DZone
Thanks for visiting DZone today,
Edit Profile
  • Manage Email Subscriptions
  • How to Post to DZone
  • Article Submission Guidelines
Sign Out View Profile
  • Post an Article
  • Manage My Drafts
Over 2 million developers have joined DZone.
Log In / Join
Refcards Trend Reports
Events Video Library
Refcards
Trend Reports

Events

View Events Video Library

Related

  • Quality Assurance in AI-Driven Business Evolution
  • The New Testing Pattern: Standardizing Regression for Cloud Migrations
  • From Test Automation to Autonomous Quality: Designing AI Agents for Data Validation at Scale
  • When Memory Overflows: Too Many ApplicationContexts in Spring Integration Tests

Trending

  • Bringing Intelligence Closer to the Source: Why Real-Time Processing is the Heart of Edge AI
  • AI Agents in Java: Architecting Intelligent Health Data Systems
  • Ujorm3: A New Lightweight ORM for JavaBeans and Records
  • When Perfect Data Breaks: The Journey from Data Quality to Data Observability
  1. DZone
  2. Software Design and Architecture
  3. Performance
  4. Monitoring and Plotting Process Memory/CPU Usage With top and gnuplot

Monitoring and Plotting Process Memory/CPU Usage With top and gnuplot

Learn how to use top and gnuplot to monitor and plot the memory and CPU usage of a Linux process.

By 
Jakub Holý user avatar
Jakub Holý
·
Updated Oct. 23, 18 · Code Snippet
Likes (5)
Comment
Save
Tweet
Share
25.8K Views

Join the DZone community and get the full member experience.

Join For Free

If you want to monitor the memory and CPU usage of a particular Linux process for a few minutes, perhaps during a performance test, you can capture the data with top and plot them with gnuplot. Keep reading to learn how.

Run this script (perhaps via nohup) to capture the data:

#!/bin/sh
# Usage: ./monitor-usage.sh <PID of the process>
# Output: top.dat with lines such as `1539689171 305m 2.0`, i.e. unix time - memory with m/g suffix - CPU load in %
# To plot the output, see https://gist.github.com/jakubholynet/931a3441982c833f5f8fcdcf54d05c91
export PID=$1
rm top.dat
while true; do top -p $PID -bMn 1 | egrep '^[0-9]+' | awk -v now=$(date +%s) '{print now,$6,$9}' >> top.dat; done

Then plot the data via ./usage-plot.gp top.dat top.png:

#!/usr/bin/env gnuplot --persist -c
# Plot memory and CPU usage over time. Usage:
#  usage-plot.gp <input file> [<output .png file>]
# where the input file has the columns `<unix time> <memory, with m/g suffix> <% cpu>`
# To create the input file, see https://gist.github.com/jakubholynet/931a3441982c833f5f8fcdcf54d05c91

# Arguments:
infile=ARG1
outfile=ARG2
set term x11
set title 'Memory, CPU usage from' . infile
set xdata time
set timefmt "%s"
set xlabel "Time [[hh:]mm:ss]"
set ylabel "Memory usage"
set format y '%.1s%cB'

set y2label 'CPU usage'
set format y2 '%.0s%%'
set y2tics nomirror
set tics out
set autoscale y
set autoscale y2

resolveUnit(u,mul,x)=(pos=strstrt(x,u), pos > 0 ? sprintf("%f",real(substr(x,1,pos - 1))*mul) : x)
resolveUnits(x)=(resolveUnit("g",1024*1024*1024,resolveUnit("m",1024*1024,x)))
check(x)=(real(resolveUnits(x)))
if (exists("outfile") && strlen(outfile) > 0) {
    print "Outputting to the file ", outfile
    set term png # 640,480
    set output outfile
}

# Styling
set style line 1 linewidth 2 linecolor 'blue'
set style line 2 linecolor 'light-green'
#set xtics font ", 10"
set tics font ", 10"
set xtics rotate 60 # put label every 60s, make vertical so they don't clash in .png if too many

plot infile u 1:3 with lp axes x1y2 title "cpu" linestyle 2, \
    infile using 1:(check(stringcolumn(2))) with linespoints title "memory" linestyle 1


Gnuplot Data (computing) Memory (storage engine) Testing Linux (operating system) Monitor (synchronization)

Published at DZone with permission of Jakub Holý. See the original article here.

Opinions expressed by DZone contributors are their own.

Related

  • Quality Assurance in AI-Driven Business Evolution
  • The New Testing Pattern: Standardizing Regression for Cloud Migrations
  • From Test Automation to Autonomous Quality: Designing AI Agents for Data Validation at Scale
  • When Memory Overflows: Too Many ApplicationContexts in Spring Integration Tests

Partner Resources

×

Comments

The likes didn't load as expected. Please refresh the page and try again.

  • RSS
  • X
  • Facebook

ABOUT US

  • About DZone
  • Support and feedback
  • Community research

ADVERTISE

  • Advertise with DZone

CONTRIBUTE ON DZONE

  • Article Submission Guidelines
  • Become a Contributor
  • Core Program
  • Visit the Writers' Zone

LEGAL

  • Terms of Service
  • Privacy Policy

CONTACT US

  • 3343 Perimeter Hill Drive
  • Suite 215
  • Nashville, TN 37211
  • [email protected]

Let's be friends:

  • RSS
  • X
  • Facebook