Files
kubeshark/performance_analysis/run_tapper_benchmark.sh
Nimrod Gilboa Markevich ab38f4c011 Add profiling tools (#1087)
* Add gin-contrib/pprof dependency

* Run pprof server on agent with --profiler flag

* Add --profiler flag to cli

* Fix error message

* Print cpu usage percentage

* measure cpu of current pid instead of globaly on the system

* Add scripts to plot performance

* Plot packetsCount in analysis

* Concat to DataFrame

* Plot in turbo colorscheme

* Make COLORMAP const

* Fix rss units

* Reduce code repetition by adding function for plotting

* Allow grouping based on filenames

* Temporary: Marked with comments where to disable code for experiments

* Add newline at end of file

* Add tap.cpuprofile flag. Change memprofile flag to tap.memprofile

* create tapper modes for debugging using env vars

* Fix rss plot units (MB instead of bytes)

* Remove comment

* Add info to plot script

* Remove tap.cpumemprofile. Rename tap.memprofile to memprofile

* Remove unused import

* Remove whitespaces

Co-authored-by: M. Mert Yıldıran <mehmet@up9.com>

* Remove whitespaces

Co-authored-by: M. Mert Yıldıran <mehmet@up9.com>

* Remove whitespaces

Co-authored-by: M. Mert Yıldıran <mehmet@up9.com>

* Remove whitespaces

Co-authored-by: M. Mert Yıldıran <mehmet@up9.com>

* Remove whitespaces

Co-authored-by: M. Mert Yıldıran <mehmet@up9.com>

* Remove whitespaces

Co-authored-by: M. Mert Yıldıran <mehmet@up9.com>

* Rename debug env vars

* Create package for debug env vars, read each env var once

* Run go mod tidy

* Increment MatchedPairs before emitting

* Only count cores once

* Count virtual and physical cores

* Add dbgctl replace in cli

* Fix lint: Check return values

* Add tap/dbgctl to test-lint make rule

* Replace tap/dbgctl in all modules

* #run_acceptance_tests

* Copy dbgctl module to docker image

* Debug/profile tapper benchmark (#1093)

* add mizu debug env to avoid all extensions

* add readme + run_tapper_benchmark.sh

* temporary change branch name

* fix readme

* fix MIZU_BENCHMARK_CLIENTS_COUNT env

* change tap target to tcp stream

* track live tcp streams

* pr fixes

* rename tapperPacketsCount to ignored_packets_count

* change mizu tapper to mizu debugg

Co-authored-by: David Levanon <dvdlevanon@gmail.com>
Co-authored-by: M. Mert Yıldıran <mehmet@up9.com>
2022-05-18 15:42:13 +03:00

