src/main/resources/pirk.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.
###############################################################################

##
## Default System Configuration Properties
##

#Name of the local properties file - used when running with the
#hadoop jar command
local.pirk.properties.dir=/root/

##
##Spark path for SparkLauncher
##
spark.home = /usr

##
## Data schema properties
##
## Each data schema should be specified in an xml file of the form:
##
##<schema>
##  <schemaName> name of the schema </schemaName>
##  <element>
##      <name> element name </name>
##      <type> class name or type name (if Java primitive type) of the element </type>
##      <isArray> true or false -- whether or not the schema element is an array within the data </isArray>
##      <partitioner> optional - Partitioner class for the element; defaults to primitive java type partitioner </partitioner> 
##  </element>
## </schema>
##
## Primitive Java types must be one of the following: "byte", "short", "int", "long", "float", 
##                  "double", "char", "string", "boolean"
##

#Comma separated list of local data schema files to load, fully qualified file names
data.schemas = none

##
## Query schema properties
##
## Each query schema should be specified in an xml file of the form; 
## 		all items are treated in a case insensitive manner:
##
##<schema>
##  <schemaName> name of the schema </schemaName>
##	<dataSchemaName> name of the data schema over which this query is run </dataSchemaName>
##	<selectorName> name of the element in the data schema that will be the selector </selectorName>
##  <elements>
##     	<name> element name of element in the data schema to include in the query response </name>
## 	</elements>
##  <filter> (optional) name of the filter class to use to filter the data </filter>
##  <filterNames>
##       <name> (optional) element name of element in the data schema to apply pre-processing filters </name>
##  </filterNames>
##  <additional> (optional) additional fields for the query schema, in <key,value> pairs
##		<field>
##			<key> key corresponding the the field </key>
##			<value> value corresponding to the field </value>
##		</field>	
##  </additional>
## </schema>
##
##

#Comma separated list of local query schema files to load, fully qualified file names
query.schemas = none

##
##Properties for ElasticSearch compatibility
##

#ES host address - One Elasticsearch node in the cluster - may include port specification
es.nodes=none
          
#Default HTTP/REST port used for connecting to Elasticsearch 
es.port=9200

#Number of results/items returned by each individual scroll
es.scroll.size = 1000

#Whether elasticsearch-hadoop will treat empty fields as null
es.field.read.empty.as.null=yes


##
##Properties for functional testing
##

#Test index to create in ES (created and deleted within the tests)
#Need final form to be: <host>:<port>/<test.esTestIndex>
test.es.index = testindex

#Type of elements to insert in ES
test.es.type = pkt

#Elasticsearch resource - Elasticsearch resource location where data is read and written to. 
#Requires the format <index>/<type> 
test.es.resource= /testindex/pkt

#Pathname in hdfs to place input JSON file testing
test.inputJSONFile = /tmp/testJSONInput

#Pathname in hdfs to place output file for testing
test.outputHDFSFile = /tmp/testOutput

#Query input dir in hdfs for testing
test.queryInputDir = none

#Test stoplist file
test.stopListFile = /tmp/testStopListFile

#Whether or not we are running PIR testing (used as a flag to dump intermediate RDDs for checking)
#This should default to false; it is changed to true in the test suite, as applicable
pir.test = false

#HDFS output dir for PIR intermediate testing
#Should default to none; changed to a real hdfs path in the test suite, as applicable
pir.testOut = none


##
## Properties to enable/disable JNA-GMP modPow acceleration for Paillier 
##

paillier.useGMPForModPow = true

# The JNA-GMP library we use to invoke the much faster, natively compiled GMP
# can be called in a way that tries to make all modPow calls take a constant amount 
# of time to execute. This will slow down the modPow performance (still better than
# Java's BigInteger.modPow() ). 
# If you are using this package in a multi-tenant computing environment and have 
# concerns about other tenants being able to closely inspect the runtime of your
# software, you may want to enable this property. 
paillier.GMPConstantTimeMode = false

# This property controls the more rigorous prime generation checks in PrimeMaker.java
# which are based on FIPS SP 800-56B and FIPS 186-4 (extra Miller-Rabin rounds, limits
# on how close the primes p and q can be, and bounds on the values of the primes)
# These checks slow down prime generation considerably
pallier.FIPSPrimeGenerationChecks = true

## These properties control the secure random number generator algorithm and provider.
## You can specify just the algorithm, or both algorithm and provider.  The system's
## default secure random is used when the algorithm is left unspecified. 
pallier.secureRandom.algorithm=NativePRNG
#pallier.secureRandom.provider=SUN

##
## Properties for PIR query and response
##

#Number of bits to return when encoding/returning string values as part of return data elements
pir.stringBits = 64

#Number of array elements to return when encoding/returning array valued elements
pir.numReturnArrayElements = 2

#Default prime certainty
pir.primeCertainty = 128

#Fully qualified dir in hdfs of Query files
pir.queryInput = none

#Data input format type -- 'base' or 'elasticsearch' (given in InputFormatsConst class)
pir.dataInputFormat = base

#Fully qualified name of input file/directory in hdfs; used if pir.dataInputFormat = base
pir.inputData = none

#Inputformat for 'base' data input format type -- must extend BaseInputFormat
pir.baseInputFormat = none

#ElasticSearch-like query if using 'base' input format
pir.baseQuery = none

#ES resource for input data
pir.esResource = null

#ES query for input data
pir.esQuery = none

#Fully qualified name of output file in hdfs
pir.outputFile = none

#Fully qualified dir in hdfs of file containing stoplist terms
pir.stopListFile = stopListFile

#Number of reduce tasks
pir.numReduceTasks = 100

#Whether or not to use the local cache during PIR computations
pir.useLocalCache = true

#Whether or not to limit the hits for each query term
pir.limitHitsPerSelector = true

#Number of hits to limit for each query term, if pir.limitHitsPerSelector = true
pir.maxHitsPerSelector = 100

#Whether or not to embed the selector in the results for false positive reduction
pir.embedSelector = true

#Whether or not to generate and use the HDFS modular exponentiation lookup table
pir.useHDFSLookupTable = false

#Number of partitions to coalesce the input data into in Spark
pir.numDataPartitions = 1500

#Mapreduce memory options
mapreduce.map.memory.mb = 3000
mapreduce.reduce.memory.mb = 3000
mapreduce.map.java.opts = -Xmx2800m
mapreduce.reduce.java.opts = -Xmx2800m

#HDFS directory for the expLookupTable 
pir.expDir = none
		
#Parallelism for expLookupTable creation in hdfs 
pir.expCreationSplits = 600