20 Linux System Monitoring Tools Every SysAdmin Should Know

Need to monitor Linux server performance? Try these built-in command and a few add-on tools. Most Linux distributions are equipped with tons of monitoring. These tools provide metrics which can be used to get information about system activities. You can use these tools to find the possible causes of a performance problem. The commands discussed below are some of the most basic commands when it comes to system analysis and debugging server issues such as:

  1. Finding out bottlenecks.
  2. Disk (storage) bottlenecks.
  3. CPU and memory bottlenecks.
  4. Network bottlenecks.

#1: top - Process Activity Command

The top program provides a dynamic real-time view of a running system i.e. actual process activity. By default, it displays the most CPU-intensive tasks running on the server and updates the list every five seconds.

Fig.01: Linux top command

Fig.01: Linux top command

Commonly Used Hot Keys

The top command provides several useful hot keys:

Hot Key Usage
t Displays summary information off and on.
m Displays memory information off and on.
A Sorts the display by top consumers of various system resources. Useful for quick identification of performance-hungry tasks on a system.
f Enters an interactive configuration screen for top. Helpful for setting up top for a specific task.
o Enables you to interactively select the ordering within top.
r Issues renice command.
k Issues kill command.
z Turn on or off color/mono

#2: vmstat - System Activity, Hardware and System Information

The command vmstat reports information about processes, memory, paging, block IO, traps, and cpu activity.

# vmstat 3

Sample Outputs:

procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------ 
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 0  0      0 2540988 522188 5130400    0    0     2    32    4    2  4  1 96  0  0
 1  0      0 2540988 522188 5130400    0    0     0   720 1199  665  1  0 99  0  0
 0  0      0 2540956 522188 5130400    0    0     0     0 1151 1569  4  1 95  0  0
 0  0      0 2540956 522188 5130500    0    0     0     6 1117  439  1  0 99  0  0
 0  0      0 2540940 522188 5130512    0    0     0   536 1189  932  1  0 98  0  0
 0  0      0 2538444 522188 5130588    0    0     0     0 1187 1417  4  1 96  0  0
 0  0      0 2490060 522188 5130640    0    0     0    18 1253 1123  5  1 94  0  0

Display Memory Utilization Slabinfo

# vmstat -m

Get Information About Active / Inactive Memory Pages

# vmstat -a

#3: w - Find Out Who Is Logged on And What They Are Doing

w command displays information about the users currently on the machine, and their processes.
# w username
# w Snehal
Sample Outputs:

  17:58:47 up 5 days, 20:28,  2 users,  load average: 0.36, 0.26, 0.24 
USER     TTY      FROM              LOGIN@   IDLE   JCPU   PCPU WHAT 
root     pts/0       14:55    5.00s  0.04s  0.02s vim /etc/resolv.conf
root     pts/1       17:43    0.00s  0.03s  0.00s w

#4: uptime - Tell How Long The System Has Been Running

The uptime command can be used to see how long the server has been running. The current time, how long the system has been running, how many users are currently logged on, and the system load averages for the past 1, 5, and 15 minutes.
# uptime


 18:02:41 up 41 days, 23:42,  1 user,  load average: 0.00, 0.00, 0.00

1 can be considered as optimal load value. The load can change from system to system. For a single CPU system 1 - 3 and SMP systems 6-10 load value might be acceptable.

#5: ps - Displays The Processes

ps command will report a snapshot of the current processes. To select all processes use the -A or -e option:
# ps -A

Sample Outputs:

  PID TTY          TIME CMD
    1 ?        00:00:02 init
    2 ?        00:00:02 migration/0
    3 ?        00:00:01 ksoftirqd/0
    4 ?        00:00:00 watchdog/0
    5 ?        00:00:00 migration/1
    6 ?        00:00:15 ksoftirqd/1
 4881 ?        00:53:28 java
 4885 tty1     00:00:00 mingetty
 4886 tty2     00:00:00 mingetty
 4887 tty3     00:00:00 mingetty
 4888 tty4     00:00:00 mingetty
 4891 tty5     00:00:00 mingetty
 4892 tty6     00:00:00 mingetty
 4893 ttyS1    00:00:00 agetty
12853 ?        00:00:00 cifsoplockd
12854 ?        00:00:00 cifsdnotifyd
14231 ?        00:10:34 lighttpd
14232 ?        00:00:00 php-cgi
54981 pts/0    00:00:00 vim
55465 ?        00:00:00 php-cgi
55546 ?        00:00:00 bind9-snmp-stat
55704 pts/1    00:00:00 ps

ps is just like top but provides more information.

