|Home||Switchboard||Unix Administration||Red Hat||TCP/IP Networks||Neoliberalism||Toxic Managers|
May the source be with you, but remember the KISS principle ;-)
Bigger doesn't imply better. Bigger often is a sign of obesity, of lost control, of overcomplexity, of cancerous cells
|Linux Performance Tuning||Recommended Links||Linux Kernel Tuning||Disk subsystem tuning||
Suse might have an edge over RHEL as it does not have SELinux (until version 11) but it does requre tuning to cut the bloat in order to demostrate better performance. In old version such resource hogs as Beagle should probably be de-installed for good. Cups and other rarely used or unused subsystems should not be installed at all. For example [UCI-Linux] Suse vs. Redhat - 2-CPU Recommendations states the following:
"I chose SuSE for the Opteron specifically because a brainy IBM solutions person told me that SuSE has been tuning for AMD for a few years (e.g., they wrote the GCC x86-64 back-end) and the threading performance on SuSE was much superior to RedHat so that's what he always recommended to customers. RedHat is apparently playing catch-up to SuSE, at least on x86-64, which may not affect you."
Note: Proposed changes need to be verified with Dell and Novell staff, tested and if they help implemented one at a time.
The value stored in /proc/sys/vm/dirty_background_ratio defines at what percentage of main memory the pdflush daemon should write data out to the disk.
If larger flushes are desired then increasing the default value of 10% to a larger value will cause less frequent flushes.
As in the example above the value can be changed to 25 as shown in
# sysctl -w vm.dirty_background_ratio=25
The default value 10 means that data will be written into system memory until the file system cache has a size of 10% of the server’s RAM.
The ratio at which dirty pages are written to disk can be altered as follows to a setting of 20% of the system memory
# sysctl -w vm.dirty_ratio=20
The best single advice that can be found on Net about Suse tuning is Patrick's Homepage/Getting SuSE Linux faster
Did you ever have the feeling, that every new SuSE Linux version runs more sluggish than the previous one? Well, you might be right about it. SuSE, like Microsoft tries to be a jack of all trades. That is, providing means for every demand, but in consequence optimizing for none. So what can be done to speed things up a bit?
First of all, let's define the scope of this howto (note: even if your system does not match these specs entirely, you might still be able to use one tip or the other):
Please note, that the following tips address experienced users, that are basicly familiar with the system and only need a hint of where to look.
- AMD 64 Bit CPU/512 MB RAM
- Desktopsystem (GNOME Desktop, but KDE installed as well)
- SuSE Linux 10.1
Get rid of unnecessary librariesNowadays, a lot of code is outsourced in shared libraries. But in order to use such code, the dynamic link loader is needed. This program requires the file /etc/ld.so.cache to find these libs (which is build up by /sbin/ldconfig). Naturally you want the cache file as small as possibly. Hence have a look at your system and deinstall every library you deem useless for you (easiest way: startup YaST softwaremanagement and search for the term "lib"). Users of 64 bit systems are advised to remove as many 32 bit libraries as possible (just search for "-32" in YaST). Note: it is not possible to remove them all, as certain "office" applications depend on them.
Shoot the beagleSuSE Linux 10.1 comes with a nifty app called "beagle", kind of a search engine, that tries to index your whole system (emails, IM, visited websites,... ). While in theory this sounds like a good idea, it'll be also a considerable performance hog. To get rid of it, simply remove the package "beagle". Note: You cannot (easily) deinstall the "libbeagle" shared libs.
Don't do it the ZEN wayAnother nifty idea, someone came up with was the ZEN installer (not to confuse with XEN virtualization). Also "just being able to install" packages (without fuss) sounds tempting, this means, that a demon is always running in the background, waiting to install/upgrade something. Having systems running, that do nothing 99% of the time is usually not desireable, so just stop the ZMD demon in YaST's runlevel editor. Note: This may mean, gui integrated "ZEN clients" start bugging you to reenable it. Deinstall the "zen-updater" package in this case as well.
