong();
// For measuring latency of pread
private static final AtomicInteger preadOps = new AtomicInteger();
private static final AtomicLong preadTimeNano = new AtomicLong();
// For measuring number of checksum failures
static final AtomicLong checksumFailures = new AtomicLong();
// For getting more detailed stats on FS latencies
// If, for some reason, the metrics subsystem stops polling for latencies,
// I don't want data to pile up in a memory leak
// so, after LATENCY_BUFFER_SIZE items have been enqueued for processing,
// fs latency stats will be dropped (and this behavior will be logged)
private static final int LATENCY_BUFFER_SIZE = 5000;
private static final BlockingQueue<Long> fsReadLatenciesNanos =
new ArrayBlockingQueue<Long>(LATENCY_BUFFER_SIZE);
private static final BlockingQueue<Long> fsWriteLatenciesNanos =
new ArrayBlockingQueue<Long>(LATENCY_BUFFER_SIZE);
private static final BlockingQueue<Long> fsPreadLatenciesNanos =
new ArrayBlockingQueue<Long>(LATENCY_BUFFER_SIZE);
public static final void offerReadLatency(long latencyNanos, boolean pread) {
if (pread) {
fsPreadLatenciesNanos.offer(latencyNanos); // might be silently dropped, if the queue is full
preadOps.incrementAndGet();
preadTimeNano.addAndGet(latencyNanos);
} else {
fsReadLatenciesNanos.offer(latencyNanos); // might be silently dropped, if the queue is full
readTimeNano.addAndGet(latencyNanos);
readOps.incrementAndGet();
}
}
public static final void offerWriteLatency(long latencyNanos) {
fsWriteLatenciesNanos.offer(latencyNanos); // might be silently dropped, if the queue is full
writeTimeNano.addAndGet(latencyNanos);
writeOps.incrementAndGet();
}
public static final Collection<Long> getReadLatenciesNanos() {
final List<Long> latencies =
Lists.newArrayListWithCapacity(fsReadLatenciesNanos.size());
fsReadLatenciesNanos.drainTo(latencies);
return latencies;
}
public static final Collection<Long> getPreadLatenciesNanos() {
final List<Long> latencies =
Lists.newArrayListWithCapacity(fsPreadLatenciesNanos.size());
fsPreadLatenciesNanos.drainTo(latencies);
return latencies;
}
public static final Collection<Long> getWriteLatenciesNanos() {
final List<Long> latencies =
Lists.newArrayListWithCapacity(fsWriteLatenciesNanos.size());
fsWriteLatenciesNanos.drainTo(latencies);
return latencies;
}
// for test purpose
public static final AtomicLong dataBlockReadCnt = new AtomicLong(0);
// number of sequential reads
public static final int getReadOps() {
return readOps.getAndSet(0);
}
public static final long getReadTimeMs() {
return readTimeNano.getAndSet(0) / 1000000;
}
// number of positional reads
public static final int getPreadOps() {
return preadOps.getAndSet(0);
}
public static final long getPreadTimeMs() {
return preadTimeNano.getAndSet(0) / 1000000;
}
public static final int getWriteOps() {
return writeOps.getAndSet(0);
}
public static final long getWriteTimeMs() {
return writeTimeNano.getAndSet(0) / 1000000;
}
/**
* Number of checksum verification failures. It also
* clears the counter.

8
hbase-server/src/main/java/org/apache/hadoop/hbase/io/hfile/HFileReaderV2.java
View File

@ -323,7 +323,6 @@ public class HFileReaderV2 extends AbstractHFileReader {
if (block == -1)
return null;
long blockSize = metaBlockIndexReader.getRootBlockDataSize(block);
long startTimeNs = System.nanoTime();
// Per meta key from any given file, synchronize reads for said block. This
// is OK to do for meta blocks because the meta block index is always
@ -348,9 +347,6 @@ public class HFileReaderV2 extends AbstractHFileReader {
HFileBlock metaBlock = fsBlockReader.readBlockData(metaBlockOffset,
blockSize, -1, true);
final long delta = System.nanoTime() - startTimeNs;
HFile.offerReadLatency(delta, true);
// Cache the block
if (cacheBlock) {
cacheConf.getBlockCache().cacheBlock(cacheKey, metaBlock,
@ -449,14 +445,10 @@ public class HFileReaderV2 extends AbstractHFileReader {
traceScope.getSpan().addTimelineAnnotation("blockCacheMiss");
}
// Load block from filesystem.