93 lines
3.3 KiB
Bash
Executable File

#!/bin/bash
[ -z "$MIZU_HOME" ] && { echo "MIZU_HOME is missing"; exit 1; }
[ -z "$MIZU_BENCHMARK_OUTPUT_DIR" ] && export MIZU_BENCHMARK_OUTPUT_DIR="/tmp/mizu-benchmark-results-$(date +%d-%m-%H-%M)"
[ -z "$MIZU_BENCHMARK_CLIENT_PERIOD" ] && export MIZU_BENCHMARK_CLIENT_PERIOD="1m"
[ -z "$MIZU_BENCHMARK_URL" ] && export MIZU_BENCHMARK_URL="http://localhost:8081/data/b.1000.json"
[ -z "$MIZU_BENCHMARK_RUN_COUNT" ] && export MIZU_BENCHMARK_RUN_COUNT="3"
[ -z "$MIZU_BENCHMARK_QPS" ] && export MIZU_BENCHMARK_QPS="500"
[ -z "$MIZU_BENCHMARK_CLIENTS_COUNT" ] && export MIZU_BENCHMARK_CLIENTS_COUNT="5"
function log() {
local message=$@
printf "[%s] %s\n" "$(date "+%d-%m %H:%M:%S")" "$message"
}
function run_single_bench() {
local mode_num=$1
local mode_str=$2
log "Starting ${mode_num}_${mode_str} (runs: $MIZU_BENCHMARK_RUN_COUNT) (period: $MIZU_BENCHMARK_CLIENT_PERIOD)"
for ((i=0;i<"$MIZU_BENCHMARK_RUN_COUNT";i++)); do
log " $i: Running tapper"
rm -f tapper.log
nohup ./agent/build/mizuagent --tap --api-server-address ws://localhost:8899/wsTapper -i lo -stats 10 > tapper.log 2>&1 &
log " $i: Running client (hey)"
hey -z $MIZU_BENCHMARK_CLIENT_PERIOD -c $MIZU_BENCHMARK_CLIENTS_COUNT -q $MIZU_BENCHMARK_QPS $MIZU_BENCHMARK_URL > /dev/null || return 1
log " $i: Killing tapper"
kill -9 $(ps -ef | grep agent/build/mizuagent | grep tap | grep -v grep | awk '{ print $2 }') > /dev/null 2>&1
local output_file=$MIZU_BENCHMARK_OUTPUT_DIR/${mode_num}_${mode_str}_${i}.log
log " $i: Moving output to $output_file"
mv tapper.log $output_file || return 1
done
}
function generate_bench_graph() {
cd performance_analysis/ || return 1
source venv/bin/activate
python plot_from_tapper_logs.py $MIZU_BENCHMARK_OUTPUT_DIR/*.log || return 1
mv graph.png $MIZU_BENCHMARK_OUTPUT_DIR || return 1
}
mkdir -p $MIZU_BENCHMARK_OUTPUT_DIR
rm -f $MIZU_BENCHMARK_OUTPUT_DIR/*
log "Writing output to $MIZU_BENCHMARK_OUTPUT_DIR"
cd $MIZU_HOME || exit 1
export HOST_MODE=0
export MIZU_DEBUG_DISABLE_PCAP=false
export MIZU_DEBUG_DISABLE_TCP_REASSEMBLY=false
export MIZU_DEBUG_DISABLE_TCP_STREAM=false
export MIZU_DEBUG_DISABLE_NON_HTTP_EXTENSSION=false
export MIZU_DEBUG_DISABLE_DISSECTORS=false
export MIZU_DEBUG_DISABLE_EMITTING=false
export MIZU_DEBUG_DISABLE_SENDING=false
export MIZU_DEBUG_DISABLE_PCAP=true
run_single_bench "01" "no_pcap" || exit 1
export MIZU_DEBUG_DISABLE_PCAP=false
export MIZU_DEBUG_DISABLE_TCP_REASSEMBLY=true
run_single_bench "02" "no_assembler" || exit 1
export MIZU_DEBUG_DISABLE_TCP_REASSEMBLY=false
export MIZU_DEBUG_DISABLE_TCP_STREAM=true
run_single_bench "03" "no_tcp_stream" || exit 1
export MIZU_DEBUG_DISABLE_TCP_STREAM=false
export MIZU_DEBUG_DISABLE_NON_HTTP_EXTENSSION=true
run_single_bench "04" "only_http" || exit 1
export MIZU_DEBUG_DISABLE_NON_HTTP_EXTENSSION=false
export MIZU_DEBUG_DISABLE_DISSECTORS=true
run_single_bench "05" "no_dissectors" || exit 1
export MIZU_DEBUG_DISABLE_DISSECTORS=false
export MIZU_DEBUG_DISABLE_EMITTING=true
run_single_bench "06" "no_emit" || exit 1
export MIZU_DEBUG_DISABLE_EMITTING=false
export MIZU_DEBUG_DISABLE_SENDING=true
run_single_bench "07" "no_send" || exit 1
export MIZU_DEBUG_DISABLE_SENDING=false
run_single_bench "08" "normal" || exit 1
generate_bench_graph || exit 1
log "Output written to to $MIZU_BENCHMARK_OUTPUT_DIR"