conf/pig.properties

# 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.

# Pig configuration file. All values can be overwritten by command line arguments.

# Use the "-h properties" command to see description of the properties

# log4jconf log4j configuration file
# log4jconf=./conf/log4j.properties

# a file that contains pig script
#file=

# load jarfile, colon separated
#jar=

#verbose print all log messages to screen (default to print only INFO and above to screen)
#verbose=true

#exectype local|mapreduce, mapreduce is default
#exectype=local

#the default timezone: if it is not set, the default timezone for this host is used.
#the correct timezone format is the UTC offset: e.g., +08:00. 
#pig.datetime.default.tz=

#pig.logfile=

#Do not spill temp files smaller than this size (bytes)
#pig.spill.size.threshold=5000000

#EXPERIMENT: Activate garbage collection when spilling a file bigger than this size (bytes)
#This should help reduce the number of files being spilled.
#pig.spill.gc.activation.size=40000000

#the following two parameters are to help estimate the reducer number
#pig.exec.reducers.bytes.per.reducer=1000000000
#pig.exec.reducers.max=999

#Logging properties
#verbose=false
#brief=false
#debug=INFO
#aggregate.warning=true

#Performance tuning properties
#pig.cachedbag.memusage=0.2
#pig.skewedjoin.reduce.memusage=0.3
#pig.exec.nocombiner=false
#opt.multiquery=true

#Following parameters are for configuring intermediate storage format
#Supported storage types are seqfile and tfile
#Supported codec types: tfile supports gz(gzip) and lzo, seqfile support gz(gzip), lzo, snappy, bzip2
#pig.tmpfilecompression=false
#pig.tmpfilecompression.storage=seqfile
#pig.tmpfilecompression.codec=gz

#pig.noSplitCombination=true

#pig.exec.mapPartAgg=false
#pig.exec.mapPartAgg.minReduction=10

#exectype=mapreduce
#pig.additional.jars=<comma seperated list of jars>
#udf.import.list=<comma seperated list of imports>
#stop.on.failure=false

#Use this option only when your Pig job will otherwise die because of
#using more counters than hadoop configured limit
#pig.disable.counter=true

# By default, pig will allow 1GB of data to be replicated using
# the distributed cache when doing fragment-replicated join.
# pig.join.replicated.max.bytes=1000000000

# Use this option to turn on UDF timers. This will cause two
# counters to be tracked for every UDF and LoadFunc in your script:
# approx_microsecs measures approximate time spent inside a UDF
# approx_invocations reports the approximate number of times the UDF was invoked
# pig.udf.profile=false

#When enabled, 'describe' prints a multi-line formatted schema
#(similar to an indended json) rather than on a single line.
#pig.pretty.print.schema=true

#pig.sql.type=hcat
hcat.bin=/usr/local/hcat/bin/hcat

############################ SchemaTuple ############################

# Setting this value will turn on the SchemaTuple feature (PIG-2632)
# This will attempt to use code generation for more efficient within
# the pig code. This can lead to both CPU, serialization, and memory
# benefits (currently, the potential memory benefits are the largest).

# This parameter will enable the optimization in all available cases
#pig.schematuple=true

# Certain cases can be turned off by uncommenting the following. These will
# all be off by default, but will all be turned on if pig.schematuple is set
# to true.

# This will disable SchemaTuples in the case of udfs. Currently,
# the input to UDF's will be SchemaTuples.

#pig.schematuple.udf=false

# This is currently not implemented. In the future, LoadFunc's with known
# schema's should output SchemaTuples

#pig.schematuple.load=false

# This will use SchemaTuples in replicated joins. The potential memory saving
# here is significant. It will use SchemaTuples when it builds the HashMap of
# the join key to related values.

#pig.schematuple.fr_join=false

# In the current implementation of merge join, all of the Tuples in the left relation
# that share a given key will be stored in a List in memory. This will use SchemaTuples
# instead in that List.

#pig.schematuple.merge_join=false

#####################################################################

##### Set up optional Pig Progress Notification Listener ############

# Note that only one PPNL can be set up. If you need several, write a PPNL that will chain them.
# pig.notification.listener = <fully qualified class name of a PPNL implementation>

# Optionally, you can supply a single String argument to pass to your PPNL. 
# pig.notification.listener.arg = <somevalue>

#####################################################################

########## Override the default Reducer Estimator logic #############

# By default, the logic to estimate the number of reducers to use for a given job lives in:
#   org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.InputSizeReducerEstimator
# This logic can be replaced by implementing the following interface:
#   org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigReducerEstimator

# This class will be invoked to estimate the number of reducers to use.
# pig.exec.reducer.estimator = <fully qualified class name of a PigReducerEstimator implementation>

# Optionally, you can supply a single String argument to pass to your PigReducerEstimator.
# pig.exec.reducer.estimator.arg = <somevalue>

#####################################################################

###### Override the default Pig Stats Output Size Reader logic ######

# By default, the size of reducers output is computed as the total size of
# output files. But since not every storage is file-based, this logic is not
# always applicable. If that is the case, the logic can be replaced by
# implementing the following interface:
#   org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigStatsOutputSizeReader

# This class will be invoked to compute the size of reducers output.
# pig.stats.output.size.reader = <fully qualified class name of a PigStatsOutputSizeReader implementation>

# If you need to register more than one reader, you can register them as a comma
# separated list. Every reader implements a boolean supports(POStore sto) method.
# When there are more than one reader, they are consulted in order, and the
# first one whose supports() method returns true will be used.
#
#####################################################################

#pig.load.default.statements=

#####################################################################

########### Override hadoop configs programatically #################

# By default, Pig expects hadoop configs (hadoop-site.xml and core-site.xml)
# to be present on the classpath. There are cases when these configs are 
# needed to be passed programatically, such as while using the PigServer API.
# In such cases, you can override hadoop configs by setting the property
# "pig.use.overriden.hadoop.configs". 
# 
# When this property is set to true, Pig ignores looking for hadoop configs
# in the classpath and instead picks it up from Properties/Configuration
# object passed to it.

# pig.use.overriden.hadoop.configs=false
#
######################################################################

# Check if the script needs to check multiple stores writing
# to the same location. When set to true, stops the execution
# of script right away.
pig.location.check.strict=false

######################################################################

# This key is used to define the default load func. Pig will fallback 
# on PigStorage as default in case this is undefined.

# pig.default.load.func=<fully qualified class name of a LoadFunc implementation>
# For eg, pig.default.load.func=org.apache.pig.custom.MyCustomStorage

# This key is used to define the default store func. Pig will fallback 
# on PigStorage as default in case this is undefined.

# pig.default.store.func=<fully qualified class name of a StoreFunc implementation>
# For eg, pig.default.store.func=org.apache.pig.custom.MyCustomStorage

# This option is used to define whether to support recovery to handle the
# application master getting restarted.
# pig.output.committer.recovery.support=true

# Set this option to true if you need to use the old partition filter optimizer. 
# Note: Old filter optimizer PColFilterOptimizer will be deprecated in the future.
# pig.exec.useOldPartitionFilterOptimizer=true