long startTimeNs = System.nanoTime();
HFileBlock hfileBlock = fsBlockReader.readBlockData(dataBlockOffset, onDiskBlockSize, -1,
pread);
validateBlockType(hfileBlock, expectedBlockType);
final long delta = System.nanoTime() - startTimeNs;
HFile.offerReadLatency(delta, pread);
// Cache the block if necessary
if (cacheBlock && cacheConf.shouldCacheBlockOnRead(hfileBlock.getBlockType().getCategory())) {
cacheConf.getBlockCache().cacheBlock(cacheKey, hfileBlock, cacheConf.isInMemory(),

2
hbase-server/src/main/java/org/apache/hadoop/hbase/io/hfile/HFileWriterV2.java
View File

@ -148,7 +148,6 @@ public class HFileWriterV2 extends AbstractHFileWriter {
if (!fsBlockWriter.isWriting() || fsBlockWriter.blockSizeWritten() == 0)
return;
long startTimeNs = System.nanoTime();
// Update the first data block offset for scanning.
if (firstDataBlockOffset == -1) {
firstDataBlockOffset = outputStream.getPos();
@ -161,7 +160,6 @@ public class HFileWriterV2 extends AbstractHFileWriter {
byte[] indexKey = comparator.calcIndexKey(lastKeyOfPreviousBlock, firstKeyInBlock);
dataBlockIndexWriter.addEntry(indexKey, lastDataBlockOffset, onDiskSize);
totalUncompressedBytes += fsBlockWriter.getUncompressedSizeWithHeader();
HFile.offerWriteLatency(System.nanoTime() - startTimeNs);
if (cacheConf.shouldCacheDataOnWrite()) {
doCacheOnWrite(lastDataBlockOffset);
}

824
hbase-server/src/test/java/org/apache/hadoop/hbase/regionserver/HFileReadWriteTest.java
View File

@ -1,824 +0,0 @@
/*
* 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.hadoop.hbase.regionserver;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collections;
import java.util.List;
import java.util.Random;
import java.util.SortedSet;
import java.util.concurrent.Callable;
import java.util.concurrent.ConcurrentSkipListSet;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.ExecutorCompletionService;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.atomic.AtomicLong;
import org.apache.commons.cli.CommandLine;
import org.apache.commons.cli.CommandLineParser;
import org.apache.commons.cli.HelpFormatter;
import org.apache.commons.cli.Option;
import org.apache.commons.cli.OptionGroup;
import org.apache.commons.cli.Options;
import org.apache.commons.cli.ParseException;
import org.apache.commons.cli.PosixParser;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.hbase.Cell;
import org.apache.hadoop.hbase.HBaseConfiguration;
import org.apache.hadoop.hbase.HColumnDescriptor;
import org.apache.hadoop.hbase.HRegionInfo;
import org.apache.hadoop.hbase.HTableDescriptor;
import org.apache.hadoop.hbase.KeyValue;
import org.apache.hadoop.hbase.KeyValueUtil;
import org.apache.hadoop.hbase.TableName;
import org.apache.hadoop.hbase.client.IsolationLevel;
import org.apache.hadoop.hbase.client.Scan;
import org.apache.hadoop.hbase.io.compress.Compression;
import org.apache.hadoop.hbase.io.encoding.DataBlockEncoding;
import org.apache.hadoop.hbase.io.hfile.BlockCache;
import org.apache.hadoop.hbase.io.hfile.CacheConfig;
import org.apache.hadoop.hbase.io.hfile.HFile;
import org.apache.hadoop.hbase.io.hfile.HFilePrettyPrinter;
import org.apache.hadoop.hbase.util.Bytes;
import org.apache.hadoop.hbase.util.LoadTestTool;
import org.apache.hadoop.hbase.util.MD5Hash;
import org.apache.hadoop.util.StringUtils;
/**
* Tests HFile read/write workloads, such as merging HFiles and random reads.
