diff --git a/tools/README.md b/tools/README.md index 0c1d5113b..bebc6333c 100644 --- a/tools/README.md +++ b/tools/README.md @@ -1,73 +1,105 @@ -### Spark Eventlog Analyzer +# Spark Eventlog Analyzer The pyspark script to analyze Gazelle's eventlog ## Prequisites -# Jupyter Installation +### Jupyter Installation ``` pip3 install jupyter ``` -# Notebook Installation +### Notebook Installation ``` pip3 install notebook ``` -# iPyKernel +### iPyKernel ``` pip3 install ipykernel ``` -# FindSpark +### FindSpark ``` pip3 install findspark ``` -# Matplotlib +### Matplotlib ``` pip3 install matplotlib ``` -# Seaborn +### Seaborn ``` pip3 install seaborn ``` -# Pandasql +### Pandasql ``` pip3 install pandasql ``` -# PyHDFS +### PyHDFS ``` pip3 install pyhdfs ``` -# PyArrow +### PyArrow ``` pip3 install pyarrow ``` -### Put below two .ipynb in the jupyter root directory +# Eventlog Analyzer Tools +The eventlog analyzer includes sparklog.ipynb and gazelle_analysis.ipynb. +Please put them into jupyter directory. ## sparklog.ipynb sparklog.ipynb is the function definition for spark eventlog analyzer ## gazelle_analysis.ipynb gazelle_analysis is the main program to call sparklog.ipynb and load the eventlog from hdfs. -###How it works: +##How it works: Launch gazelle_analysis.ipynb as main script -###Parameters: +##Parameters: In Analysis:generate_trace_view, the url for display. In App_Log_Analysis:get_basic_state, the url for display. In App_Log_Analysis:get_app_info, the url for display. In show_rst function, the url for html. In pyhdfs, the url for HDFS hosts. - - -###To run in in commandline: +##To run in in commandline: jupyter nbconvert --execute --to notebook --inplace --allow-errors --ExecutePreprocessor.timeout=-1 ./gazelle_analysis.ipynb --template classic -###To convert into HTML: +##To convert into HTML: jupyter nbconvert --to html ./gazelle_analysis.ipynb --output ./gazelle_analysis.html --template classic + +# Tools to collect sar information and generation trace view(.json): +You can also use below files to collect sar information. +The purpose for this tool is to generate a json file with sar information. +After the json has been generated, you can use catapult to view your json file. + +monitor.py: main program to collect sar information, it must be started before your application and stopped after your application. +post_process.sh: Post process the sar files after the monitor stop. +run_example.sh: An example to teach you how to use monitor.py with your application. +template.ipynb: A template to use for generating trace view(.json). + +## Usage +Before run the tool, please make sure to set up some settings including +In monitor.py, +clients:the nodes in your cluster. +base_dir: the base directory name to put logs. +local_profile_dir: the local location to put logs. +hdfs_address: the hdfs address to copy all the logs to hdfs. + +In sparklog.ipynb, +Please replace sr124 to the master in your cluster and use to process the logs. +Please replace sr525 to the catapult server. + +You can check run_example.sh to see how to use the script to collect sar information. +Please add below command in your script: +``` +appid=`yarn application -list 2>&1 | tail -n 1 | awk -F"\t" '{print $1}'` +rm -f log/memory*.csv +python3 ./monitor.py start $appid +$run_your_query +python3 ./monitor.py stop $appid "spark_logs" +``` diff --git a/tools/monitor.py b/tools/monitor.py new file mode 100644 index 000000000..2816de488 --- /dev/null +++ b/tools/monitor.py @@ -0,0 +1,146 @@ +import os +import pyhdfs +import subprocess +import sys +import time + +from inspect import