namevaluedescription
hadoop.tmp.dir/tmp/hadoop-${user.name}A base for other temporary directories.
hadoop.logfile.size10000000The max size of each log file
hadoop.logfile.count10The max number of log files
dfs.namenode.logging.levelinfoThe logging level for dfs namenode. Other values are "dir"(trac e namespace mutations), "block"(trace block under/over replications and block creations/deletions), or "all".
io.sort.factor10The number of streams to merge at once while sorting files. This determines the number of open file handles.
io.sort.mb100The total amount of buffer memory to use while sorting files, in megabytes. By default, gives each merge stream 1MB, which should minimize seeks.
io.file.buffer.size4096The size of buffer for use in sequence files. The size of this buffer should probably be a multiple of hardware page size (4096 on Intel x86), and it determines how much data is buffered during read and write operations.
io.bytes.per.checksum512The number of bytes per checksum. Must not be larger than io.file.buffer.size.
io.skip.checksum.errorsfalseIf true, when a checksum error is encountered while reading a sequence file, entries are skipped, instead of throwing an exception.
io.map.index.skip0Number of index entries to skip between each entry. Zero by default. Setting this to values larger than zero can facilitate opening large map files using less memory.
io.compression.codecsorg.apache.hadoop.io.compress.DefaultCodec,org.apache.hadoop.io.compress.GzipCodecA list of the compression codec classes that can be used for compression/decompression.
fs.default.namelocalThe name of the default file system. Either the literal string "local" or a host:port for DFS.
dfs.datanode.bindAddress0.0.0.0 the address where the datanode will listen to.
dfs.datanode.port50010The port number that the dfs datanode server uses as a starting point to look for a free port to listen on.
dfs.info.bindAddress0.0.0.0 the address where the dfs namenode web ui will listen on.
dfs.info.port50070The base port number for the dfs namenode web ui.
dfs.datanode.dns.interfacedefaultThe name of the Network Interface from which a data node should report its IP address.
dfs.datanode.dns.nameserverdefaultThe host name or IP address of the name server (DNS) which a DataNode should use to determine the host name used by the NameNode for communication and display purposes.
dfs.default.chunk.view.size2048The number of bytes to view for a file on the browser.
dfs.datanode.du.reserved0Reserved space in bytes. Always leave this much space free for non dfs use
dfs.datanode.du.pct0.98fWhen calculating remaining space, only use this percentage of the real available space
dfs.name.dir${hadoop.tmp.dir}/dfs/nameDetermines where on the local filesystem the DFS name node should store the name table.
dfs.client.buffer.dir${hadoop.tmp.dir}/dfs/tmpDetermines where on the local filesystem an DFS client should store its blocks before it sends them to the datanode.
dfs.data.dir${hadoop.tmp.dir}/dfs/dataDetermines where on the local filesystem an DFS data node should store its blocks. If this is a comma-delimited list of directories, then data will be stored in all named directories, typically on different devices. Directories that do not exist are ignored.
dfs.replication3Default block replication. The actual number of replications can be specified when the file is created. The default is used if replication is not specified in create time.
dfs.replication.max512Maximal block replication.
dfs.replication.min1Minimal block replication.
dfs.block.size67108864The default block size for new files.
dfs.df.interval3000Disk usage statistics refresh interval in msec.
dfs.client.block.write.retries3The number of retries for writing blocks to the data nodes, before we signal failure to the application.
dfs.blockreport.intervalMsec3600000Determines block reporting interval.
dfs.safemode.threshold.pct0.999f Specifies the percentage of blocks that should satisfy the minimal replication requirement defined by dfs.replication.min. Values less than or equal to 0 mean not to start in safe mode. Values greater than 1 will make safe mode permanent.
dfs.safemode.extension30000 Determines extension of safe mode in milliseconds after the threshold level is reached.
mapred.job.trackerlocalThe host and port that the MapReduce job tracker runs at. If "local", then jobs are run in-process as a single map and reduce task.
mapred.job.tracker.info.bindAddress0.0.0.0 the address where the job tracker info webserver will be binded on.
mapred.job.tracker.info.port50030The port that the MapReduce job tracker info webserver runs at.
mapred.task.tracker.report.bindAddress0.0.0.0 the address where the maperd tracker report server will be binded on.
mapred.task.tracker.report.port50050The port number that the MapReduce task tracker report server uses as a starting point to look for a free port to listen on.
mapred.local.dir${hadoop.tmp.dir}/mapred/localThe local directory where MapReduce stores intermediate data files. May be a comma-separated list of directories on different devices in order to spread disk i/o. Directories that do not exist are ignored.
local.cache.size10737418240The limit on the size of cache you want to keep, set by default to 10GB. This will act as a soft limit on the cache directory for out of band data.
