bin/rebalance-zone-expansion.sh

   1 #!/bin/bash -e
   2
   3 #
   4 #   Copyright 2013 LinkedIn, Inc
   5 #
   6 #  Licensed under the Apache License, Version 2.0 (the "License");
   7 #  you may not use this file except in compliance with the License.
   8 #  You may obtain a copy of the License at
   9 #
  10 #      http://www.apache.org/licenses/LICENSE-2.0
  11 #
  12 #  Unless required by applicable law or agreed to in writing, software
  13 #  distributed under the License is distributed on an "AS IS" BASIS,
  14 #  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  15 #  See the License for the specific language governing permissions and
  16 #  limitations under the License.
  17
  18 # This script generates a cluster.xml and a plan for the zone expansion.
  19 # The final cluster is placed in output_dir/
  20 # Argument = -c current_cluster -s current_stores -i interim_cluster -f final_stores
  21 #            -o output dir
  22
  23 # This script uses getopts which means only single character switches are allowed.
  24 # Using getopt would allow for multi charcter switch names but would come at a
  25 # cost of not being cross compatible.
  26
  27 # Function to display usage
  28 usage_and_exit() {
  29   echo "ERROR: $1."
  30   cat <<EOF
  31
  32   Usage: $0 options
  33   OPTIONS:
  34    -h     Show this message
  35    -c     Current cluster that describes the cluster
  36    -s     Current stores that describes the store. If you do not have info about the stores yet, look
  37           under 'voldemort_home/config/tools/' for some store examples.
  38    -i     Interim Cluster that corresponds to zone expansion.
  39    -f     Final Stores that corresponds to zone expansion.
  40    -o     Output dir where all interim and final files will be stored.
  41 EOF
  42 exit 1
  43 }
  44
  45 # initialize  variables to an empty string
  46 current_cluster=""
  47 current_stores=""
  48 interim_cluster=""
  49 final_stores=""
  50 output_dir=""
  51
  52 # Figure out voldemort home directory
  53 dir="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
  54 vold_home="$(dirname "$dir")"
  55
  56 # Parse options
  57 while getopts “hc:s:i:f:o:” OPTION
  58 do
  59   case $OPTION in
  60   h)
  61     usage_and_exit
  62     exit 1
  63     ;;
  64   c)
  65     current_cluster=$OPTARG
  66     echo "[rebalance-zone-expansion] Will rebalance on the cluster described in '$current_cluster'."
  67     ;;
  68   s)
  69     current_stores=$OPTARG
  70     echo "[rebalance-zone-expansion] Will rebalance on the stores described in '$current_stores'."
  71     ;;
  72   i)
  73     interim_cluster=$OPTARG
  74     echo "[rebalance-zone-expansion] Will rebalance on the interim cluster described in '$interim_cluster'."
  75     ;;
  76   f)
  77     final_stores=$OPTARG
  78     echo "[rebalance-zone-expansion] Will rebalance on the final stores described in '$final_stores'."
  79     ;;
  80   o)
  81     output_dir=$OPTARG
  82     output_dir=$(readlink -m $output_dir)
  83     mkdir -p $output_dir
  84     echo "[rebalance-zone-expansion] Using '$output_dir' for all interim and final files generated."
  85     ;;
  86   ?)
  87     usage_and_exit
  88   ;;
  89      esac
  90 done
  91
  92 if  [[ -z $current_cluster ]] || [[ -z $current_stores ]] || [[ -z $interim_cluster ]] \
  93     || [[ -z $final_stores ]] || [[ -z $output_dir ]]
  94 then
  95      printf "\n"
  96      echo "[rebalance-zone-expansion] Missing argument. Check again."
  97      usage_and_exit
  98      exit 1
  99 fi
 100
 101 if [ ! -e $current_cluster ]; then
 102     usage_and_exit "File '$current_cluster' does not exist."
 103 fi
 104
 105 if [ ! -e $current_stores ]; then
 106     usage_and_exit "File '$current_stores' does not exist."
 107 fi
 108
 109 if [ ! -e $interim_cluster ]; then
 110     usage_and_exit "File '$interim_cluster' does not exist."
 111 fi
 112
 113 if [ ! -e $final_stores ]; then
 114     usage_and_exit "File '$final_stores' does not exist."
 115 fi
 116
 117
 118 # The final cluster.xml for zone expansion is generated in three steps.
 119 # Step 1 : In step 1 1000 separate iterations of the repartitioner is executed and the one
 120 #          with the minimal utility value is chose for step 2.
 121 # Step 2: In step 2, the cluster.xml from step 1 is fed to the repartitioner along with random swap
 122 #         attempts. The repartitioner randomly swaps the partitions only in zone 2
 123 #         and tries to balance the ring.
 124 # Step 3: Finally, a plan is generated on how to reach from the orignal cluster topology to
 125 #         the one that is generated in step 2.
 126 #
 127
 128 step2_zoneid=2
 129 step2_swap_attempts=1000
 130 step2_overall_iterations=5
 131
 132
 133 # Step 1
 134 for i in {1..1000}
 135 do
 136     mkdir -p $output_dir/step1/$i
 137     $vold_home/bin/run-class.sh voldemort.tools.RepartitionerCLI \
 138                                 --current-cluster $current_cluster \
 139                                 --current-stores $current_stores \
 140                                 --interim-cluster $interim_cluster \
 141                                 --final-stores $final_stores \
 142                                 --attempts 1 \
 143                                 --output-dir $output_dir/step1/$i
 144 done
 145
 146 #find the run with the best (minimal) utility value
 147 bestUtil="$(grep "Utility" $output_dir/step1/*/final-cluster.xml.analysis | sort -nk 3 | head -n 1 | cut -f1 -d ' ')"
 148 remove=".analysis:Utility"
 149 path_to_bestUtil=${bestUtil%%$remove}
 150
 151 if [ ! -e $path_to_bestUtil ]; then
 152     usage_and_exit "final cluster.xml from step1 does not exist"
 153 fi
 154
 155 # Step 2
 156 $vold_home/bin/run-class.sh voldemort.tools.RepartitionerCLI \
 157                             --current-cluster $path_to_bestUtil \
 158                             --current-stores $final_stores \
 159                             --output-dir $output_dir/step2 \
 160                             --enable-random-swaps \
 161                             --random-swap-zoneids $step2_zoneid \
 162                             --attempts $step2_overall_iterations \
 163                             --random-swap-attempts $step2_swap_attempts \
 164
 165 # Step 3
 166 mkdir -p $output_dir/step3/
 167 $vold_home/bin/run-class.sh voldemort.tools.RebalancePlanCLI \
 168                              --current-cluster $current_cluster \
 169                              --current-stores $current_stores \
 170                              --final-cluster $output_dir/step2/final-cluster.xml \
 171                              --final-stores $final_stores \
 172                              --output-dir $output_dir/step3/
 173
 174 echo "[rebalance-zone-expansion] Placing final-cluster.xml in '$output_dir'"
 175 cp $output_dir/step3/final-cluster.xml $output_dir/final-cluster.xml
 176 echo "[rebalance-zone-expansion] Placing plan.out in '$output_dir'"
 177 cp $output_dir/step3/plan.out $output_dir/plan.out
 178 echo "[rebalance-zone-expansion] Placing final-cluster.xml.analysis in '$output_dir'"
 179 cp $output_dir/step2/final-cluster.xml $output_dir/final-cluster.xml.analysis