Show Long Format Output

# ps -Al
To turn on extra full mode (it will show command line arguments passed to process):
# ps -AlF

To See Threads ( LWP and NLWP)

# ps -AlFH

To See Threads After Processes

# ps -AlLm

Print All Process On The Server

# ps ax
# ps axu

Print A Process Tree

# ps -ejH
# ps axjf
# pstree

Print Security Information

# ps -eo euser,ruser,suser,fuser,f,comm,label
# ps axZ
# ps -eM

See Every Process Running As User Vivek

# ps -U vivek -u vivek u

Set Output In a User-Defined Format

# ps -eo pid,tid,class,rtprio,ni,pri,psr,pcpu,stat,wchan:14,comm
# ps axo stat,euid,ruid,tty,tpgid,sess,pgrp,ppid,pid,pcpu,comm
# ps -eopid,tt,user,fname,tmout,f,wchan

Display Only The Process IDs of Lighttpd

# ps -C lighttpd -o pid=
# pgrep lighttpd

# pgrep -u vivek php-cgi

Display The Name of PID 55977

# ps -p 55977 -o comm=

Find Out The Top 10 Memory Consuming Process

# ps -auxf | sort -nr -k 4 | head -10

Find Out top 10 CPU Consuming Process

# ps -auxf | sort -nr -k 3 | head -10

#6: free - Memory Usage

The command free displays the total amount of free and used physical and swap memory in the system, as well as the buffers used by the kernel.
# free

Sample Output:

            total       used       free     shared    buffers     cached
Mem:      12302896    9739664    2563232          0     523124    5154740
-/+ buffers/cache:    4061800    8241096
Swap:      1052248          0    1052248

#7: iostat - Average CPU Load, Disk Activity

The command iostat report Central Processing Unit (CPU) statistics and input/output statistics for devices, partitions and network filesystems (NFS).
# iostat

Sample Outputs:

Linux 2.6.18-128.1.14.el5 (www03.nixcraft.in)  06/26/2009

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           3.50    0.09    0.51    0.03    0.00   95.86

Device:            tps   Blk_read/s   Blk_wrtn/s   Blk_read   Blk_wrtn
sda              22.04        31.88       512.03   16193351  260102868
sda1              0.00         0.00         0.00       2166        180
sda2             22.04        31.87       512.03   16189010  260102688
sda3              0.00         0.00         0.00       1615          0

#8: sar - Collect and Report System Activity

The sar command is used to collect, report, and save system activity information. To see network counter, enter:
# sar -n DEV | more

To display the network counters from the 24th:
# sar -n DEV -f /var/log/sa/sa24 | more

You can also display real time usage using sar:
# sar 4 5

Sample Outputs:

Linux 2.6.18-128.1.14.el5 (www03.nixcraft.in)   06/26/2009

06:45:12 PM       CPU     %user     %nice   %system   %iowait    %steal     %idle
06:45:16 PM       all      2.00      0.00      0.22      0.00      0.00     97.78
06:45:20 PM       all      2.07      0.00      0.38      0.03      0.00     97.52
06:45:24 PM       all      0.94      0.00      0.28      0.00      0.00     98.78
06:45:28 PM       all      1.56      0.00      0.22      0.00      0.00     98.22
06:45:32 PM       all      3.53      0.00      0.25      0.03      0.00     96.19
Average:          all      2.02      0.00      0.27      0.01      0.00     97.70

#9: mpstat - Multiprocessor Usage

The mpstat command displays activities for each available processor, processor 0 being the first one. mpstat -P ALL to display average CPU utilization per processor:
# mpstat -P ALL

Sample Output:

Linux 2.6.18-128.1.14.el5 (www03.nixcraft.in)   06/26/2009

06:48:11 PM  CPU   %user   %nice    %sys %iowait    %irq   %soft  %steal   %idle    intr/s
06:48:11 PM  all    3.50    0.09    0.34    0.03    0.01    0.17    0.00   95.86   1218.04
06:48:11 PM    0    3.44    0.08    0.31    0.02    0.00    0.12    0.00   96.04   1000.31
06:48:11 PM    1    3.10    0.08    0.32    0.09    0.02    0.11    0.00   96.28     34.93
06:48:11 PM    2    4.16    0.11    0.36    0.02    0.00    0.11    0.00   95.25      0.00
06:48:11 PM    3    3.77    0.11    0.38    0.03    0.01    0.24    0.00   95.46     44.80
06:48:11 PM    4    2.96    0.07    0.29    0.04    0.02    0.10    0.00   96.52     25.91
06:48:11 PM    5    3.26    0.08    0.28    0.03    0.01    0.10    0.00   96.23     14.98
06:48:11 PM    6    4.00    0.10    0.34    0.01    0.00    0.13    0.00   95.42      3.75
06:48:11 PM    7    3.30    0.11    0.39    0.03    0.01    0.46    0.00   95.69     76.89