mapred.system.dir${hadoop.tmp.dir}/mapred/systemThe shared directory where MapReduce stores control files.
mapred.temp.dir${hadoop.tmp.dir}/mapred/tempA shared directory for temporary files.
mapred.local.dir.minspacestart0If the space in mapred.local.dir drops under this, do not ask for more tasks. Value in bytes.
mapred.local.dir.minspacekill0If the space in mapred.local.dir drops under this, do not ask more tasks until all the current ones have finished and cleaned up. Also, to save the rest of the tasks we have running, kill one of them, to clean up some space. Start with the reduce tasks, then go with the ones that have finished the least. Value in bytes.
mapred.map.tasks2The default number of map tasks per job. Typically set to a prime several times greater than number of available hosts. Ignored when mapred.job.tracker is "local".
mapred.reduce.tasks1The default number of reduce tasks per job. Typically set to a prime close to the number of available hosts. Ignored when mapred.job.tracker is "local".
mapred.reduce.parallel.copies5The default number of parallel transfers run by reduce during the copy(shuffle) phase.
mapred.task.timeout600000The number of milliseconds before a task will be terminated if it neither reads an input, writes an output, nor updates its status string.
mapred.tasktracker.tasks.maximum2The maximum number of tasks that will be run simultaneously by a task tracker.
mapred.child.java.opts-Xmx200mJava opts for the task tracker child processes. Subsumes 'mapred.child.heap.size' (If a mapred.child.heap.size value is found in a configuration, its maximum heap size will be used and a warning emitted that heap.size has been deprecated). Also, the following symbols, if present, will be interpolated: @taskid@ is replaced by current TaskID; and @port@ will be replaced by mapred.task.tracker.report.port + 1 (A second child will fail with a port-in-use if mapred.tasktracker.tasks.maximum is greater than one). Any other occurrences of '@' will go unchanged. For example, to enable verbose gc logging to a file named for the taskid in /tmp and to set the heap maximum to be a gigabyte, pass a 'value' of: -Xmx1024m -verbose:gc -Xloggc:/tmp/@taskid@.gc
mapred.combine.buffer.size100000The number of entries the combining collector caches before combining them and writing to disk.
mapred.speculative.executiontrueIf true, then multiple instances of some map tasks may be executed in parallel.
mapred.min.split.size0The minimum size chunk that map input should be split into. Note that some file formats may have minimum split sizes that take priority over this setting.
mapred.submit.replication10The replication level for submitted job files. This should be around the square root of the number of nodes.
mapred.tasktracker.dns.interfacedefaultThe name of the Network Interface from which a task tracker should report its IP address.
mapred.tasktracker.dns.nameserverdefaultThe host name or IP address of the name server (DNS) which a TaskTracker should use to determine the host name used by the JobTracker for communication and display purposes.
tasktracker.http.threads40The number of worker threads that for the http server. This is used for map output fetching
tasktracker.http.bindAddress0.0.0.0 the address where the task tracker http server will be binded on.
tasktracker.http.port50060The default port for task trackers to use as their http server.
keep.failed.task.filesfalseShould the files for failed tasks be kept. This should only be used on jobs that are failing, because the storage is never reclaimed. It also prevents the map outputs from being erased from the reduce directory as they are consumed.
mapred.output.compressfalseShould the outputs of the reduces be compressed?
mapred.output.compression.codecorg.apache.hadoop.io.compress.DefaultCodecIf the reduce outputs are compressed, how should they be compressed?
mapred.compress.map.outputfalseShould the outputs of the maps be compressed before being sent across the network. Uses SequenceFile compression.
io.seqfile.compress.blocksize1000000The minimum block size for compression in block compressed SequenceFiles.
io.seqfile.lazydecompresstrueShould values of block-compressed SequenceFiles be decompressed only when necessary.
io.seqfile.sorter.recordlimit1000000The limit on number of records to be kept in memory in a spill in SequenceFiles.Sorter
io.seqfile.compression.typeRECORDThe default compression type for SequenceFile.Writer.
ipc.client.timeout60000Defines the timeout for IPC calls in milliseconds.
ipc.client.idlethreshold4000Defines the threshold number of connections after which connections will be inspected for idleness.
ipc.client.maxidletime120000Defines the maximum idle time for a connected client after which it may be disconnected.
ipc.client.kill.max10Defines the maximum number of clients to disconnect in one go.
ipc.client.connection.maxidletime1000The maximum time after which a client will bring down the connection to the server.
ipc.client.connect.max.retries10Indicates the number of retries a client will make to establish a server connection.
ipc.server.listen.queue.size128Indicates the length of the listen queue for servers accepting client connections.