path: root/src/ns3/simulate.org

* Run simulations
  To run all the simulations, execute the following call:
  #+NAME: runSim
  #+CALL: runBW(lat=runLat(nbSens=runNbSensors(nbHop=runNbHop())))

  #+RESULTS: runSim

** Experiments
   
*** Sensors Position
    #+NAME: runSensorsPos
    #+BEGIN_SRC bash :noweb yes :results output
      <<singleRun>>
      simKey="SENSORSPOS"
      for sensorsNumber in $(seq 5 2 15)
      do
          for positionSeed in $(seq 1 10)
          do
              run
          done      
      done
    #+END_SRC

*** Sensors Send Interval
    #+NAME: runSendInterval
    #+BEGIN_SRC bash :noweb yes :results output
      <<singleRun>>
      simKey="SENDINTERVAL"
      for sensorsNumber in $(seq 5 2 15)
      do
          for sensorsSendInterval in $(seq 10 10 100)
          do
              run
          done      
      done
    #+END_SRC

*** Bandwidth
    #+NAME: runBW
    #+BEGIN_SRC bash :noweb yes :results output
      <<singleRun>>
      simKey="BW"
      for sensorsNumber in $(seq 1 10)
      do
          for linksBandwidth in $(seq 10 20 100)
          do
              run
          done      
      done
    #+END_SRC

    #+RESULTS: runBW

    #+RESULTS:

*** Latency
    #+NAME: runLat
    #+BEGIN_SRC bash :noweb yes :results output
      <<singleRun>>
      simKey="LATENCY"
      for sensorsNumber in $(seq 1 10)
      do
          for linksLatency in $(seq 1 1 10)
          do
              run
          done      
      done
    #+END_SRC

    #+RESULTS: runLat

    #+RESULTS:
    
*** Number of sensors
    #+NAME: runNbSensors
    #+BEGIN_SRC bash :noweb yes :results output
      <<singleRun>>
      simKey="NBSENSORS"
      for sensorsNumber in $(seq 1 15)
      do
          run
      done      
    #+END_SRC

    #+RESULTS:

*** Number of Hop
    #+NAME: runNbHop
    #+BEGIN_SRC bash :noweb yes :results output
      <<singleRun>>
      simKey="NBHOP"
      for sensorsNumber in $(seq 1 10)
      do
          for nbHop in $(seq 1 10)
          do
              run
          done
      done
    #+END_SRC

    #+RESULTS: runNbHop


** Single Run
   
   #+NAME: singleRun
   #+BEGIN_SRC bash :eval never :noweb yes :results output
     simulator="simulator/simulator"
     parseEnergyScript="./parseEnergy.awk"
     parseDelayScript="./parseDelay.awk"
     logFolder="logs/"
     export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:${NS3_PATH}/build/lib

     # Default Parameters
     sensorsSendInterval=10 # DON'T GO BELOW 1 SECONDS !!!!!!! Simulator will stay stuck
     sensorsPktSize=192 # 1 byte temperature (-128 à +128 °C) and 4Byte sensorsId
     sensorsNumber=5
     nbHop=10 # Cf paper AC/Yunbo
     linksBandwidth=10000 # 10Ge links (to be coherent with energy values of ECOFEN/literature)
     linksLatency=11
     positionSeed=5
     simKey="NOKEY"

     run () {
         # If another function want to handle simulation (tipically used on g5k)
         type -t handleSim > /dev/null && { handleSim; return; }

         local logFile="${logFolder}/${simKey}_${sensorsSendInterval}SSI_${sensorsPktSize}SPS_${sensorsNumber}SN_${nbHop}NH_${linksBandwidth}LB_${linksLatency}LL_${positionSeed}PS.org"
         [ -f "$logFile" ] && return
         local simCMD="$simulator --sensorsSendInterval=${sensorsSendInterval} --sensorsPktSize=${sensorsPktSize} --sensorsNumber=${sensorsNumber} --nbHop=${nbHop} --linksBandwidth=${linksBandwidth} --linksLatency=${linksLatency} --positionSeed=${positionSeed} 2>&1"
         local log=$(bash -c "$simCMD")