currentframe, getframeinfo +from pathlib import Path + +clients=['sr124'] +home = str(Path.home()) +base_dir = 'profile' +local_profile_dir=home+"/"+base_dir +hdfs_address='10.1.0.24:50070' + +def killsar(): + for l in clients: + try: + cmd="ssh "+l+" ps aux | grep -w sar | grep -v grep | tr -s ' ' | cut -d' ' -f2" + out=subprocess.check_output(cmd).decode('ascii').strip().split("\n") + for x in out: + cmd="ssh "+l+" kill "+x+" > /dev/null 2>&1" + subprocess.call(cmd,shell=True) + except Exception as e: + print(e) + pass + for l in clients: + try: + cmd="ssh "+l+" ps aux | grep -w pidstat | grep -v grep | tr -s ' ' | cut -d' ' -f2" + out=subprocess.check_output(cmd, shell=True).decode('ascii').strip().split("\n") + for x in out: + cmd="ssh "+l+" kill "+x+" > /dev/null 2>&1" + subprocess.call(cmd,shell=True) + except Exception as e: + print(e) + pass + for l in clients: + try: + cmd="ssh "+l+" ps aux | grep -w perf | grep -v grep | tr -s ' ' | cut -d' ' -f2" + out=subprocess.check_output(cmd,shell=True).decode('ascii').strip().split("\n") + except Exception as e: + print(e) + pass + +def startmonitor(appid, **kwargs): + print("[monitor.py]Starting system monitoring ...") + print(clients) + appid_profile_dir=local_profile_dir+"/"+appid + cmd="mkdir -p "+appid_profile_dir + print("Launching CMD create application id profile dir: %s" % cmd) + subprocess.call(cmd,shell=True) + + for l in clients: + cmd="ssh "+l+" date" + print(subprocess.check_output(cmd,shell=True).decode('ascii')) + + killsar() + + for l in clients: + print("[monitor.py]create profile directory") + client_profile_dir=appid_profile_dir+"/"+l + cmd="mkdir -p "+client_profile_dir + print("[monitor.py]Launching CMD create server profile dir: %s" % cmd) + subprocess.call(cmd,shell=True) + cmd="ssh "+l+" mkdir -p "+client_profile_dir + print("[monitor.py]Launching CMD create client profile dir: %s" % cmd) + subprocess.call(cmd,shell=True) + cmd="ssh "+l+" sar -o "+client_profile_dir+"/sar.bin -r -u -d -B -n DEV 1 >/dev/null 2>&1 &" + print("[monitor.py]Launching CMD create sar.bin file: %s" % cmd) + subprocess.call(cmd,shell=True) + if kwargs.get("collect_pid",False): + cmd="ssh "+l+" jps | grep CoarseGrainedExecutorBackend | head -n 1 | cut -d' ' -f 1 | xargs -I % pidstat -h -t -p % 1 > "+client_profile_dir+"/pidstat.out 2>/dev/null &" + print("Launching CMD collect pid: %s" % cmd) + subprocess.call(cmd,shell=True) + return appid_profile_dir + +def stopmonitor(appid, eventlogdir, basedif): + + appid_profile_dir=local_profile_dir+"/"+appid + cmd="mkdir -p "+appid_profile_dir + print("Launching CMD create application id profile dir: %s" % cmd) + subprocess.call(cmd,shell=True) + + killsar() + + with open("%s/starttime" % appid_profile_dir,"w") as f: + f.write("{:d}".format(int(time.time()*1000))) + + hadoophome=os.environ["HADOOP_HOME"] + userlogdir="/opt/hadoop/yarn/logs" + + for l in clients: + client_profile_dir=appid_profile_dir+"/"+l + cmd="ssh "+l+" sar -f "+client_profile_dir+"/sar.bin -r > "+client_profile_dir+"/sar_mem.sar;sar -f "+client_profile_dir+"/sar.bin -u > "+client_profile_dir+"/sar_cpu.sar;sar -f "+client_profile_dir+"/sar.bin -d > "+client_profile_dir+"/sar_disk.sar;sar -f "+client_profile_dir+"/sar.bin -n DEV > "+client_profile_dir+"/sar_nic.sar;sar -f "+client_profile_dir+"/sar.bin -B > "+client_profile_dir+"/sar_page.sar;" + print("Launching CMD: %s" % cmd) + subprocess.call(cmd,shell=True) + cmd="ssh "+l+" grep -rI xgbtck --no-filename "+userlogdir+"/"+appid+"/* | sed 's/^ //g' > "+client_profile_dir+"/xgbtck.txt" + print("Launching CMD: %s" % cmd) + subprocess.call(cmd,shell=True) + cmd="scp -r "+l+":"+client_profile_dir+" "+appid_profile_dir+"/ > /dev/null 2>&1" + print("Launching CMD: %s" % cmd) + subprocess.call(cmd, shell=True) + cmd="ssh "+l+" jps | grep CoarseGrainedExecutorBackend | head -n 2 | tail -n 1 | cut -d' ' -f 1 | xargs -I % ps -To tid p % > "+client_profile_dir+"/sched_threads.txt" + subprocess.call(cmd, shell=True) + cmd="ssh "+l+" sar -V > "+client_profile_dir+"/sarv.txt" + print("Launching CMD: %s" % cmd) + subprocess.call(cmd, shell=True) + cmd="test -f "+client_profile_dir+"/perfstat.txt && head -n 1 "+client_profile_dir+"/perfstat.txt > "+client_profile_dir+"/perfstarttime" + print("Launching CMD: %s" % cmd) + subprocess.call(cmd,shell=True) + + logfile=eventlogdir+"/"+appid + cmd="hadoop fs -copyToLocal "+logfile+" "+appid_profile_dir+"/app.log" + print("Launching CMD hadoop fs copytolocal: %s" % cmd) + subprocess.call(cmd,shell=True) + + fs = pyhdfs.HdfsClient(hosts=hdfs_address, user_name='root') + + print("Launching CMD hadoop fs -mkdir /%s" % basedif) + fs.mkdirs("/" + basedif + "/") + v=[os.path.join(dp, f) for dp, dn, fn in os.walk(os.path.expanduser(local_profile_dir+"/"+appid)) for f in fn] + for f in v: + paths=os.path.split(f) + fs.mkdirs("/"+ basedif + paths[0][len(local_profile_dir):]) + fs.copy_from_local(f,"/"+ basedif + paths[0][len(local_profile_dir):]+"/"+paths[1],overwrite=True) + + +if __name__ == '__main__': + if sys.argv[1]=="start": + startmonitor( sys.argv[2]) + elif sys.argv[1]=="stop": + import datetime + from datetime import date + basedir=base_dir+"/"+date.today().strftime("%Y_%m_%d") + stopmonitor( sys.argv[2],sys.argv[3],basedir) + + lastnightrun=["","",""] + with open("log/runs.txt") as f: + for l in f.readlines(): + x=l.strip().split(" ") + if ( x[0]=="05" and x[2]!=sys.argv[2] ) or ( len(sys.argv)==5 and x[2]==sys.argv[4] ): + lastnightrun[0]=x[0] + lastnightrun[1]=x[1] + lastnightrun[2]=x[2] + os.system(("./post_process.sh {} {} {} {}").format(date.today().strftime("%Y_%m_%d"), sys.argv[2], lastnightrun[1],lastnightrun[2])) + diff --git a/tools/post_process.sh b/tools/post_process.sh new file mode 100755 index 000000000..7506004e0 --- /dev/null +++ b/tools/post_process.sh @@ -0,0 +1,35 @@ +# /bin/sh +echo -e "Starting Post Processing ..." +echo -e "Start notebook processing - template.ipynb" +sed 's/BASEDIR_TEMP/profile\/'$1'/g' template.ipynb > tpcxx_$1_$2.ipynb +sed -i 's/APPID_TEMP/'$2'/g' tpcxx_$1_$2.ipynb +if [ $# -eq 4 ] +then + sed -i 's/LAST_BASEDIR/profile\/'$3'/g' tpcxx_$1_$2.ipynb + sed -i 's/LAST_APPID/'$4'/g' tpcxx_$1_$2.ipynb +fi + +hadoop fs -mkdir /history +hadoop fs -cp /profile/$1/$2/app.log /history/$2 +echo -e "Finish notebook processing - template.ipynb" + +echo -e "Start notebook execution - tpcxx.ipynb" +mkdir -p html +jupyter nbconvert --execute --to notebook --inplace --allow-errors --ExecutePreprocessor.timeout=-1 ./tpcxx_$1_$2.ipynb --template classic +jupyter nbconvert --to html ./tpcxx_$1_$2.ipynb --output html/tpcxx_$1_$2.html --template classic + +#echo -e "notebook processing - tpch_summary.ipynb" +#sed 's/BASEDIR_TEMP/profile\/'$1'/g' tpch_template_summary.ipynb > tpch_summary_$1_$2.ipynb +#sed -i 's/APPID_TEMP/'$2'/g' tpch_summary_$1_$2.ipynb +#if [ $# -eq 4 ] +#then +# sed -i 's/LAST_BASEDIR/profile\/'$3'/g' tpch_summary_$1_$2.ipynb +# sed -i 's/LAST_APPID/'$4'/g' tpch_summary_$1_$2.ipynb +#fi + +#jupyter nbconvert --execute --to notebook --inplace --allow-errors --ExecutePreprocessor.timeout=-1 ./tpch_summary_$1_$2.ipynb --template classic +#jupyter nbconvert --to html --no-input ./tpch_summary_$1_$2.ipynb --output html/tpch_summary_$1_$2.html --template classic +#echo -e "Finish notebook processing - tpch.ipynb" + +#rm -rf ./tpch_summary_$1_$2.ipynb +echo -e "Finish Post Processing !!!" diff --git a/tools/run_example.sh b/tools/run_example.sh new file mode 100755 index 000000000..646339d82 --- /dev/null +++ b/tools/run_example.sh @@ -0,0 +1,41 @@ +#!