public static void main(String[] args) throws Exception {
String inputDirectory = "/home/cs246/Desktop/HW2/input";
String outputDirectory = "/home/cs246/Desktop/HW2/output";
String centroidDirectory = "/home/cs246/Desktop/HW2/config";
int iterations = 20;
for (int i = 1; i <= iterations; i++) {
Configuration conf = new Configuration();
String cFile = centroidDirectory + "/c" + i + ".txt";
String nextCFile = centroidDirectory + "/c" + (i + 1) + ".txt";
conf.set("CFILE", cFile);
conf.set("NEXTCFILE", nextCFile);
String cFile = centroidDirectory + "/c" + i + ".txt";
String nextCFile = centroidDirectory + "/c" + (i + 1) + ".txt";
conf.set("CFILE", cFile);
conf.set("NEXTCFILE", nextCFile);
Job job = new Job(conf, "HW2_Q4." + i);
job.setJarByClass(HW2_Q4.class);
job.setOutputKeyClass(IntWritable.class);
job.setOutputValueClass(Text.class);
job.setMapperClass(Map1.class);
job.setReducerClass(Reduce1.class);
job.setInputFormatClass(TextInputFormat.class);
job.setOutputFormatClass(TextOutputFormat.class);
FileInputFormat.addInputPath(job, new Path(inputDirectory));
FileOutputFormat.setOutputPath(job, new Path(outputDirectory + "/output" + i));
job.waitForCompletion(true);
}
}
/** Initialize SecondaryNameNode. */
private void initialize(Configuration conf) throws IOException {
// initiate Java VM metrics
JvmMetrics.init("SecondaryNameNode", conf.get("session.id"));
// Create connection to the namenode.
shouldRun = true;
nameNodeAddr = NameNode.getAddress(conf);
this.conf = conf;
this.namenode =
(NamenodeProtocol)
RPC.waitForProxy(
NamenodeProtocol.class, NamenodeProtocol.versionID, nameNodeAddr, conf);
// initialize checkpoint directories
fsName = getInfoServer();
checkpointDirs = FSImage.getCheckpointDirs(conf, "/tmp/hadoop/dfs/namesecondary");
checkpointEditsDirs = FSImage.getCheckpointEditsDirs(conf, "/tmp/hadoop/dfs/namesecondary");
checkpointImage = new CheckpointStorage(conf);
checkpointImage.recoverCreate(checkpointDirs, checkpointEditsDirs);
// Initialize other scheduling parameters from the configuration
checkpointPeriod = conf.getLong("fs.checkpoint.period", 3600);
checkpointSize = conf.getLong("fs.checkpoint.size", 4194304);
// initialize the webserver for uploading files.
String infoAddr =
NetUtils.getServerAddress(
conf,
"dfs.secondary.info.bindAddress",
"dfs.secondary.info.port",
"dfs.secondary.http.address");
InetSocketAddress infoSocAddr = NetUtils.createSocketAddr(infoAddr);
infoBindAddress = infoSocAddr.getHostName();
int tmpInfoPort = infoSocAddr.getPort();
infoServer = new HttpServer("secondary", infoBindAddress, tmpInfoPort, tmpInfoPort == 0, conf);
infoServer.setAttribute("name.system.image", checkpointImage);
this.infoServer.setAttribute("name.conf", conf);
infoServer.addInternalServlet("getimage", "/getimage", GetImageServlet.class);
infoServer.start();
// The web-server port can be ephemeral... ensure we have the correct info
infoPort = infoServer.getPort();
conf.set("dfs.secondary.http.address", infoBindAddress + ":" + infoPort);
LOG.info("Secondary Web-server up at: " + infoBindAddress + ":" + infoPort);
LOG.warn(
"Checkpoint Period :"
+ checkpointPeriod
+ " secs "
+ "("
+ checkpointPeriod / 60
+ " min)");
LOG.warn(
"Log Size Trigger :"
+ checkpointSize
+ " bytes "
+ "("
+ checkpointSize / 1024
+ " KB)");
}
/**
* Add a file path to the current set of classpath entries. It adds the file to cache as well.
* Intended to be used by user code.
*
* @param file Path of the file to be added
* @param conf Configuration that contains the classpath setting
* @param fs FileSystem with respect to which {@code archivefile} should be interpreted.
*/
public static void addFileToClassPath(Path file, Configuration conf, FileSystem fs)
throws IOException {
String filepath = file.toUri().getPath();
String classpath = conf.get("mapred.job.classpath.files");
conf.set(
"mapred.job.classpath.files",
classpath == null ? filepath : classpath + System.getProperty("path.separator") + filepath);
URI uri = fs.makeQualified(file).toUri();
addCacheFile(uri, conf);
}
/**
* Add an archive path to the current set of classpath entries. It adds the archive to cache as
* well. Intended to be used by user code.