*/
public class HFileReadWriteTest {
private static final String TABLE_NAME = "MyTable";
private static enum Workload {
MERGE("merge", "Merge the specified HFiles", 1, Integer.MAX_VALUE),
RANDOM_READS("read", "Perform a random read benchmark on the given HFile",
1, 1);
private String option;
private String description;
public final int minNumInputFiles;
public final int maxNumInputFiles;
Workload(String option, String description, int minNumInputFiles,
int maxNumInputFiles) {
this.option = option;
this.description = description;
this.minNumInputFiles = minNumInputFiles;
this.maxNumInputFiles = maxNumInputFiles;
}
static OptionGroup getOptionGroup() {
OptionGroup optionGroup = new OptionGroup();
for (Workload w : values())
optionGroup.addOption(new Option(w.option, w.description));
return optionGroup;
}
private static String getOptionListStr() {
StringBuilder sb = new StringBuilder();
for (Workload w : values()) {
if (sb.length() > 0)
sb.append(", ");
sb.append("-" + w.option);
}
return sb.toString();
}
static Workload fromCmdLine(CommandLine cmdLine) {
for (Workload w : values()) {
if (cmdLine.hasOption(w.option))
return w;
}
LOG.error("No workload specified. Specify one of the options: " +
getOptionListStr());
return null;
}
public String onlyUsedFor() {
return ". Only used for the " + this + " workload.";
}
}
private static final String OUTPUT_DIR_OPTION = "output_dir";
private static final String COMPRESSION_OPTION = "compression";
private static final String BLOOM_FILTER_OPTION = "bloom";
private static final String BLOCK_SIZE_OPTION = "block_size";
private static final String DURATION_OPTION = "duration";
private static final String NUM_THREADS_OPTION = "num_threads";
private static final Log LOG = LogFactory.getLog(HFileReadWriteTest.class);
private Workload workload;
private FileSystem fs;
private Configuration conf;
private CacheConfig cacheConf;
private List<String> inputFileNames;
private Path outputDir;
private int numReadThreads;
private int durationSec;
private DataBlockEncoding dataBlockEncoding;
private BloomType bloomType = BloomType.NONE;
private int blockSize;
private Compression.Algorithm compression = Compression.Algorithm.NONE;
private byte[] firstRow, lastRow;
private AtomicLong numSeeks = new AtomicLong();
private AtomicLong numKV = new AtomicLong();
private AtomicLong totalBytes = new AtomicLong();
private byte[] family;
private long endTime = Long.MAX_VALUE;
private SortedSet<String> keysRead = new ConcurrentSkipListSet<String>();
private List<StoreFile> inputStoreFiles;
public HFileReadWriteTest() {
conf = HBaseConfiguration.create();
cacheConf = new CacheConfig(conf);
}
@SuppressWarnings("unchecked")
public boolean parseOptions(String args[]) {
Options options = new Options();
options.addOption(OUTPUT_DIR_OPTION, true, "Output directory" +