#10: pmap - Process Memory Usage

The command pmap report memory map of a process. Use this command to find out causes of memory bottlenecks.
# pmap -d PID

To display process memory information for pid # 47394, enter:
# pmap -d 47394

Sample Outputs:

47394:   /usr/bin/php-cgi
Address           Kbytes Mode  Offset           Device    Mapping
0000000000400000    2584 r-x-- 0000000000000000 008:00002 php-cgi
0000000000886000     140 rw--- 0000000000286000 008:00002 php-cgi
00000000008a9000      52 rw--- 00000000008a9000 000:00000   [ anon ]
0000000000aa8000      76 rw--- 00000000002a8000 008:00002 php-cgi
000000000f678000    1980 rw--- 000000000f678000 000:00000   [ anon ]
000000314a600000     112 r-x-- 0000000000000000 008:00002 ld-2.5.so
000000314a81b000       4 r---- 000000000001b000 008:00002 ld-2.5.so
000000314a81c000       4 rw--- 000000000001c000 008:00002 ld-2.5.so
000000314aa00000    1328 r-x-- 0000000000000000 008:00002 libc-2.5.so
000000314ab4c000    2048 ----- 000000000014c000 008:00002 libc-2.5.so
00002af8d48fd000       4 rw--- 0000000000006000 008:00002 xsl.so
00002af8d490c000      40 r-x-- 0000000000000000 008:00002 libnss_files-2.5.so
00002af8d4916000    2044 ----- 000000000000a000 008:00002 libnss_files-2.5.so
00002af8d4b15000       4 r---- 0000000000009000 008:00002 libnss_files-2.5.so
00002af8d4b16000       4 rw--- 000000000000a000 008:00002 libnss_files-2.5.so
00002af8d4b17000  768000 rw-s- 0000000000000000 000:00009 zero (deleted)
00007fffc95fe000      84 rw--- 00007ffffffea000 000:00000   [ stack ]
ffffffffff600000    8192 ----- 0000000000000000 000:00000   [ anon ]
mapped: 933712K    writeable/private: 4304K    shared: 768000K

The last line is very important:

  • mapped: 933712K total amount of memory mapped to files
  • writeable/private: 4304K the amount of private address space
  • shared: 768000K the amount of address space this process is sharing with others

#11 and #12: netstat and ss - Network Statistics

The command netstat displays network connections, routing tables, interface statistics, masquerade connections, and multicast memberships. ss command is used to dump socket statistics. It allows showing information similar to netstat. 

#13: iptraf - Real-time Network Statistics

The iptraf command is interactive colorful IP LAN monitor. It is an ncurses-based IP LAN monitor that generates various network statistics including TCP info, UDP counts, ICMP and OSPF information, Ethernet load info, node stats, IP checksum errors, and others. It can provide the following info in easy to read format:

  • Network traffic statistics by TCP connection
  • IP traffic statistics by network interface
  • Network traffic statistics by protocol
  • Network traffic statistics by TCP/UDP port and by packet size
  • Network traffic statistics by Layer2 address

Fig.02: General interface statistics: IP traffic statistics by network interface

Fig.02: General interface statistics: IP traffic statistics by network interface 

Fig.03 Network traffic statistics by TCP connection

Fig.03 Network traffic statistics by TCP connection


#14: tcpdump - Detailed Network Traffic Analysis

The tcpdump is simple command that dump traffic on a network. However, you need good understanding of TCP/IP protocol to utilize this tool. For.e.g to display traffic info about DNS, enter:
# tcpdump -i eth1 'udp port 53'

To display all IPv4 HTTP packets to and from port 80, i.e. print only packets that contain data, not, for example, SYN and FIN packets and ACK-only packets, enter:
# tcpdump 'tcp port 80 and (((ip[2:2] - ((ip[0]&0xf)<<2)) - ((tcp[12]&0xf0)>>2)) != 0)'

To display all FTP session to, enter:
# tcpdump -i eth1 'dst and (port 21 or 20'

To display all HTTP session to
# tcpdump -ni eth0 'dst and tcp and port http'

Use wireshark to view detailed information about files, enter:
# tcpdump -n -i eth1 -s 0 -w output.txt src or dst port 80

#15: strace - System Calls

Trace system calls and signals. This is useful for debugging webserver and other server problems. See how to use to trace the process and see What it is doing.