*
* @param archive Path of the archive to be added
* @param conf Configuration that contains the classpath setting
* @param fs FileSystem with respect to which {@code archive} should be interpreted.
*/
public static void addArchiveToClassPath(Path archive, Configuration conf, FileSystem fs)
throws IOException {
String archivepath = archive.toUri().getPath();
String classpath = conf.get("mapred.job.classpath.archives");
conf.set(
"mapred.job.classpath.archives",
classpath == null
? archivepath
: classpath + System.getProperty("path.separator") + archivepath);
URI uri = fs.makeQualified(archive).toUri();
addCacheArchive(uri, conf);
}
@Override
public int run(String[] args) throws Exception {
String locatorHost = args[0];
int locatorPort = Integer.parseInt(args[1]);
String hdfsHomeDir = args[2];
System.out.println(
"KnownKeysMRv2 invoked with args (locatorHost = "
+ locatorHost
+ " locatorPort = "
+ locatorPort
+ " hdfsHomeDir = "
+ hdfsHomeDir);
Configuration conf = getConf();
conf.set(GFInputFormat.INPUT_REGION, "partitionedRegion");
conf.set(GFInputFormat.HOME_DIR, hdfsHomeDir);
conf.setBoolean(GFInputFormat.CHECKPOINT, false);
conf.set(GFOutputFormat.REGION, "validationRegion");
conf.set(GFOutputFormat.LOCATOR_HOST, locatorHost);
conf.setInt(GFOutputFormat.LOCATOR_PORT, locatorPort);
Job job = Job.getInstance(conf, "knownKeysMRv2");
job.setInputFormatClass(GFInputFormat.class);
job.setOutputFormatClass(GFOutputFormat.class);
job.setMapperClass(KnownKeysMRv2Mapper.class);
job.setMapOutputKeyClass(GFKey.class);
job.setMapOutputValueClass(PEIWritable.class);
job.setReducerClass(KnownKeysMRv2Reducer.class);
// job.setOutputKeyClass(String.class);
// job.setOutputValueClass(ValueHolder.class);
return job.waitForCompletion(false) ? 0 : 1;
}
private static URI addArchiveToClassPathHelper(Path archive, Configuration conf)
throws IOException {
String classpath = conf.get("mapred.job.classpath.archives");
// the scheme/authority use ':' as separator. put the unqualified path in classpath
String archivePath = archive.toUri().getPath();
conf.set(
"mapred.job.classpath.archives",
classpath == null
? archivePath
: classpath + System.getProperty("path.separator") + archivePath);
return archive.makeQualified(archive.getFileSystem(conf)).toUri();
}
/**
* This is to check the timestamp of the archives to be localized
*
* @param conf Configuration which stores the timestamp's
* @param timestamps comma separated list of timestamps of archives. The order should be the same
* as the order in which the archives are added.
*/
public static void setArchiveTimestamps(Configuration conf, String timestamps) {
conf.set("mapred.cache.archives.timestamps", timestamps);
}
/**
* Set the configuration with the given set of files
*
* @param files The list of files that need to be localized
* @param conf Configuration which will be changed
*/
public static void setCacheFiles(URI[] files, Configuration conf) {
String sfiles = StringUtils.uriToString(files);
conf.set("mapred.cache.files", sfiles);
}
/**
* Set the configuration with the given set of archives
*
* @param archives The list of archives that need to be localized
* @param conf Configuration which will be changed
*/
public static void setCacheArchives(URI[] archives, Configuration conf) {
String sarchives = StringUtils.uriToString(archives);
conf.set("mapred.cache.archives", sarchives);
}
/**
* Add a archive that has been localized to the conf. Used by internal DistributedCache code.
*
* @param conf The conf to modify to contain the localized caches
* @param str a comma separated list of local archives
*/
public static void addLocalArchives(Configuration conf, String str) {
String archives = conf.get(CACHE_LOCALARCHIVES);
conf.set(CACHE_LOCALARCHIVES, archives == null ? str : archives + "," + str);
}
FileDataGenNew(String HDFSMaster) {
fsConf.set("fs.default.name", HDFSMaster);
}
/**
* Add a archives to be localized to the conf
*
* @param uri The uri of the cache to be localized
* @param conf Configuration to add the cache to
*/
public static void addSharedCacheArchive(URI uri, Configuration conf) {
String archives = conf.get("mapred.cache.shared.archives");
conf.set(
"mapred.cache.shared.archives",
archives == null ? uri.toString() : archives + "," + uri.toString());
}
/**
* Set the conf to contain the location for localized archives
*
* @param conf The conf to modify to contain the localized caches
* @param str a comma separated list of local archives
*/
public static void setLocalArchives(Configuration conf, String str) {
conf.set("mapred.cache.localArchives", str);
}
/**
* This method allows you to create symlinks in the current working directory of the task to all
* the cache files/archives. Intended to be used by user code.