The exorcistSuSE usually starts a lot more demons and other boottime programs, than you actually need or want. Some things to check:
Keep in mind: Even though the gains seem little, any program not running will save you tripple: A program not running does not take time to startup and it does not consume RAM, which can then be used for filecaching instead (hence speeding up the system).
- DHCP - It is convinient, but it also slows down booting till the server answers. If your network topology is simple and set, you just might want to switch to static IP address settings.
- Auditd - nice thing to have, if you are interested in watching what your programs do. If you don't, disable it.
- AppArmour - nice thing to have, if you don't trust yourself (or use untrusted third party software). Apparmor has it's own menuitem in YaST controlcenter, where it can be disabled.
- CUPS - if all your printjobs go to a remote printer, you do not need to run a local cups demon. In this case, you might also want to prevent loading of parallelport modules, which can be done by moving the according file in /etc/sysconfig/hardware away.
- mdsnd - Got a Mac? No? Ok!
- Once a day, CRON runs updatedb. One culd consider this as an earlier version of beagle and expendable as well. if you do not use locate(1), than you can probably also live without this service. Disable it in /etc/sysconfig/locate.
- YaST has been offering Software RAID as a partitioning option for some time now. Unless you explicitly choose to enable it, it is not used by default. However, upon bootup the according modules are loaded into the kernel. Easiest way to stop this, is to deinstall the "mdadm" package.
- By default, SuSE starts six getties (linux textconsole). As X is nowadays standard, you can probably live with less. Disable as many as you like in /etc/inittab (but keep at least one).
- Do you need gpg-agent and ssh-agent? They can both be disabled in /etc/X11/xdm/sys.xsession
- Do you need ipv6 networking? You can disable it by writing the following two lines to /etc/modprobe.conf.local:
alias net-pf-10 off
alias ipv6 off
Same can be done for other modules, that are loaded, but not needed.
Time does not matterA curious and often overlooked fact on Linux systems is, that filesystems not only keep track on when a file was modified/created but also when it was accessed last time. That is, simply reading a file (e.g. starting a program) will result in disc write operations. This may make some sense for pathes like /tmp (to find abbandonned tmpfiles) and /home ,but it is utterly useless for the static portions of the filesystem. To disable accesstime logging, the respective filesystem can be mounted with the "noatime" option in /etc/fstab. Note: /home and /tmp should live on partitions of their own (and have atime enabled).
Recompile the kernelOne should think, that this was an advise of the past, since stock kernels nowadays support anything and the kitchen sink. But basically, that's also the problem with them: They support way to much and are a compromise between desktop and server system.
For those who are not familiar with the build process (any more):
Things to watch out for:
- Set a "Local version" name under "General setup". Any string except "-default" will do (this prevents from clashing with the stock kernels).
- "Processor type" and features" is the main optimization goal. Specify your machiene exactly and set "Timer frequency" to "1000 Hz". Also set "Preemption Model" to "Voluntary Kernel Preemption" (not "Preemptible Kernel" - it will seriously stall the system upon disc IO).
- Throw out any device drivers, you do not have fitting hardware for.
ApplicationsApplication programmers tend to not pay attention to the resource hunger of their software nowadays. With some tweaks you can quite a boost in speed.
- Webbrowsers. Web browsers come with page caching facilities. While diskcache can considerably unburden the line, (big) memory cache does little good. It will only make the application use more RAM over time and therefore result in more pages getting swapped out (that is, a big memory cache will degrade into a diskcache eventually). While it is not feasible to disable memory cache completely, limiting it to a sane size of a few megabytes is certainly a good idea.
- No auditing please. If you feel unfompfortable with the X server being audited, you can disable it in /etc/opt/gnome/gdm/gdm.conf (no graphical means seem to be provided for this).
- Displaymanager. Prefer KDM in favour of GDM, as it has a smaller memory footprint (changeable via /etc/sysconfig/displaymanager).
- X itself. Do you really need 24 bit colordepth? Switching down to 16 bit is barely noticeable (unless you are into image manipulation) but will cut down X server memory consumption by 1/3.
