Over a million developers have joined DZone.

Java Thread CPU Analysis on Windows

DZone 's Guide to

Java Thread CPU Analysis on Windows

Free Resource
This article will provide you with a tutorial on how you can quickly pinpoint the Java Thread contributors to a high CPU problem on the Windows OS. Windows, like other OS such as Linux, Solaris & AIX allow you to monitor the CPU utilization at the process level but also for individual Thread executing a task within a process.

For this tutorial, we created a simple Java program that will allow you to learn this technique in a step by step manner.

Troubleshooting tools

The following tools will be used below for this tutorial:

-        Windows Process Explorer (to pinpoint high CPU Thread contributors)
-        JVM Thread Dump (for Thread correlation and root cause analysis at code level)

High CPU simulator Java program

The simple program below is simply looping and creating new String objects. It will allow us to perform this CPU per Thread analysis. I recommend that you import it in an IDE of your choice e.g. Eclipse and run it from there. You should observe an increase of CPU on your Windows machine as soon as you execute it.
package org.ph.javaee.tool.cpu;

 * HighCPUSimulator
 * @author Pierre-Hugues Charbonneau
 * http://javaeesupportpatterns.blogspot.com
public class HighCPUSimulator {
       private final static int NB_ITERATIONS = 500000000;
       // ~1 KB data footprint
       private final static String DATA_PREFIX = "datadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadatadata";

        * @param args
       public static void main(String[] args) {
             System.out.println("HIGH CPU Simulator 1.0");
             System.out.println("Author: Pierre-Hugues Charbonneau");

             try {

                    for (int i = 0; i < NB_ITERATIONS; i++) {
                           // Perform some String manipulations to slowdown and expose looping process...
                           String data = DATA_PREFIX + i;                

             } catch (Throwable any) {
                    System.out.println("Unexpected Exception! " + any.getMessage()
                                        + " [" + any + "]");

             System.out.println("HighCPUSimulator done!");

Step #1 – Launch Process Explorer

The Process Explorer tool visually shows the CPU usage dynamically. It is good for live analysis. If you need historical data on CPU per Thread then you can also use Windows perfmon with % Processor Time & Thread Id data counters. You can download Process Explorer from the link below:

In our example, you can see that the Eclipse javaw.exe process is now using ~25% of total CPU utilization following the execution of our sample program.

Step #2 – Launch Process Explorer Threads view

The next step is to display the Threads view of the javaw.exe process. Simply right click on the javaw.exe process and select Properties. The Threads view will be opened as per below snapshot:

-        The first column is the Thread Id (decimal format)
-        The second column is the CPU utilization % used by each Thread
-        The third column is also another counter indicating if Thread is running on the CPU

In our example, we can see our primary culprit is Thread Id #5996 using ~ 25% of CPU.

Step #3 – Generate a JVM Thread Dump

At this point, Process Explorer will no longer be useful. The goal was to pinpoint one or multiple Java Threads consuming most of the Java process CPU utilization which is what we achieved. In order to go the next level in your analysis you will need to capture a JVM Thread Dump. This will allow you to correlate the Thread Id with the Thread Stack Trace so you can pinpoint that type of processing is consuming such high CPU.

JVM Thread Dump generation can be done in a few manners. If you are using JRockit VM you can simply use the jrcmd tool as per below example:

Once you have the Thread Dump data, simply search for the Thread Id and locate the Thread Stack Trace that you are interested in.

For our example, the Thread “Main Thread” which was fired from Eclipse got exposed as the primary culprit which is exactly what we wanted to demonstrate.
"Main Thread" id=1 idx=0x4 tid=5996 prio=5 alive, native_blocked
 at org/ph/javaee/tool/cpu/HighCPUSimulator.main
 at jrockit/vm/RNI.c2java(IIIII)V(Native Method)
 -- end of trace
Step #4 – Analyze the culprit Thread(s) Stack Trace and determine root cause

At this point you should have everything that you need to move forward with the root cause analysis. You will need to review each Thread Stack Trace and determine what type of problem you are dealing with. That final step is typically where you will spend most of your time and problem can be simple such as infinite looping or complex such as garbage collection related problems.

In our example, the Thread Dump did reveal the high CPU originates from our sample Java program around line 31. As expected, it did reveal the looping condition that we engineered on purpose for this tutorial.
for (int i = 0; i < NB_ITERATIONS; i++) {
                           // Perform some String manipulations to slowdown and expose looping process...
                           String data = DATA_PREFIX + i;                


I hope this tutorial has helped you understand how you can analyze and help pinpoint root cause of Java CPU problems on Windows OS. Please stay tuned for more updates, the next article will provide you with a Java CPU troubleshooting guide including how to tackle that last analysis step along with common problem patterns.



Published at DZone with permission of

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}