*
* @param conf the jobconf
*/
public static void createSymlink(Configuration conf) {
conf.set(CACHE_SYMLINK, "yes");
}
/**
* This is to check the timestamp of the files to be localized. Used by internal MapReduce code.
*
* @param conf Configuration which stores the timestamp's
* @param timestamps comma separated list of timestamps of files. The order should be the same as
* the order in which the files are added.
*/
public static void setFileTimestamps(Configuration conf, String timestamps) {
conf.set(CACHE_FILES_TIMESTAMPS, timestamps);
}
/**
* Set the conf to contain the location for localized archives. Used by internal DistributedCache
* code.
*
* @param conf The conf to modify to contain the localized caches
* @param str a comma separated list of local archives
*/
public static void setLocalArchives(Configuration conf, String str) {
conf.set(CACHE_LOCALARCHIVES, str);
}
/**
* Add a file to be localized to the conf. Intended to be used by user code.
*
* @param uri The uri of the cache to be localized
* @param conf Configuration to add the cache to
*/
public static void addCacheFile(URI uri, Configuration conf) {
String files = conf.get(CACHE_FILES);
conf.set(CACHE_FILES, files == null ? uri.toString() : files + "," + uri.toString());
}
/**
* Add a archives to be localized to the conf. Intended to be used by user code.
*
* @param uri The uri of the cache to be localized
* @param conf Configuration to add the cache to
*/
public static void addCacheArchive(URI uri, Configuration conf) {
String archives = conf.get(CACHE_ARCHIVES);
conf.set(CACHE_ARCHIVES, archives == null ? uri.toString() : archives + "," + uri.toString());
}
/**
* Add a file that has been localized to the conf.. Used by internal DistributedCache code.
*
* @param conf The conf to modify to contain the localized caches
* @param str a comma separated list of local files
*/
public static void addLocalFiles(Configuration conf, String str) {
String files = conf.get(CACHE_LOCALFILES);
conf.set(CACHE_LOCALFILES, files == null ? str : files + "," + str);
}
/**
* This is to check the timestamp of the files to be localized
*
* @param conf Configuration which stores the timestamp's
* @param timestamps comma separated list of timestamps of files. The order should be the same as
* the order in which the files are added.
*/
public static void setFileTimestamps(Configuration conf, String timestamps) {
conf.set("mapred.cache.files.timestamps", timestamps);
}
public static void setSharedFileLength(Configuration conf, String length) {
conf.set("mapred.cache.shared.files.length", length);
}
/**
* This is to check the timestamp of the archives to be localized. Used by internal MapReduce
* code.
*
* @param conf Configuration which stores the timestamp's
* @param timestamps comma separated list of timestamps of archives. The order should be the same
* as the order in which the archives are added.
*/
public static void setArchiveTimestamps(Configuration conf, String timestamps) {
conf.set(CACHE_ARCHIVES_TIMESTAMPS, timestamps);
}
/**
* Set the conf to contain the location for localized files
*
* @param conf The conf to modify to contain the localized caches
* @param str a comma separated list of local files
*/
public static void setLocalSharedFiles(Configuration conf, String str) {
conf.set("mapred.cache.shared.localFiles", str);
}
/**
* Set the configuration with the given set of files. Intended to be used by user code.
*
* @param files The list of files that need to be localized
* @param conf Configuration which will be changed
*/
public static void setCacheFiles(URI[] files, Configuration conf) {
String sfiles = StringUtils.uriToString(files);
conf.set(CACHE_FILES, sfiles);
}
/**
* Add a file to be localized to the conf
*
* @param uri The uri of the cache to be localized
* @param conf Configuration to add the cache to
*/
public static void addCacheFile(URI uri, Configuration conf) {
String files = conf.get("mapred.cache.files");
conf.set("mapred.cache.files", files == null ? uri.toString() : files + "," + uri.toString());
}
/**
* This method allows you to create symlinks in the current working directory of the task to all
* the cache files/archives
*
* @param conf the jobconf
*/
public static void createSymlink(Configuration conf) {
conf.set("mapred.create.symlink", "yes");
}
/**
* Set the configuration with the given set of archives. Intended to be used by user code.
*
* @param archives The list of archives that need to be localized
* @param conf Configuration which will be changed
*/
public static void setCacheArchives(URI[] archives, Configuration conf) {
String sarchives = StringUtils.uriToString(archives);
conf.set(CACHE_ARCHIVES, sarchives);
}
/**
* Set the conf to contain the location for localized files. Used by internal DistributedCache
* code.
*
* @param conf The conf to modify to contain the localized caches
* @param str a comma separated list of local files
*/
public static void setLocalFiles(Configuration conf, String str) {
conf.set(CACHE_LOCALFILES, str);
}