Check your locale settingsSystemlanguage (glibc) is controlled through the $RC_LANG variable in /etc/sysconfig/language. For german users it seems to default to "de_DE.UTF-8", which is wrong, as there is no /usr/locale/de_DE.UTF8 directory. However, /usr/locale/de_DE.utf8 does exist (which somehow seems to be the fallback, too ). Trying to access the wrong place first results in several unsuccessfull open(3) systemcalls, whenever an application launches. German users are therefore recommended to set $RC_LANG to "de_DE.utf8" (you can check the result by using "strace" and looking for open(3) calls).
Tweaking gnomeGnome can be quite a resource hog and unfortunately it comes badly configured.
Things not to installThe following list contains packages, that should be checked, whether the according software is actually used. Merely having the package installed will result in software loaded and/or configs parsed upon login (obliviously loading unused programs/libs/configs is not so desireable).
Most of these packages install hooks into the bonobo-activation server and will therefore use resources even if not in active use.
Stop the resapplet from loadingThe resapplet takes care of switching screen resolution. This can be quite handy, if fullscreen aps (e.g. games) forget to switch back to the original resolution. However in most cases it's just another applet running and eating resources. The tricky part about disabling it is, that it is not in the users but the systems autostart. Hence the resapplet.desktop file in /opt/gnome/share/autostart has to be deleted.
Reduce themesSuSE's perception of icon themes seems to be: The more the merrier. Hence, they try to bind all installed themes together (even cross desktop, if the according KDE packages are installed). Selecting one icon theme may mean, that several others are loaded as well, since themes can depend on each other. To reduce icon themes, go to /opt/gnome/share/icons and edit the index.theme file of the theme of your choice. Unwanted themes can be thrown out by removing them from the "Inherits" line. Afterwards the theme cache must be rebuild using /sbin/conf.d/SuSEconfig.gtk2. Also try reducing fonts. Chances are, that way more fonts are installed than needed.
Cut down applnk usageKDE and GNOME use "*.desktop" files to describe programs. These files associate icons, localized names and other things with the binary. While this is good for useability, it also hit's performance hard, since on startup, GNOME tries to load every *.desktop it can find on the system (it will not only load but also watch them!) - This can mean hundreds of files. To lessen the load, either don't install desktop applications you do not need or limit the search pathes for those .desktop files in the environment variable $XDG_DATA_DIRS (it can be set in /etc/profiled.d/xdg*). Limiting the XDG datadirs to only /opt/gnome/share gains the biggest performance benefit, but will also strip down the desktop seriously (expect to loose program access through the panel menu).
Use gconf-editorA lot of things about gnome can be tweaked using gconf-editor. Though little can be done regarding performance here, it can greatly boost usability.
Create missing filesBy using strace you can find out, which syscalls a program invokes. Of particular interest ist the open(3) syscall, as it is very costly (doing disk io). Trying to open() a file, just to find out, it does not exist and then try to find it in several other locations will considerably slow down things. SuSE is missing some files and hence the desktop tries to guess where they could by (trying several places before giving up):
- ~/.Xdefaults (basicly a relict of older times. Not really needed any more today, but GNOME looks for it anyway. Simply create an empty file if none exists).
The obvious stuffThings worth remembering:
- Try to use "simple" wallpapers. Photorealistic, high resolution wallpapers will eat several MB of RAM. Especially of transparancy effects for the panel are activated.
- Try to keep the number of applets low. Throw out, whats not used.
- Disable animations (using gconf-editor) in the panel.
- Use lightweight themes. Small can also be beautiful.
- Disable preview in nautilus.
05 July 2007 | IBM
Chapter 1, Understanding the Linux operating system
Chapter 2. Monitoring and benchmark tools
Chapter 3. Analyzing performance bottlenecks
Chapter 4. Tuning the operating system
Download PDF (4.1 MB)
bonnie++ 1.03a. (Hard drive benchmark). netio 1.23. (Network benchmark). Nepim 0.27. (Network test). Netperf 2.4.2. (Network test). Iperf 2.0.2 ...
Benchmarks for Native IPsec in the 2.6 Kernel | Linux Journal
NETIO - Network Throughput Benchmark, Version 1.14. (c) 1997-2001 Kai Uwe Rommel.