         # Compute some metrics
         energyLog=$(echo "$log" | $parseEnergyScript)
         avgDelay=$(echo "$log" | $parseDelayScript)
         totalEnergy=$(echo "$energyLog" | awk 'BEGIN{power=0;FS=","}NR!=1{power+=$2}END{print(power)}')
         sensorsEnergy=$(echo "$energyLog" |awk -F',' 'BEGIN{sumW=0}$1<0{sumW+=$2}END{print sumW}')
         networkEnergy=$(echo "$energyLog" |awk -F',' 'BEGIN{sumN=0}$1>=0{sumN+=$2}END{print sumN}')
         nbPacketCloud=$(echo "$log"|grep -c "CloudSwitch receive")
         nbNodes=$(echo "$log"|awk '/Simulation used/{print($3)}')
         ns3Version=$(echo "$log"|awk '/NS-3 Version/{print($3)}')

         # Save logs
         echo -e "#+TITLE: $(date) ns-3 (version ${ns3Version}) simulation\n" > $logFile
         echo "* Environment Variables" >> $logFile
         env >> $logFile
         echo "* Full Command" >> $logFile
         echo "$simCMD" >> $logFile
         echo "* Output" >> $logFile
         echo "$log" >> $logFile
         echo "* Energy CSV (negative nodeId = WIFI, 0 = AP (Wireless+Wired), positive nodeId = ECOFEN" >> $logFile
         echo "$energyLog" >> $logFile
         echo "* Metrics" >> $logFile
         echo "-METRICSLINE- sensorsSendInterval:${sensorsSendInterval} sensorsPktSize:${sensorsPktSize} sensorsNumber:${sensorsNumber} nbHop:${nbHop} linksBandwidth:${linksBandwidth} linksLatency:${linksLatency} totalEnergy:$totalEnergy nbPacketCloud:$nbPacketCloud nbNodes:$nbNodes avgDelay:${avgDelay} ns3Version:${ns3Version} simKey:${simKey} positionSeed:${positionSeed} sensorsEnergy:${sensorsEnergy} networkEnergy:${networkEnergy}" >> $logFile
     }

   #+END_SRC

** Grid 5000
*** Master Node Script
    This code generate and distribute simulation argument to the slave worker nodes and
    run start their simulations processes:
    #+BEGIN_SRC sh :tangle ./g5k-root.sh :shebang "#!/bin/bash" :noweb yes
      ##### Arguments #####
      nHost=20 # At least 20 host x)
      nProcesses=3 # Max number of parrallel simulations (don't go too high, your process will be killed (arround 8))
      nHours=4 # Reservation duration
      simArgsLoc=~/args/ # Don't change this path witouth changing it in workder scripts
      finishedFile="$simArgsLoc/finished-microBenchmarks.txt"
      logsFinalDst=~/logs/
      #####################

      # Check
      [ "$1" == "subscribe" ] && subscribe=1 ||subscribe=0
      [ "$1" == "deploy" ] && deploy=1 || deploy=0
      [ "$1" == "-p" ] && progress=1 || progress=0

      handleSim () {
          [ -z "${argId}" ] && argId=1 || argId=$(( argId + 1 ))
          outF="$simArgsLoc/${argId}.sh" # Args file based on host name (avoid conflict)

          # Add Shebang
          echo '#!/bin/bash' > $outF
          echo "finishedFile=\"$finishedFile\"" >> $outF
          echo "nProcesses=$nProcesses" >> $outF
          echo "logsFinalDst=\"$logsFinalDst\"" >> $outF
          # Save arguments
          echo "sensorsSendInterval=${sensorsSendInterval}" >> $outF
          echo "sensorsPktSize=${sensorsPktSize}" >> $outF
          echo "nbHop=${nbHop}" >> $outF
          echo "simKey=\"${simKey}\"" >> $outF
          echo "sensorsNumber=${sensorsNumber}" >> $outF
          echo "linksLatency=${linksLatency}" >> $outF
          echo "sensorsNumber=${sensorsNumber}" >> $outF
          echo "linksBandwidth=${linksBandwidth}" >> $outF
          echo "positionSeed=${positionSeed}" >> $outF
      }


      # Start subscribe/deploy
      if [ $subscribe -eq 1 ]
      then
          echo "Starting oarsub..."
          oarsub -l host=$nHost,walltime=$nHours 'sleep "10d"' # Start reservation
          echo "Please join your node manually when your reservation is ready by using oarsub -C <job-id>"
          exit 0
      elif [ $deploy -eq 1 ]
      then
          echo "Starting deployment..."