#16: /Proc file system - Various Kernel Statistics

/proc file system provides detailed information about various hardware devices and other Linux kernel information. See Linux kernel /proc documentations for further details. Common /proc examples:
# cat /proc/cpuinfo
# cat /proc/meminfo
# cat /proc/zoneinfo
# cat /proc/mounts


#17: Nagios - Server And Network Monitoring

Nagios is a popular open source computer system and network monitoring application software. You can easily monitor all your hosts, network equipment and services. It can send alert when things go wrong and again when they get better. FAN is "Fully Automated Nagios". FAN goals are to provide a Nagios installation including most tools provided by the Nagios Community. FAN provides a CDRom image in the standard ISO format, making it easy to easilly install a Nagios server. Added to this, a wide bunch of tools are including to the distribution, in order to improve the user experience around Nagios.

#18: Cacti - Web-based Monitoring Tool

Cacti is a complete network graphing solution designed to harness the power of RRDTool's data storage and graphing functionality. Cacti provides a fast poller, advanced graph templating, multiple data acquisition methods, and user management features out of the box. All of this is wrapped in an intuitive, easy to use interface that makes sense for LAN-sized installations up to complex networks with hundreds of devices. It can provide data about network, CPU, memory, logged in users, Apache, DNS servers and much more. See how to install and configure Cacti network graphing tool under CentOS / RHEL.

#19: KDE System Guard - Real-time Systems Reporting and Graphing

KSysguard is a network enabled task and system monitor application for KDE desktop. This tool can be run over ssh session. It provides lots of features such as a client/server architecture that enables monitoring of local and remote hosts. The graphical front end uses so-called sensors to retrieve the information it displays. A sensor can return simple values or more complex information like tables. For each type of information, one or more displays are provided. Displays are organized in worksheets that can be saved and loaded independently from each other. So, KSysguard is not only a simple task manager but also a very powerful tool to control large server farms.

Fig.05 KDE System Guard

Fig.05 KDE System Guard {Image credit: Wikipedia}

See the KSysguard handbook for detailed usage.

#20: Gnome System Monitor - Real-time Systems Reporting and Graphing

The System Monitor application enables you to display basic system information and monitor system processes, usage of system resources, and file systems. You can also use System Monitor to modify the behavior of your system. Although not as powerful as the KDE System Guard, it provides the basic information which may be useful for new users:

  • Displays various basic information about the computer's hardware and software.
  • Linux Kernel version
  • GNOME version
  • Hardware
  • Installed memory
  • Processors and speeds
  • System Status
  • Currently available disk space
  • Processes
  • Memory and swap space
  • Network usage
  • File Systems
  • Lists all mounted filesystems along with basic information about each.

Fig.06 The Gnome System Monitor application

Fig.06 The Gnome System Monitor application

Bonus: Additional Tools

A few more tools:

  • nmap - scan your server for open ports.
  • lsof - list open files, network connections and much more.
  • ntop web based tool - ntop is the best tool to see network usage in a way similar to what top command does for processes i.e. it is network traffic monitoring software. You can see network status, protocol wise distribution of traffic for UDP, TCP, DNS, HTTP and other protocols.
  • Conky - Another good monitoring tool for the X Window System. It is highly configurable and is able to monitor many system variables including the status of the CPU, memory, swap space, disk storage, temperatures, processes, network interfaces, battery power, system messages, e-mail inboxes etc.
  • GKrellM - It can be used to monitor the status of CPUs, main memory, hard disks, network interfaces, local and remote mailboxes, and many other things.
  • vnstat - vnStat is a console-based network traffic monitor. It keeps a log of hourly, daily and monthly network traffic for the selected interface(s).
  • htop - htop is an enhanced version of top, the interactive process viewer, which can display the list of processes in a tree form.
  • mtr - mtr combines the functionality of the traceroute and ping programs in a single network diagnostic tool.

Did I miss something? Please add your favorite system motoring tool in the comments.

Java concurrency (multi-threading) - Tutorial

This article describes how to do concurrent programming with Java. It covers the concepts of parallel programming, immutability, threads, the executor framework (thread pools), futures, callables and the fork-join framework.