/bin/bash + +#/home/spark-sql/collect_sar.sh +eventlogdir="spark_logs" + +echo "Stoping Thrift Server ..." +./run_spark_thrift_server.sh stop +echo "Done" +sleep 20 +echo "Cleaning Cache ..." +./clean_cache.sh +echo "Done" +sleep 1 +echo "Starting Thrift Server ..." +./run_spark_thrift_server.sh start +echo "Done" +sleep 40 +echo "Start Resource Monitoring ..." +appid=`yarn application -list 2>&1 | tail -n 1 | awk -F"\t" '{print $1}'` +echo `date +'%H %Y_%m_%d'` $appid ${1} >> log/runs.txt +rm -f log/memory*.csv +python3 ./monitor.py start $appid +echo "Running TPCH Query" +./run_tpch.py 2>&1 >> tpch_query.log | tee -a tpch_query.txt +echo "Done" +sleep 1 +echo "Stop Thrift Server" +./run_spark_thrift_server.sh stop +sleep 10 +python3 ./monitor.py stop $appid "spark_logs" +echo '' > log/link.html +echo 'history event: http://10.1.0.24:18080/history/'$appid'/jobs/
' >> log/link.html +echo 'history on sr124: http://10.1.0.24:18080/history/'$appid'/jobs/
' >> log/link.html +echo 'notebook on sr124: http://10.1.0.24:8888/notebooks/jenkins/tpch_'`date +'%Y_%m_%d'`'_'$appid'.ipynb
' >> log/link.html +echo 'notebook html on sr124: http://10.1.0.24:8888/notebooks/jenkins/html/tpch_'`date +'%Y_%m_%d'`'_'$appid'.html
' >> log/link.html +echo 'traceview on sr124: http://10.1.0.24:1088/tracing_examples/trace_viewer.html#/tracing/test_data/'$appid'.json
' >> log/link.html + +echo "
" >> log/link.html + +echo "All Jobs Are Done." +#/home/spark-sql/stop_sar.sh diff --git a/tools/template.ipynb b/tools/template.ipynb new file mode 100644 index 000000000..87b4d351b --- /dev/null +++ b/tools/template.ipynb @@ -0,0 +1,456 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Imports" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from __future__ import nested_scopes\n", + "from IPython.core.display import display, HTML\n", + "display(HTML(\"\"))\n", + "display(HTML(''))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import logging\n", + "logger = logging.getLogger()\n", + "logger.setLevel(logging.ERROR)\n", + "\n", + "import warnings\n", + "warnings.filterwarnings('ignore')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import findspark\n", + "findspark.init()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import re\n", + "import os\n", + "import pandas\n", + "pandas.set_option('display.max_rows', None)\n", + "\n", + "import matplotlib\n", + "%matplotlib inline\n", + "import matplotlib.pyplot as plt\n", + "import matplotlib.ticker as mtick\n", + "from matplotlib import colors\n", + "from matplotlib import rcParams\n", + "rcParams['font.sans-serif'] = 'Courier New'\n", + "rcParams['font.family'] = 'Courier New'\n", + "rcParams['font.size'] = '12'\n", + "%matplotlib inline\n", + "\n", + "from IPython.display import display,HTML\n", + "import threading\n", + "import collections\n", + "\n", + "from IPython.display import display\n", + "from ipywidgets import IntProgress,Layout\n", + "import time\n", + "import threading\n", + "import gzip\n", + "import pyspark\n", + "import pyspark.sql\n", + "from pyspark.sql import SparkSession\n", + "from pyspark.sql.types import (StructType, StructField, DateType,\n", + " TimestampType, StringType, LongType, IntegerType, DoubleType,FloatType)\n", + "from pyspark.sql.functions import