          ##### Usefull Variables #####
          wai=$(dirname "$(readlink -f $0)") # Where Am I ?
          hostList=($(cat $OAR_NODE_FILE | uniq))
          #############################

          # Initialize logsFinalDst
          mkdir -p $logsFinalDst
          rm -rf $logsFinalDst/* # Clean log dst just in case (it is dangerous but avoid conflict)
          mkdir -p $simArgsLoc
          rm -rf $simArgsLoc/* # Clean old args

          # Add your simulation code block here
          <<runNbSensors>>
          <<runSendInterval>>
          <<runSensorsPos>>

          # Distribute argument according to subsribed nodes
          cd $simArgsLoc
          curHostId=0
          for file in $(find ./ -type f)
          do
              [ $curHostId -eq $nHost ] && curHostId=0  
              mv -- ${file} ${hostList[$curHostId]}-$(basename ${file})
              curHostId=$(( curHostId + 1 ))
          done
          cd -


          # Run simulations
          echo "Host who finished their work:" > $finishedFile
          for host in ${hostList[@]}
          do
              echo "Start simulations on node $host"
              oarsh lguegan@$host bash g5k-worker.sh &
          done

          exit 0
      elif [ $progress -eq 1 ]
      then
          alreadyFinished=$(cat $finishedFile| tail -n +2| wc -l)
          percent=$(echo $alreadyFinished $nHost| awk '{print $1/$2*100}')
          echo "Progression: " $alreadyFinished/$nHost "(${percent}%)"
      else
          echo "Invalid arguments, make sure you know what you are doing !"
          exit 1
      fi
    #+END_SRC
*** Worker Node Script
    Almost like the [[microBenchmarksSingle][single run script]] but with additionnal code to handle g5k simulation platform (arguments,logs etc..).
    #+BEGIN_SRC sh :tangle ./g5k-worker.sh :shebang "#!/bin/bash" :noweb yes
      export NS3_PATH=~/.bin/ns-3/ns-3.29/
      g5kLogFolder="/tmp/logs/"
      mkdir -p $g5kLogFolder # Create log folder just in case
      rm -rf $g5kLogFolder/* # Clean previous logs just in case

      hostname=$(hostname)

      # Run simulations with sourced arguments :D
      simArgsLoc=~/args/ # Don't change this path without changing it in root scripts

      argsId=0
      argsFile="$simArgsLoc/${hostname}-args-${argsId}.sh" # Arguments generated by Root Node
      curNProcesses=0 # Start with no processes
      for argsFile in $(find $simArgsLoc -type f -name "$hostname*")
      do
          <<singleRun>>
          logFolder=$g5kLogFolder # Don't forget override default g5kLogFolder
          source  $argsFile # Fetch argument
          run & # Run async 
          ((curNProcesses+=1)) # Increase by 2           
          [ $curNProcesses -ge $nProcesses ] && { curNProcesses=0; wait; }
      done
      wait # Wait until the end of all simulations

      cp -r $g5kLogFolder/* "$logsFinalDst" # Fetch log from tmp into nfs
      echo $(hostname) >> $finishedFile # Just say I finished
    #+END_SRC

* Logs Analysis
  To Generate all the plots, please execute the following line:
  #+NAME: runAnalysis
  #+CALL: plotToPDF(plots=genAllPlots(data=logToCSV()))

  #+RESULTS: runAnalysis

** R Scripts
*** Generate all plots script
    Available variables:
    |---------------------|
    | Name                |
    |---------------------|
    | sensorsSendInterval |
    | sensorsPktSize      |
    | sensorsNumber       |
    | nbHop               |
    | linksBandwidth      |
    | linksLatency        |
    | totalEnergy         |
    | nbPacketCloud       |
    | nbNodes             |
    | avgDelay            |
    | simKey              |
    |---------------------|