Linux tip: Bash parameters and parameter expansions
Do you sometimes wonder how to use parameters with your scripts, and how to pass them to internal functions or other scripts? Do you need to do simple validity tests on parameters or options, or perform simple extraction and replacement operations on the parameter strings? This tip helps you with parameter use and the various parameter expansions available in the bash shell. Articles 16 May 2007
Writing endian-independent code in C
Architectures, processors, network stacks, and communication protocols all have to define endianness at some point. This article explains how endianness affects code, how to determine endianness at run time, and how to write code that can reverse byte order and free you from being bound to a certain endian. Articles 15 May 2007
Setting up a multicluster environment using General Parallel File System
Learn how to construct and deconstruct a simple multicluster of System x(TM)and System p(TM) computers using the General Parallel File System (GPFS). You can remotely add an existing GPFS cluster to another cluster. See how to mount a file system from the remote cluster using the GPFS secure communication protocol. Articles 11 May 2007
UNIX tips and tricks for a new user, Part 4: Some nifty shell tricks
When writing a shell program, you often come across some special situation that you'd like to handle automatically. This tutorial includes examples of such situations from small Bourne shell scripts. These situations include base conversion from one string to another (decimal to hex, hex to decimal, decimal to octal, and so on), reading the keyboard while in a piped loop, subshell execution, inline input, executing a command once for each file in a directory, and multiple ways to construct a continuous loop. Part 4 of this series wraps up with a collection of shell one-liners that perform useful functions. Tutorial 20 Feb 2007
Linux tip: Bash test and comparison functions
Are you confused by the plethora of testing and comparison options in the Bash shell? This tip helps you demystify the various types of file, arithmetic, and string tests so you will always know when to use test, [ ], [[ ]], (( )), or if-then-else constructs. Articles 20 Feb 2007
Linux tip: Printing DVI files with CUPS
Have you ever tried to print DVI or other files in Linux and gotten an "unsupported format" message? This tip shows you how to combine existing tools to make a CUPS print filter for printing DVI files. Articles 07 Feb 2007
Create uniform namespace using autofs with NFS Version 3 clients and servers
Do you have trouble accessing data exported from multiple file servers? If so, try using open source implementations of autofs and Lightweight Directory Access Protocol (LDAP), with Network File System (NFS) Version 3, to access data under the same global mount point. In this article, study and compare five different methods to create a uniform namespace using autofs. A handy table with a comparative evaluation is available to help you choose the best technique for your scenario. Articles 30 Jan 2007
Delve into UNIX process creation
Examine the life cycle of a process so that you can relate what you see happening on your system to what's going on within the kernel. System administrators must know how processes are created and destroyed within the UNIX(R) environment in order to understand how the system fits together and how to manage misbehaving processes. Similarly, developers must understand the UNIX processes model in order to write solid applications that run unattended and won't cause problems for system administrators. Articles 03 Jan 2007
IBM Cluster Systems Management: Installing the Webmin tool for Web access
Learn to install and use the Webmin UNIX(R) system administration tool so you can use a standard Web interface to remotely manage a clustered environment This article is an update to the 2005 developerWorks article about the IBM Cluster Systems Management tool. Articles 22 Dec 2006
System Administration Toolkit: Get the most out of zsh
Examine key parts of the Z shell (zsh) and how to use it's features to ease your UNIX(R) system administration tasks. zsh is a popular alternative to the original Bourne and Korn shells. It provides an impressive range of additional functionality, including improvements for completing different commands, files, and paths automatically, and for binding keys to functions and operations. Articles 19 Dec 2006
System Administration Toolkit: Get the most out of bash
Ease your system administration tasks by taking advantage of key parts of the Bourne-again shell (bash) and its features. Bash is a popular alternative to the original Bourne and Korn shells. It provides an impressive range of additional functionality that includes improvements to the scripting environment, extensive aliasing techniques, and improved methods for automatically completing different commands, files, and paths. Articles 12 Dec 2006
UNIX tools for exploring object files
The programs that run on a UNIX(R) system follow a careful design known as the object file format. Learn more about the object file format and the tools that you can use for exploring object files found on your system. Articles 21 Nov 2006