Workload.MERGE.onlyUsedFor());
options.addOption(COMPRESSION_OPTION, true, " Compression type, one of "
+ Arrays.toString(Compression.Algorithm.values()) +
Workload.MERGE.onlyUsedFor());
options.addOption(BLOOM_FILTER_OPTION, true, "Bloom filter type, one of "
+ Arrays.toString(BloomType.values()) +
Workload.MERGE.onlyUsedFor());
options.addOption(BLOCK_SIZE_OPTION, true, "HFile block size" +
Workload.MERGE.onlyUsedFor());
options.addOption(DURATION_OPTION, true, "The amount of time to run the " +
"random read workload for" + Workload.RANDOM_READS.onlyUsedFor());
options.addOption(NUM_THREADS_OPTION, true, "The number of random " +
"reader threads" + Workload.RANDOM_READS.onlyUsedFor());
options.addOption(NUM_THREADS_OPTION, true, "The number of random " +
"reader threads" + Workload.RANDOM_READS.onlyUsedFor());
options.addOption(LoadTestTool.OPT_DATA_BLOCK_ENCODING, true,
LoadTestTool.OPT_DATA_BLOCK_ENCODING_USAGE);
options.addOptionGroup(Workload.getOptionGroup());
if (args.length == 0) {
HelpFormatter formatter = new HelpFormatter();
formatter.printHelp(HFileReadWriteTest.class.getSimpleName(),
options, true);
return false;
}
CommandLineParser parser = new PosixParser();
CommandLine cmdLine;
try {
cmdLine = parser.parse(options, args);
} catch (ParseException ex) {
LOG.error(ex);
return false;
}
workload = Workload.fromCmdLine(cmdLine);
if (workload == null)
return false;
inputFileNames = (List<String>) cmdLine.getArgList();
if (inputFileNames.size() == 0) {
LOG.error("No input file names specified");
return false;
}
if (inputFileNames.size() < workload.minNumInputFiles) {
LOG.error("Too few input files: at least " + workload.minNumInputFiles +
" required");
return false;
}
if (inputFileNames.size() > workload.maxNumInputFiles) {
LOG.error("Too many input files: at most " + workload.minNumInputFiles +
" allowed");
return false;
}
if (cmdLine.hasOption(COMPRESSION_OPTION)) {
compression = Compression.Algorithm.valueOf(
cmdLine.getOptionValue(COMPRESSION_OPTION));
}
if (cmdLine.hasOption(BLOOM_FILTER_OPTION)) {
bloomType = BloomType.valueOf(cmdLine.getOptionValue(
BLOOM_FILTER_OPTION));
}
if (cmdLine.hasOption(LoadTestTool.OPT_DATA_BLOCK_ENCODING)) {
dataBlockEncoding = DataBlockEncoding.valueOf(
cmdLine.getOptionValue(LoadTestTool.OPT_DATA_BLOCK_ENCODING));
}
blockSize = conf.getInt("hfile.min.blocksize.size", 65536);
if (cmdLine.hasOption(BLOCK_SIZE_OPTION))
blockSize = Integer.valueOf(cmdLine.getOptionValue(BLOCK_SIZE_OPTION));
if (workload == Workload.MERGE) {
String outputDirStr = cmdLine.getOptionValue(OUTPUT_DIR_OPTION);
if (outputDirStr == null) {
LOG.error("Output directory is not specified");
return false;
}
outputDir = new Path(outputDirStr);
// Will be checked for existence in validateConfiguration.
