#! /bin/bash # # # # # # # # # # # # # # # # # # # # # # # @file :extract_device.sh # @author :liushuai # @revision:2017-01-18 16:52 # # # # # # # # # # # # # # # # # # # # # # source ../dmp_env.sh LOG_TIME=$(date +%Y%m%d -d "-1 day $ScheduleTime") year=${LOG_TIME:0:4} month=${LOG_TIME:4:2} day=${LOG_TIME:6:2} INPUT_PATH="${GENDER_MERGE_INSTALL}/${year}/${month}/${day}/" OUTPUT_PATH="${GENDER_CALC_PACKAGE_DICT}/${year}/${month}/${day}/" LOW=50 HIGH=50000 MALE_RATIO=0.85 FEMALE_RATIO=0.15 spark-submit --class mobvista.dmp.datasource.age_gender.CalcPackageGender \ --name "CalcPackageGender.${LOG_TIME}" \ --conf spark.default.parallelism=1000 \ --conf spark.storage.memoryFraction=0.4 \ --conf spark.shuffle.memoryFraction=0.4 \ --conf spark.sql.files.maxPartitionBytes=134217728 \ --master yarn --deploy-mode cluster --executor-memory 8g --driver-memory 4g --executor-cores 4 --num-executors 100 \ ../${JAR} -inputPath ${INPUT_PATH} -outputPath ${OUTPUT_PATH} -parallelism 1 \ -lowThreshold ${LOW} -highThreshold ${HIGH} -mRatio ${MALE_RATIO} -fRatio ${FEMALE_RATIO} if [[ $? -ne 0 ]]; then exit 255 fi