    #+NAME: genAllPlots
    #+BEGIN_SRC R :noweb yes :results output
      <<RUtils>>
#      easyPlotGroup("linksLatency","totalEnergy", "LATENCY","sensorsNumber")
#      easyPlotGroup("linksBandwidth","totalEnergy", "BW","sensorsNumber")
      easyPlot("sensorsNumber","totalEnergy", "NBSENSORS")
      easyPlotGroup("positionSeed", "totalEnergy","SENSORSPOS","sensorsNumber")
      easyPlotGroup("positionSeed", "avgDelay","SENSORSPOS","sensorsNumber")
      easyPlotGroup("sensorsSendInterval","sensorsEnergy","SENDINTERVAL","sensorsNumber")
      easyPlotGroup("sensorsSendInterval","networkEnergy","SENDINTERVAL","sensorsNumber")

    #+END_SRC

    #+RESULTS: genAllPlots

*** R Utils
    RUtils is intended to load logs (data.csv) and providing
    simple plot function for them.

    #+NAME: RUtils
    #+BEGIN_SRC R :eval never
      library("tidyverse")

      # Fell free to update the following
      labels=c(nbNodes="Number of nodes",sensorsNumber="Number of sensors",totalEnergy="Total Energy (J)",
        nbHop="Number of hop (AP to Cloud)", linksBandwidth="Links Bandwidth (Mbps)", avgDelay="Average Application Delay (s)",
        linksLatency="Links Latency (ms)", sensorsSendInterval="Sensors Send Interval (s)", positionSeed="Position Seed",
        sensorsEnergy="Sensors Wifi Energy Consumption (J)", networkEnergy="Network Energy Consumption (J)")

      # Load Data
      data=read_csv("logs/data.csv")

      # Get label according to varName
      getLabel=function(varName){
        if(is.na(labels[varName])){
          return(varName)
        }
        return(labels[varName])
      }

      easyPlot=function(X,Y,KEY){
        curData=data%>%filter(simKey==KEY)
        stopifnot(NROW(curData)>0)
        ggplot(curData,aes_string(x=X,y=Y))+geom_point()+geom_line()+xlab(getLabel(X))+ylab(getLabel(Y))
        ggsave(paste0("plots/",KEY,"-",X,"_",Y,".png"))
      }

      easyPlotGroup=function(X,Y,KEY,GRP){
        curData=data%>%filter(simKey==KEY) %>% mutate(!!GRP:=as.character(UQ(rlang::sym(GRP)))) # %>%mutate(sensorsNumber=as.character(sensorsNumber))
        stopifnot(NROW(curData)>0)
        ggplot(curData,aes_string(x=X,y=Y,color=GRP,group=GRP))+geom_point()+geom_line()+xlab(getLabel(X))+ylab(getLabel(Y)) + labs(color = getLabel(GRP))
        ggsave(paste0("plots/",KEY,"-",X,"_",Y,".png"))
      }
    #+END_SRC

** Plots -> PDF
   Merge all plots in plots/ folder into a pdf file.
   #+NAME: plotToPDF
   #+BEGIN_SRC bash :results output :noweb yes
     orgFile="plots/plots.org"
     <<singleRun>> # To get all default arguments

     # Write helper function
     function write {
         echo "$1" >> $orgFile
     }

     echo "#+TITLE: Analysis" > $orgFile
     write "#+LATEX_HEADER: \usepackage{fullpage}"
     write "#+OPTIONS: toc:nil"
     # Default arguments
     write '\begin{center}'
     write '\begin{tabular}{lr}'
     write 'Parameters & Values\\'
     write '\hline'
     write "sensorsPktSize & ${sensorsPktSize} bytes\\\\"
     write "sensorsSendInterval & ${sensorsSendInterval}s\\\\"
     write "sensorsNumber & ${sensorsNumber}\\\\"
     write "nbHop & ${nbHop}\\\\"
     write "linksBandwidth & ${linksBandwidth}Mbps\\\\"
     write "linksLatency & ${linksLatency}ms\\\\"
     write '\end{tabular}'
     write '\newline'
     write '\end{center}'

     for plot in $(find plots/ -type f -name "*.png")
     do
         write "\includegraphics[width=0.5\linewidth]{$(basename ${plot})}"
     done