to_date, floor\n", + "from pyspark.ml.feature import StringIndexer, VectorAssembler\n", + "from pyspark.ml import Pipeline\n", + "from pyspark.sql.functions import lit\n", + "import datetime\n", + "import time\n", + "from pyspark.storagelevel import StorageLevel\n", + "from pyspark.sql.window import Window\n", + "from pyspark.sql.functions import rank, col\n", + "from pyspark.ml import Pipeline\n", + "import pandas\n", + "import numpy\n", + "\n", + "\n", + "import re\n", + "import math\n", + "from functools import reduce\n", + "import json\n", + "\n", + "\n", + "from pyspark.sql.types import *\n", + "from pyspark.sql import functions as F\n", + "from datetime import date" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# start analysis cluster and run" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "code_folding": [] + }, + "outputs": [], + "source": [ + "executors_per_node = 5\n", + "nodes=3\n", + "cores_per_executor=4\n", + "task_per_core=4\n", + "\n", + "\n", + "cache_size=20\n", + "total_size=100000\n", + "print('executor per node: {:d}\\nparallelism: {:d}\\nmemory: {:d}m\\noffheap:{:d}m'.format(executors_per_node,nodes*executors_per_node*cores_per_executor*task_per_core,int(math.floor(nodes*total_size/(nodes*executors_per_node)))-1024-int(math.floor(cache_size*1024/(nodes*executors_per_node))),int(math.floor(cache_size*1024/(nodes*executors_per_node)))))\n", + "\n", + "from pyspark import SparkConf, SparkContext\n", + "from pyspark.sql import SQLContext\n", + "conf = (SparkConf()\n", + " .set(\"spark.default.parallelism\", \"{:d}\".format(nodes*executors_per_node))\n", + " .set('spark.executor.instances', '{:d}'.format(nodes*executors_per_node))\n", + " .set('spark.sql.files.maxPartitionBytes', '256m')\n", + " .set('spark.app.name', 'pyspark_final')\n", + " .set('spark.rdd.compress', 'False')\n", + " .set('spark.serializer','org.apache.spark.serializer.KryoSerializer') \n", + " .set('spark.executor.cores','{:d}'.format(cores_per_executor))\n", + " .set('spark.sql.adaptive.coalescePartitions.initialPartitionNum','{:d}'.format(nodes*executors_per_node*cores_per_executor*task_per_core))\n", + " .set('spark.sql.adaptive.enabled',True)\n", + " .set('spark.sql.adaptive.advisoryPartitionSizeInBytes', '256m')\n", + " .set('spark.executor.memory', '{:d}m'.format(int(math.floor(nodes*total_size/(nodes*executors_per_node)))-1024-int(math.floor(cache_size*1024/(nodes*executors_per_node)))))\n", + " .set('spark.task.cpus','1')\n", + " .set('spark.driver.memory','128g')\n", + " .set('spark.sql.inMemoryColumnarStorage.compressed','False')\n", + " .set('spark.sql.inMemoryColumnarStorage.batchSize','100000')\n", + "# .set('spark.memory.storageFraction','0.8')\n", + "# .set('spark.memory.fraction','0.7')\n", + " \n", + " .set('spark.sql.execution.arrow.fallback.enabled','True')\n", + " .set('spark.sql.execution.arrow.enabled','True')\n", + " .set('spark.sql.execution.arrow.maxRecordsPerBatch','100000')\n", + " .set(\"spark.sql.repl.eagerEval.enabled\", True)\n", + " \n", + " .set('spark.memory.offHeap.enabled','True')\n", + " .set('spark.memory.offHeap.size','{:d}m'.format(int(math.floor(cache_size*1024/(nodes*executors_per_node)))))\n", + " .set('spark.executor.memoryOverhead','{:d}m'.format(int(math.floor(cache_size*1024/(nodes*executors_per_node)))+3000))\n", + " .set('spark.sql.join.preferSortMergeJoin','False')\n", + "# .set('spark.executor.extraJavaOptions',\n", + "# '-XX:+UseG1GC -XX:+PrintFlagsFinal -XX:+PrintReferenceGC -verbose:gc -XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:InitiatingHeapOccupancyPercent=35')\n", + " .set('spark.executor.extraJavaOptions',\n", + " '-XX:+UseParallelGC -XX:+UseParallelOldGC -verbose:gc -XX:+PrintGCDetails -XX:+PrintGCTimeStamps')\n", + " .set('spark.dynamicAllocation.enabled', 'False')\n", + " .set('spark.shuffle.service.enabled', 'False')\n", + " .set('spark.dynamicAllocation.shuffleTracking.enabled', 'Falze')\n", + " .set(\"spark.sql.legacy.timeParserPolicy\",\"LEGACY\")\n", + "\n", + " \n", + "# .set(\"spark.sql.session.timeZone\", \"PST\")\n", + " .set(\"spark.sql.session.timeZone\", \"Etc/GMT-8\")\n", + "# .set(\"spark.driver.extraJavaOptions\",\"-Duser.timezone=UTC+8\")\n", + " \n", + " )\n", + "\n", + "sc = SparkContext(conf=conf,master='yarn')\n", + "sc.setLogLevel(\"ERROR\")\n", + "spark = SQLContext(sc)\n", + "time.sleep(10)\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%%html\n", + "" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Sparklog" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "scrolled": false + }, + "outputs": [], + "source": [ + "%run ./sparklog_monitor.ipynb" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import warnings\n", + "warnings.filterwarnings('ignore')" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "basedir=\"BASEDIR_TEMP\"\n", + "appid=\"APPID_TEMP\"\n", + "\n", + "last_basedir=\"LAST_BASEDIR\"\n", + "last_appid=\"LAST_APPID\"" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# native, partition table" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "app=Application_Run(appid,basedir=basedir)\n", + "appals=app.analysis['app']['als']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "appals.get_basic_state()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "appals.get_app_name()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#shuffle_df, dfx=appals.get_shuffle_stat()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# APP info" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "app.generate_trace_view(showemon=False,show_metric=emonmetric,disk_prefix='nvme%n1p1',nic_prefix=[\"'enp134s0f1'\",\"'eno1'\"])" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "appals=app.analysis['app']['als']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "appals.get_app_info(disk_prefix='nvme%n1p1',nic_prefix=[\"'enp134s0f1'\",\"'eno1'\"])" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "appals.show_critical_path_time_breakdown()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "hide_input": false, + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.6.8" + }, + "nbTranslate": { + "displayLangs": [ + "*" + ], + "hotkey": "alt-t", + "langInMainMenu": true, + "sourceLang": "en", + "targetLang": "fr", + "useGoogleTranslate": true + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": true, + "sideBar": false, + "skip_h1_title": false, + "title_cell": "Table of Contents", + "title_sidebar": "Contents", + "toc_cell": false, + "toc_position": { + "height": "197px", + "left": "2188px", + "top": "111px", + "width": "269px" + }, + "toc_section_display": true, + "toc_window_display": true + }, + "varInspector": { + "cols": { + "lenName": 16, + "lenType": 16, + "lenVar": 40 + }, + "kernels_config": { + "python": { + "delete_cmd_postfix": "", + "delete_cmd_prefix": "del ", + "library": "var_list.py", + "varRefreshCmd": "print(var_dic_list())" + }, + "r": { + "delete_cmd_postfix": ") ", + "delete_cmd_prefix": "rm(", + "library": "var_list.r", + "varRefreshCmd": "cat(var_dic_list()) " + } + }, + "types_to_exclude": [ + "module", + "function", + "builtin_function_or_method", + "instance", + "_Feature" + ], + "window_display": false + } + }, + "nbformat": 4, + "nbformat_minor": 2 +}