}
if (workload == Workload.RANDOM_READS) {
if (!requireOptions(cmdLine, new String[] { DURATION_OPTION,
NUM_THREADS_OPTION })) {
return false;
}
durationSec = Integer.parseInt(cmdLine.getOptionValue(DURATION_OPTION));
numReadThreads = Integer.parseInt(
cmdLine.getOptionValue(NUM_THREADS_OPTION));
}
Collections.sort(inputFileNames);
return true;
}
/** @return true if all the given options are specified */
private boolean requireOptions(CommandLine cmdLine,
String[] requiredOptions) {
for (String option : requiredOptions)
if (!cmdLine.hasOption(option)) {
LOG.error("Required option -" + option + " not specified");
return false;
}
return true;
}
public boolean validateConfiguration() throws IOException {
fs = FileSystem.get(conf);
for (String inputFileName : inputFileNames) {
Path path = new Path(inputFileName);
if (!fs.exists(path)) {
LOG.error("File " + inputFileName + " does not exist");
return false;
}
if (fs.getFileStatus(path).isDirectory()) {
LOG.error(inputFileName + " is a directory");
return false;
}
}
if (outputDir != null &&
(!fs.exists(outputDir) || !fs.getFileStatus(outputDir).isDirectory())) {
LOG.error(outputDir.toString() + " does not exist or is not a " +
"directory");
return false;
}
return true;
}
public void runMergeWorkload() throws IOException {
long maxKeyCount = prepareForMerge();
HColumnDescriptor columnDescriptor = new HColumnDescriptor(
HFileReadWriteTest.class.getSimpleName());
columnDescriptor.setBlocksize(blockSize);
columnDescriptor.setBloomFilterType(bloomType);
columnDescriptor.setCompressionType(compression);
columnDescriptor.setDataBlockEncoding(dataBlockEncoding);
HRegionInfo regionInfo = new HRegionInfo();
HTableDescriptor htd = new HTableDescriptor(TableName.valueOf(TABLE_NAME));
HRegion region = new HRegion(outputDir, null, fs, conf, regionInfo, htd, null);
HStore store = new HStore(region, columnDescriptor, conf);
List<StoreFileScanner> scanners =
StoreFileScanner.getScannersForStoreFiles(inputStoreFiles, false,
false, region.getReadpoint(IsolationLevel.READ_COMMITTED));
StoreFile.Writer writer = store.createWriterInTmp(maxKeyCount, compression, false, true, false);
StatisticsPrinter statsPrinter = new StatisticsPrinter();
statsPrinter.startThread();
try {
performMerge(scanners, store, writer);
writer.close();
} finally {
statsPrinter.requestStop();
}
Path resultPath = writer.getPath();
resultPath = tryUsingSimpleOutputPath(resultPath);
long fileSize = fs.getFileStatus(resultPath).getLen();
LOG.info("Created " + resultPath + ", size " + fileSize);
System.out.println();
System.out.println("HFile information for " + resultPath);
System.out.println();
HFilePrettyPrinter hfpp = new HFilePrettyPrinter();
hfpp.run(new String[] { "-m", "-f", resultPath.toString() });
}
private Path tryUsingSimpleOutputPath(Path resultPath) throws IOException {
if (inputFileNames.size() == 1) {
// In case of only one input set output to be consistent with the
// input name.
Path inputPath = new Path(inputFileNames.get(0));
Path betterOutputPath = new Path(outputDir,
inputPath.getName());
if (!fs.exists(betterOutputPath)) {
fs.rename(resultPath, betterOutputPath);
resultPath = betterOutputPath;
}
}
return resultPath;
}
private void performMerge(List<StoreFileScanner> scanners, HStore store,
StoreFile.Writer writer) throws IOException {
InternalScanner scanner = null;
try {
Scan scan = new Scan();
// Include deletes
scanner = new StoreScanner(store, store.getScanInfo(), scan, scanners,
ScanType.COMPACT_DROP_DELETES, Long.MIN_VALUE, Long.MIN_VALUE);
ArrayList<Cell> kvs = new ArrayList<Cell>();
while (scanner.next(kvs) || kvs.size() != 0) {
numKV.addAndGet(kvs.size());
for (Cell c : kvs) {
KeyValue kv = KeyValueUtil.ensureKeyValue(c);
totalBytes.addAndGet(kv.getLength());
writer.append(kv);
}
kvs.clear();
}
} finally {
if (scanner != null)
scanner.close();
}
}
/**
* @return the total key count in the files being merged
* @throws IOException
*/
private long prepareForMerge() throws IOException {
LOG.info("Merging " + inputFileNames);
LOG.info("Using block size: " + blockSize);
inputStoreFiles = new ArrayList<StoreFile>();
long maxKeyCount = 0;
for (String fileName : inputFileNames) {
Path filePath = new Path(fileName);
// Open without caching.