1. Concurrency

1.1. What is concurrency?

Concurrency is the ability to run several programs or several parts of a program in parallel. If a time consuming task can be performed asynchronously or in parallel, this improve the throughput and the interactivity of the program.
A modern computer has several CPU's or several cores within one CPU. The ability to leverage these multi-cores can be the key for a successful high-volume application.

1.2. Process vs. threads

process runs independently and isolated of other processes. It cannot directly access shared data in other processes. The resources of the process, e.g. memory and CPU time, are allocated to it via the operating system.
thread is a so called lightweight process. It has its own call stack, but can access shared data of other threads in the same process. Every thread has its own memory cache. If a thread reads shared data it stores this data in its own memory cache. A thread can re-read the shared data.
A Java application runs by default in one process. Within a Java application you work with several threads to achieve parallel processing or asynchronous behavior.

2. Improvements and issues with concurrency

Within a Java application you work with several threads to achieve parallel processing or asynchronous behavior.

2.1. Limits of concurrency gains

Concurrency promises to perform certain task faster as these tasks can be divided into subtasks and these subtasks can be executed in parallel. Of course the runtime is limited by parts of the task which can be performed in parallel.
The theoretical possible performance gain can be calculated by the following rule which is referred to as Amdahl's Law.
If F is the percentage of the program which can not run in parallel and N is the number of processes, then the maximum performance gain is 1/ (F+ ((1-F)/n)).

2.2. Concurrency issues

Threads have their own call stack, but can also access shared data. Therefore you have two basic problems, visibility and access problems.
A visibility problem occurs if thread A reads shared data which is later changed by thread B and thread A is unaware of this change.
An access problem can occur if several thread access and change the same shared data at the same time.
Visibility and access problem can lead to
  • Liveness failure: The program does not react anymore due to problems in the concurrent access of data, e.g. deadlocks.
  • Safety failure: The program creates incorrect data.

3. Concurrency in Java

3.1. Processes and Threads

A Java program runs in its own process and by default in one thread. Java supports threads as part of the Java language via the Thread code. The Java application can create new threads via this class.
Java 1.5 also provides improved support for concurrency with the in the java.util.concurrent package.

3.2. Locks and thread synchronization

Java provides locks to protect certain parts of the code to be executed by several threads at the same time. The simplest way of locking a certain method or Java class is to define the method or class with the synchronizedkeyword.
The synchronized keyword in Java ensures:
  • that only a single thread can execute a block of code at the same time
  • that each thread entering a synchronized block of code sees the effects of all previous modifications that were guarded by the same lock
Synchronization is necessary for mutually exclusive access to blocks of and for reliable communication between threads.
You can use the synchronized keyword for the definition of a method. This would ensure that only one thread can enter this method at the same time. Another threads which is calling this method would wait until the first threads leaves this method.
public synchronized void critial() {
  // some thread critical stuff
  // here
You can also use the synchronized keyword to protect blocks of code within a method. This block is guarded by a key, which can be either a string or an object. This key is called the lock. All code which is protected by the same lock can only be executed by one thread at the same time
For example the following datastructure will ensure that only one thread can access the inner block of the add() andnext() methods.
package de.vogella.pagerank.crawler;

import java.util.ArrayList;
import java.util.List;

/** * Data structure for a web crawler. Keeps track of the visited sites and keeps * a list of sites which needs still to be crawled. * * @author Lars Vogel * */
public class CrawledSites { private List crawledSites = new ArrayList(); private List linkedSites = new ArrayList(); public void add(String site) { synchronized (this) { if (!crawledSites.contains(site)) { linkedSites.add(site); } } }
/** * Get next site to crawl. Can return null (if nothing to crawl) */
public String next() { if (linkedSites.size() == 0) { return null; } synchronized (this) { // Need to check again if size has changed if (linkedSites.size() > 0) { String s = linkedSites.get(0); linkedSites.remove(0); crawledSites.add(s); return s; } return null; } } }

3.3. Volatile

If a variable is declared with the volatile keyword then it is guaranteed that any thread that reads the field will see the most recently written value. The volatile keyword will not perform any mutual exclusive lock on the variable.
As of Java 5 write access to a volatile variable will also update non-volatile variables which were modified by the same thread. This can also be used to update values within a reference variable, e.g. for a volatile variable person. In this case you must use a temporary variable person and use the setter to initialize the variable and then assign the temporary variable to the final variable. This will then make the address changes of this variable and the values visible to other threads.

4. The Java memory model

4.1. Overview

The Java memory model describes the communication between the memory of the threads and the main memory of the application.
It defines the rules how changes in the memory done by threads are propagated to other threads. The Java memory model also defines the situations in which a thread re-fresh its own memory from the main memory.
It also describes which operations are atomic and the ordering of the operations.