     # Export to pdf
     emacs $orgFile --batch -f org-latex-export-to-pdf --kill  
   #+END_SRC

   #+RESULTS:

** Log -> CSV
   logToCSV extract usefull data from logs and put them into logs/data.csv.

   #+NAME: logToCSV
   #+BEGIN_SRC bash :results none
     csvOutput="logs/data.csv"

     # First save csv header line
     aLog=$(find logs/ -type f -name "*.org"|head -n 1)
     metrics=$(cat $aLog|grep "\-METRICSLINE\-"|sed "s/-METRICSLINE-//g")
     echo $metrics | awk '{for(i=1;i<=NF;i++){split($i,elem,":");printf(elem[1]);if(i<NF)printf(",");else{print("")}}}' > $csvOutput

     # Second save all values
     for logFile in $(find logs/ -type f -name "*.org")
     do
         metrics=$(cat $logFile|grep "\-METRICSLINE\-"|sed "s/-METRICSLINE-//g")
         echo $metrics | awk '{for(i=1;i<=NF;i++){split($i,elem,":");printf(elem[2]);if(i<NF)printf(",");else{print("")}}}' >> $csvOutput
     done
   #+END_SRC



   
   
   
** Custom Plots

   #+NAME: ssiNet
   #+BEGIN_SRC R :noweb yes :results graphics :file plots/sensorsSendInterval-net.png
     <<RUtils>>

     data%>%filter(simKey=="SENDINTERVAL",sensorsNumber==20) %>% ggplot(aes(x=sensorsSendInterval,y=networkEnergy))+xlab(getLabel("sensorsSendInterval"))+ylab(getLabel("networkEnergy"))+
     geom_line()+labs(title="For 20 sensors")
     ggsave("plots/sensorsSendInterval-net.png",dpi=80)
   #+END_SRC

   #+RESULTS:
   [[file:plots/sensorsSendInterval-net.png]]

   
   #+NAME: ssiWifi
   #+BEGIN_SRC R :noweb yes :results graphics :file plots/sensorsSendInterval-wifi.png
     <<RUtils>>
     data%>%filter(simKey=="SENDINTERVAL",sensorsNumber==20) %>% ggplot(aes(x=sensorsSendInterval,y=sensorsEnergy))+xlab(getLabel("sensorsSendInterval"))+ylab(getLabel("sensorsEnergy"))+
     geom_line() +     geom_line()+labs(title="For 20 sensors")
     ggsave("plots/sensorsSendInterval-wifi.png",dpi=80)
   #+END_SRC

   #+RESULTS: ssiWifi
   [[file:plots/sensorsSendInterval-wifi.png]]

   #+RESULTS:
   [[file:plots/sensorsSendInterval.png]]






    #+BEGIN_SRC R :results graphics  :noweb yes  :file plot-final.png :session *R*
      <<RUtils>>
      simTime=1800

      # Network
      data=read_csv("logs/data.csv")
      data=data%>%filter(simKey=="NBSENSORS")
      dataC5=data%>%filter(sensorsNumber==5)%>% mutate(energy=networkEnergy/simTime) %>%select(energy,sensorsNumber)
      dataC10=data%>%filter(sensorsNumber==10)%>%mutate(energy=networkEnergy/simTime) %>%select(energy,sensorsNumber)
      dataNet=rbind(dataC5,dataC10)%>%mutate(type="Network")

      # Network
      dataS5=data%>%filter(sensorsNumber==5)%>% mutate(energy=sensorsEnergy/simTime) %>%select(energy,sensorsNumber) 
      dataS10=data%>%filter(sensorsNumber==10)%>%mutate(energy=sensorsEnergy/simTime) %>%select(energy,sensorsNumber) 
      dataS=rbind(dataS5,dataS10)%>%mutate(type="Sensors")
      
      data=rbind(dataNet,dataS)%>%mutate(sensorsNumber=as.character(sensorsNumber))

      ggplot(data)+geom_bar(aes(x=sensorsNumber,y=energy,fill=type),stat="identity")+xlab("Sensors Number")+ylab("Power Consumption (W)")+guides(fill=guide_legend(title="Part"))
      ggsave("plot-final.png",dpi=80)

    #+END_SRC

    #+RESULTS:
    [[file:plot-final.png]]