StoreFile sf = openStoreFile(filePath, false);
sf.createReader();
inputStoreFiles.add(sf);
StoreFile.Reader r = sf.getReader();
if (r != null) {
long keyCount = r.getFilterEntries();
maxKeyCount += keyCount;
LOG.info("Compacting: " + sf + "; keyCount = " + keyCount
+ "; Bloom Type = " + r.getBloomFilterType().toString()
+ "; Size = " + StringUtils.humanReadableInt(r.length()));
}
}
return maxKeyCount;
}
public HFile.Reader[] getHFileReaders() {
HFile.Reader readers[] = new HFile.Reader[inputStoreFiles.size()];
for (int i = 0; i < inputStoreFiles.size(); ++i)
readers[i] = inputStoreFiles.get(i).getReader().getHFileReader();
return readers;
}
private StoreFile openStoreFile(Path filePath, boolean blockCache)
throws IOException {
// We are passing the ROWCOL Bloom filter type, but StoreFile will still
// use the Bloom filter type specified in the HFile.
return new StoreFile(fs, filePath, conf, cacheConf,
BloomType.ROWCOL);
}
public static int charToHex(int c) {
if ('0' <= c && c <= '9')
return c - '0';
if ('a' <= c && c <= 'f')
return 10 + c - 'a';
return -1;
}
public static int hexToChar(int h) {
h &= 0xff;
if (0 <= h && h <= 9)
return '0' + h;
if (10 <= h && h <= 15)
return 'a' + h - 10;
return -1;
}
public static byte[] createRandomRow(Random rand, byte[] first, byte[] last)
{
int resultLen = Math.max(first.length, last.length);
int minLen = Math.min(first.length, last.length);
byte[] result = new byte[resultLen];
boolean greaterThanFirst = false;
boolean lessThanLast = false;
for (int i = 0; i < resultLen; ++i) {
// Generate random hex characters if both first and last row are hex
// at this position.
boolean isHex = i < minLen && charToHex(first[i]) != -1
&& charToHex(last[i]) != -1;
// If our key is already greater than the first key, we can use
// arbitrarily low values.
int low = greaterThanFirst || i >= first.length ? 0 : first[i] & 0xff;
// If our key is already less than the last key, we can use arbitrarily
// high values.
int high = lessThanLast || i >= last.length ? 0xff : last[i] & 0xff;
// Randomly select the next byte between the lowest and the highest
// value allowed for this position. Restrict to hex characters if
// necessary. We are generally biased towards border cases, which is OK
// for test.
int r;
if (isHex) {
// Use hex chars.
if (low < '0')
low = '0';
if (high > 'f')
high = 'f';
int lowHex = charToHex(low);
int highHex = charToHex(high);
r = hexToChar(lowHex + rand.nextInt(highHex - lowHex + 1));
} else {
r = low + rand.nextInt(high - low + 1);
}
if (r > low)
greaterThanFirst = true;
if (r < high)
lessThanLast = true;
result[i] = (byte) r;
}
if (Bytes.compareTo(result, first) < 0) {
throw new IllegalStateException("Generated key " +
Bytes.toStringBinary(result) + " is less than the first key " +
Bytes.toStringBinary(first));
}
if (Bytes.compareTo(result, last) > 0) {
throw new IllegalStateException("Generated key " +
Bytes.toStringBinary(result) + " is greater than te last key " +
Bytes.toStringBinary(last));
}
return result;
}
private static byte[] createRandomQualifier(Random rand) {
byte[] q = new byte[10 + rand.nextInt(30)];
rand.nextBytes(q);
return q;
}