4.2. Atomic operation

An atomic operation is an operation which is performed as a single unit of work without the possibility of interference from other operations.
The Java language specification guarantees that reading or writing a variable is an atomic operation(unless the variable is of type long or double). Operations variables of type long or double are only atomic if they declared with the volatile keyword. .
Assume i is defined as int. The i++ (increment) operation it not an atomic operation in Java. This also applies for the other numeric types, e.g. long. etc).
The i++ operation first reads the value which is currently stored in i (atomic operations) and then it adds one to it (atomic operation). But between the read and the write the value of i might have changed.
Since Java 1.5 the java language provides atomic variables, e.g. AtomicInteger or AtomicLong which provide methods like getAndDecrement()getAndIncrement() and getAndSet() which are atomic.

4.3. Memory updates in synchronized code

The Java memory model guarantees that each thread entering a synchronized block of code sees the effects of all previous modifications that were guarded by the same lock.

5. Immutability and Defensive Copies

5.1. Immutability

The simplest way to avoid problems with concurrency is to share only immutable data between threads. Immutable data is data which cannot changed.
To make a class immutable make
  • all its fields final
  • the class declared as final
  • the this reference is not allowed to escape during construction
  • Any fields which refer to mutable data objects are
    • private
    • have no setter method
    • they are never directly returned of otherwise exposed to a caller
    • if they are changed internally in the class this change is not visible and has no effect outside of the class
An immutable class may have some mutable data which is uses to manages its state but from the outside this class nor any attribute of this class can get changed.
For all mutable fields, e.g. Arrays, that are passed from the outside to the class during the construction phase, the class needs to make a defensive-copy of the elements to make sure that no other object from the outside still can change the data

5.2. Defensive Copies

You must protected your classes from calling code. Assume that calling code will do its best to change your data in a way you didn't expect it. While this is especially true in case of immutable data it is also true for non-immutable data which you still not expect that this data is changed outside your class.
To protect your class against that you should copy data you receive and only return copies of data to calling code.
The following example creates a copy of a list (ArrayList) and returns only the copy of the list. This way the client of this class cannot remove elements from the list.
package de.vogella.performance.defensivecopy;

import java.util.ArrayList;
import java.util.Collections;
import java.util.List;

public class MyDataStructure {
  List list = new ArrayList();

  public void add(String s) {

/** * Makes a defensive copy of the List and return it * This way cannot modify the list itself * * @return List */
public List getList() { return Collections.unmodifiableList(list); } }

6. Threads in Java

The base means for concurrency are is the java.lang.Threads class. A Thread executes an object of typejava.lang.Runnable.
Runnable is an interface with defines the run() method. This method is called by the Thread object and contains the work which should be done. Therefore the "Runnable" is the task to perform. The Thread is the worker who is doing this task.
The following demonstrates a task (Runnable) which counts the sum of a given range of numbers. Create a new Java project called de.vogella.concurrency.threads for the example code of this section.
package de.vogella.concurrency.threads;

/** * MyRunnable will count the sum of the number from 1 to the parameter * countUntil and then write the result to the console. * * MyRunnable is the task which will be performed * * @author Lars Vogel * */
public class MyRunnable implements Runnable { private final long countUntil; MyRunnable(long countUntil) { this.countUntil = countUntil; } @Override public void run() { long sum = 0; for (long i = 1; i < countUntil; i++) { sum += i; } System.out.println(sum); } }
The following example demonstrate the usage of the Thread and the Runnable class.
package de.vogella.concurrency.threads;

import java.util.ArrayList;
import java.util.List;

public class Main {

  public static void main(String[] args) {
    // We will store the threads so that we can check if they are done
    List threads = new ArrayList();
    // We will create 500 threads
    for (int i = 0; i < 500; i++) {
      Runnable task = new MyRunnable(10000000L + i);
      Thread worker = new Thread(task);
      // We can set the name of the thread
      // Start the thread, never call method run() direct
      // Remember the thread for later usage
    int running = 0;
    do {
      running = 0;
      for (Thread thread : threads) {
        if (thread.isAlive()) {
      System.out.println("We have " + running + " running threads. ");
    } while (running > 0);

Using the Thread class directly has the following disadvantages.
  • Creating a new thread causes some performance overhead
  • Too many threads can lead to reduced performance, as the CPU needs to switch between these threads.
  • You cannot easily control the number of threads, therefore you may run into out of memory errors due to too many threads.
The java.util.concurrent package offers improved support for concurrency compared to the direct usage ofThreads. This package is described in the next section.

7. Threads pools with the Executor Framework


You find this examples in the source section in Java project calledde.vogella.concurrency.threadpools.
Thread pools manage a pool of worker threads. The thread pools contains a work queue which holds tasks waiting to get executed.
A thread pool can be described as a collection of Runnable objects (work queue) and a connections of running threads. These threads are constantly running and are checking the work query for new work. If there is new work to be done they execute this Runnable. The Thread class itself provides a method, e.g. execute(Runnable r) to add a new Runnable object to the work queue.
The Executor framework provides example implementation of the java.util.concurrent.Executor interface, e.g. Executors.newFixedThreadPool(int n) which will create n worker threads. The ExecutorService adds life cycle methods to the Executor, which allows to shutdown the Executor and to wait for termination.


If you want to use one thread pool with one thread which executes several runnables you can use the Executors.newSingleThreadExecutor() method.
Create again the Runnable.
package de.vogella.concurrency.threadpools;

/** * MyRunnable will count the sum of the number from 1 to the parameter * countUntil and then write the result to the console. * * MyRunnable is the task which will be performed * * @author Lars Vogel * */
public class MyRunnable implements Runnable { private final long countUntil; MyRunnable(long countUntil) { this.countUntil = countUntil; } @Override public void run() { long sum = 0; for (long i = 1; i < countUntil; i++) { sum += i; } System.out.println(sum); } }
Now you run your runnables with the executor framework.
package de.vogella.concurrency.threadpools;

import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;

public class Main {
  private static final int NTHREDS = 10;

  public static void main(String[] args) {
    ExecutorService executor = Executors.newFixedThreadPool(NTHREDS);
    for (int i = 0; i < 500; i++) {
      Runnable worker = new MyRunnable(10000000L + i);
    // This will make the executor accept no new threads
    // and finish all existing threads in the queue
    // Wait until all threads are finish
    System.out.println("Finished all threads");
In case the threads should return some value (result-bearing threads) then you can use thejava.util.concurrent.Callable class.

8. Futures and Callables

The code examples for this section are created in a Java project called de.vogella.concurrency.callables.
The executor framework presented in the last chapter works with Runnables. Runnable do not return result.
In case you expect your threads to return a computed result you can use java.util.concurrent.Callable. TheCallable object allows to return values after completion.
The Callable object uses generics to define the type of object which is returned.
If you submit a Callable object to an Executor the framework returns an object of typejava.util.concurrent.Future. This Future object can be used to check the status of a Callable and to retrieve the result from the Callable.
On the Executor you can use the method submit to submit a Callable and to get a future. To retrieve the result of the future use the get() method.
package de.vogella.concurrency.callables;

import java.util.concurrent.Callable;

public class MyCallable implements Callable {
  public Long call() throws Exception {
    long sum = 0;
    for (long i = 0; i <= 100; i++) {
      sum += i;
    return sum;

package de.vogella.concurrency.callables;

import java.util.ArrayList;
import java.util.List;
import java.util.concurrent.Callable;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;

public class CallableFutures {
  private static final int NTHREDS = 10;

  public static void main(String[] args) {

    ExecutorService executor = Executors.newFixedThreadPool(NTHREDS);
    List> list = new ArrayList>();
    for (int i = 0; i < 20000; i++) {
      Callable worker = new MyCallable();
      Future submit = executor.submit(worker);
    long sum = 0;
    // now retrieve the result
    for (Future future : list) {
      try {
        sum += future.get();
      } catch (InterruptedException e) {
      } catch (ExecutionException e) {

9. Nonblocking algorithms

Java 5.0 provides supports for additional atomic operations. This allows to develop algorithm which are non-blocking algorithm, e.g. which do not require synchronization, but are based on low-level atomic hardware primitives such as compare-and-swap (CAS). A compare-and-swap operation check if the variable has a certain value and if it has this value it will perform this operation.
Non-blocking algorithm are usually much faster then blocking algorithms as the synchronization of threads appears on a much finer level (hardware).
For example this created a non-blocking counter which always increases. This example is contained in the project called de.vogella.concurrency.nonblocking.counter.
package de.vogella.concurrency.nonblocking.counter;

import java.util.concurrent.atomic.AtomicInteger;

public class Counter {
  private AtomicInteger value = new AtomicInteger(); 
  public int getValue(){
    return value.get();
  public int increment(){
    return value.incrementAndGet();
  // Alternative implementation as increment but just make the 
  // implementation explicit
  public int incrementLongVersion(){
    int oldValue = value.get();
    while (!value.compareAndSet(oldValue, oldValue+1)){
       oldValue = value.get();
    return oldValue+1;
And a test.
package de.vogella.concurrency.nonblocking.counter;

import java.util.ArrayList;
import java.util.HashSet;
import java.util.List;
import java.util.Set;
import java.util.concurrent.Callable;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;

public class Test {
    private static final int NTHREDS = 10;

    public static void main(String[] args) {
      final Counter counter = new Counter();
      List> list = new ArrayList>();