private class RandomReader implements Callable<Boolean> {
private int readerId;
private StoreFile.Reader reader;
private boolean pread;
public RandomReader(int readerId, StoreFile.Reader reader,
boolean pread)
{
this.readerId = readerId;
this.reader = reader;
this.pread = pread;
}
@Override
public Boolean call() throws Exception {
Thread.currentThread().setName("reader " + readerId);
Random rand = new Random();
StoreFileScanner scanner = reader.getStoreFileScanner(true, pread);
while (System.currentTimeMillis() < endTime) {
byte[] row = createRandomRow(rand, firstRow, lastRow);
KeyValue kvToSeek = new KeyValue(row, family,
createRandomQualifier(rand));
if (rand.nextDouble() < 0.0001) {
LOG.info("kvToSeek=" + kvToSeek);
}
boolean seekResult;
try {
seekResult = scanner.seek(kvToSeek);
} catch (IOException ex) {
throw new IOException("Seek failed for key " + kvToSeek + ", pread="
+ pread, ex);
}
numSeeks.incrementAndGet();
if (!seekResult) {
error("Seek returned false for row " + Bytes.toStringBinary(row));
return false;
}
for (int i = 0; i < rand.nextInt(10) + 1; ++i) {
Cell kv = scanner.next();
numKV.incrementAndGet();
if (i == 0 && kv == null) {
error("scanner.next() returned null at the first iteration for " +
"row " + Bytes.toStringBinary(row));
return false;
}
if (kv == null)
break;
KeyValue keyv = KeyValueUtil.ensureKeyValue(kv);
String keyHashStr = MD5Hash.getMD5AsHex(keyv.getKey());
keysRead.add(keyHashStr);
totalBytes.addAndGet(keyv.getLength());
}
}
return true;
}
private void error(String msg) {
LOG.error("error in reader " + readerId + " (pread=" + pread + "): "
+ msg);
}
}
private class StatisticsPrinter implements Callable<Boolean> {
private volatile boolean stopRequested;
private volatile Thread thread;
private long totalSeekAndReads, totalPositionalReads;
/**
* Run the statistics collector in a separate thread without an executor.
*/
public void startThread() {
new Thread() {
@Override
public void run() {
try {
call();
} catch (Exception e) {
LOG.error(e);
}
}
}.start();
}
@Override
public Boolean call() throws Exception {
LOG.info("Starting statistics printer");
thread = Thread.currentThread();
thread.setName(StatisticsPrinter.class.getSimpleName());
long startTime = System.currentTimeMillis();
long curTime;
while ((curTime = System.currentTimeMillis()) < endTime &&
!stopRequested) {
long elapsedTime = curTime - startTime;
printStats(elapsedTime);
try {
Thread.sleep(1000 - elapsedTime % 1000);
} catch (InterruptedException iex) {
Thread.currentThread().interrupt();
if (stopRequested)
break;
}
}
printStats(curTime - startTime);
LOG.info("Stopping statistics printer");
return true;
}
private void printStats(long elapsedTime) {
long numSeeksL = numSeeks.get();
double timeSec = elapsedTime / 1000.0;
double seekPerSec = numSeeksL / timeSec;
long kvCount = numKV.get();
double kvPerSec = kvCount / timeSec;
long bytes = totalBytes.get();
double bytesPerSec = bytes / timeSec;
// readOps and preadOps counters get reset on access, so we have to
// accumulate them here. HRegion metrics publishing thread should not
// be running in this tool, so no one else should be resetting these
// metrics.