      ExecutorService executor = Executors.newFixedThreadPool(NTHREDS);
      for (int i = 0; i < 500; i++) {
        Callable worker = new  Callable() {
          public Integer call() throws Exception {
            int number = counter.increment();
            return number ;
        Future submit= executor.submit(worker);

      // This will make the executor accept no new threads
      // and finish all existing threads in the queue
      // Wait until all threads are finish
      while (!executor.isTerminated()) {
      Set set = new HashSet();
      for (Future future : list) {
        try {
        } catch (InterruptedException e) {
        } catch (ExecutionException e) {
      if (list.size()!=set.size()){
        throw new RuntimeException("Double-entries!!!"); 


The interesting part is how the incrementAndGet() method is implemented. It uses a CAS operation.
public final int incrementAndGet() {
        for (;;) {
            int current = get();
            int next = current + 1;
            if (compareAndSet(current, next))
                return next;
The JDK itself makes more and more use of non-blocking algorithms to increase performance for every developer. Developing correct non-blocking algorithm is not a trivial task.
For more information on non-blocking algorithm, e.g. examples for a non-blocking Stack and non-block LinkedList, please see http://www.ibm.com/developerworks/java/library/j-jtp04186/index.html

10. Fork-Join in Java 7

Java 7 introduce a new parallel mechanism for compute intensive tasks, the fork-join framework. The fork-join framework allows you to distribute a certain task on several workers and then wait for the result.
E For Java 6.0 you can download the package (jsr166y) from Download site
For testing create the Java project "de.vogella.performance.forkjoin". If you are not using Java 7 you also need tojsr166y.jar to the classpath.
Create first a algorithm package and then the following class.
package algorithm;

import java.util.Random;

/** * * This class defines a long list of integers which defines the problem we will * later try to solve * */
public class Problem { private final int[] list = new int[2000000]; public Problem() { Random generator = new Random(19580427); for (int i = 0; i < list.length; i++) { list[i] = generator.nextInt(500000); } } public int[] getList() { return list; } }
Define now the Solver class as shown in the following example coding.


The API defines other top classes, e.g. RecursiveAction, AsyncAction. Check the Javadoc for details.
package algorithm;

import java.util.Arrays;

import jsr166y.forkjoin.RecursiveAction;

public class Solver extends RecursiveAction {
  private int[] list;
  public long result;

  public Solver(int[] array) {
    this.list = array;

  protected void compute() {
    if (list.length == 1) {
      result = list[0];
    } else {
      int midpoint = list.length / 2;
      int[] l1 = Arrays.copyOfRange(list, 0, midpoint);
      int[] l2 = Arrays.copyOfRange(list, midpoint, list.length);
      Solver s1 = new Solver(l1);
      Solver s2 = new Solver(l2);
      forkJoin(s1, s2);
      result = s1.result + s2.result;
Now define a small test class for testing it efficiency.
package testing;

import jsr166y.forkjoin.ForkJoinExecutor;
import jsr166y.forkjoin.ForkJoinPool;
import algorithm.Problem;
import algorithm.Solver;

public class Test {

  public static void main(String[] args) {
    Problem test = new Problem();
    // check the number of available processors
    int nThreads = Runtime.getRuntime().availableProcessors();
    Solver mfj = new Solver(test.getList());
    ForkJoinExecutor pool = new ForkJoinPool(nThreads);
    long result = mfj.getResult();
    System.out.println("Done. Result: " + result);
    long sum = 0;
    // check if the result was ok
    for (int i = 0; i < test.getList().length; i++) {
      sum += test.getList()[i];
    System.out.println("Done. Result: " + sum);

11. Deadlock

A concurrent application has the risk of a deadlock. A set of processes are deadlocked if all processes are waiting for an event which another process in the same set has to cause.
For example if thread A waits for a lock on object Z which thread B holds and thread B wait for a look on object Y which is hold be process A then these two processes are looked and cannot continue in their processing.