totalSeekAndReads += HFile.getReadOps();
totalPositionalReads += HFile.getPreadOps();
long totalBlocksRead = totalSeekAndReads + totalPositionalReads;
double blkReadPerSec = totalBlocksRead / timeSec;
double seekReadPerSec = totalSeekAndReads / timeSec;
double preadPerSec = totalPositionalReads / timeSec;
boolean isRead = workload == Workload.RANDOM_READS;
StringBuilder sb = new StringBuilder();
sb.append("Time: " + (long) timeSec + " sec");
if (isRead)
sb.append(", seek/sec: " + (long) seekPerSec);
sb.append(", kv/sec: " + (long) kvPerSec);
sb.append(", bytes/sec: " + (long) bytesPerSec);
sb.append(", blk/sec: " + (long) blkReadPerSec);
sb.append(", total KV: " + numKV);
sb.append(", total bytes: " + totalBytes);
sb.append(", total blk: " + totalBlocksRead);
sb.append(", seekRead/sec: " + (long) seekReadPerSec);
sb.append(", pread/sec: " + (long) preadPerSec);
if (isRead)
sb.append(", unique keys: " + (long) keysRead.size());
LOG.info(sb.toString());
}
public void requestStop() {
stopRequested = true;
if (thread != null)
thread.interrupt();
}
}
public boolean runRandomReadWorkload() throws IOException {
if (inputFileNames.size() != 1) {
throw new IOException("Need exactly one input file for random reads: " +
inputFileNames);
}
Path inputPath = new Path(inputFileNames.get(0));
// Make sure we are using caching.
StoreFile storeFile = openStoreFile(inputPath, true);
StoreFile.Reader reader = storeFile.createReader();
LOG.info("First key: " + Bytes.toStringBinary(reader.getFirstKey()));
LOG.info("Last key: " + Bytes.toStringBinary(reader.getLastKey()));
KeyValue firstKV = KeyValue.createKeyValueFromKey(reader.getFirstKey());
firstRow = firstKV.getRow();
KeyValue lastKV = KeyValue.createKeyValueFromKey(reader.getLastKey());
lastRow = lastKV.getRow();
byte[] family = firstKV.getFamily();
if (!Bytes.equals(family, lastKV.getFamily())) {
LOG.error("First and last key have different families: "
+ Bytes.toStringBinary(family) + " and "
+ Bytes.toStringBinary(lastKV.getFamily()));
return false;
}
if (Bytes.equals(firstRow, lastRow)) {
LOG.error("First and last row are the same, cannot run read workload: " +
"firstRow=" + Bytes.toStringBinary(firstRow) + ", " +
"lastRow=" + Bytes.toStringBinary(lastRow));
return false;
}
ExecutorService exec = Executors.newFixedThreadPool(numReadThreads + 1);
int numCompleted = 0;
int numFailed = 0;
try {
ExecutorCompletionService<Boolean> ecs =
new ExecutorCompletionService<Boolean>(exec);
endTime = System.currentTimeMillis() + 1000 * durationSec;
boolean pread = true;
for (int i = 0; i < numReadThreads; ++i)
ecs.submit(new RandomReader(i, reader, pread));
ecs.submit(new StatisticsPrinter());
Future<Boolean> result;
while (true) {
try {
result = ecs.poll(endTime + 1000 - System.currentTimeMillis(),
TimeUnit.MILLISECONDS);
if (result == null)
break;
try {
if (result.get()) {
++numCompleted;
} else {
++numFailed;
}
} catch (ExecutionException e) {
LOG.error("Worker thread failure", e.getCause());
++numFailed;
}
} catch (InterruptedException ex) {
LOG.error("Interrupted after " + numCompleted +
" workers completed");
Thread.currentThread().interrupt();
continue;
}
}
} finally {
storeFile.closeReader(true);
exec.shutdown();
BlockCache c = cacheConf.getBlockCache();
if (c != null) {
c.shutdown();
}
}
LOG.info("Worker threads completed: " + numCompleted);
LOG.info("Worker threads failed: " + numFailed);
return true;
}
public boolean run() throws IOException {
LOG.info("Workload: " + workload);
switch (workload) {
case MERGE:
runMergeWorkload();
break;
case RANDOM_READS:
return runRandomReadWorkload();
default:
LOG.error("Unknown workload: " + workload);
return false;
}
return true;
}
private static void failure() {
System.exit(1);
}
public static void main(String[] args) {
HFileReadWriteTest app = new HFileReadWriteTest();
if (!app.parseOptions(args))
failure();
try {
if (!app.validateConfiguration() ||
!app.run())
failure();
} catch (IOException ex) {
LOG.error(ex);
failure();
}
}
}