System Administration Toolkit: Swap space management and tricks
Configure your swap space (including adding space in an emergency) to get the most out of your system. In this article, you'll learn how to monitor your system to determine an effective swap space figure as well as examine methods for using swap space for more than just secondary random access memory (RAM). Articles 31 Oct 2006
Tunneling with SSH
Use OpenSource tools, such as Secure Shell (SSH), PuTTY, and Cygwin, to create secure connections to almost any resource you need to access. Current information on SSH tunneling and setup is fragmented and limited to specific applications, or it is written at a system administrator's level. With increasing security needs, the addition of boundary firewalls, and tightening of the number of allowed network ports, users need a method that is simple to configure, easy to operate and, above all, secure to accomplish day-to-day tasks and access the services that they have become accustomed to. This article describes the setup of a simple SSH client connecting to an AIX(R)- or Linux(R)-based SSH server that allows a typical, technically literate individual the ability to set up, configure, and operate a flexible means of tunneling data and services over the SSH service. Articles 17 Oct 2006
Run commands sequentially across a cluster from a UNIX server, Part 1: Secure Shell (SSH)
Configure Secure Shell (SSH) on IBM System p(TM) and System x(TM) computers so the UNIX(R) server can access a remote server without a password. Articles 21 Sep 2006
Get to know NetBSD
NetBSD runs on more hardware platforms than any other UNIX(R) derivative due to smart design decisions and a commitment to portable code. Whether you're porting an operating system to a proprietary embedded system or looking for stability and compatibility across hardware platforms in the lab, NetBSD and its open license is a compelling alternative to Linux(R) and the GNU Public License (GPL). Articles 29 Aug 2006
Port Fortran applications
Discover how to port Fortran-based High Performance Computing applications, such as computational fluid dynamic (CFD) modeling, weather modeling, and linear algebra packages, using IBM XL compilers and gnu compilers on large clusters. The solutions in this article are equally applicable in all types of porting work in Fortran or any UNIX(R)- or Linux-based systems. Though C/C++ and other programming languages are popular, with strong features and widespread use, Fortran is still the favorite for the scientific and engineering community because of its performance. Articles 29 Aug 2006
System Administration Toolkit: Standardizing your UNIX command-line tools
Examine methods for standardizing your interface to simplify movements between different UNIX(R) systems. If you manage multiple UNIX systems, particularly in a heterogeneous environment, then the hardest task can be switching between the different environments and performing the different tasks while having to consider all of the differences between the systems. This article does not cover specific differences, but you'll look at ways that can provide compatible layers, or wrappers, to support a consistent environment. Articles 22 Aug 2006
System Administration Toolkit: Backing up key information
Most UNIX(R) administrators have processes in place to back up the data and information on their UNIX machines, but what about the configuration files and other elements that provide the configuration data your machines need to operate? This article provides detailed information on techniques for achieving an effective and efficient backup system for these key files. Articles 15 Aug 2006
Take a closer look at OpenBSD
OpenBSD is quite possibly the most secure operating system on the planet. Every step of the development process focuses on building a secure, open, and free platform. UNIX(R) and Linux(R) administrators take note: Without realizing it, you probably use tools ported from OpenBSD every day. Maybe it's time to give the whole operating system a closer look. Articles 08 Aug 2006
System Administration Toolkit: Managing NIS services for authorizations
Examine how to set up, configure, and update a Network Information System (NIS) installation for sharing information, and learn how NIS can be merged with other solutions, such as files and Domain Name System (DNS), to provide subnet, network, and worldwide data sharing facilities. In a large UNIX(R) network, the ability to share information among the many systems helps to alleviate many problems, such as sharing permissions across different systems with Network File System (NFS), or simply providing a single login for the entire network. Articles 01 Aug 2006
System Administration Toolkit: Migrating and moving UNIX directory trees
Occasionally, you need to copy around an entire UNIX(R) directory tree, either between areas on the same system or between different systems. There are many different methods of achieving this, but not all preserve the right amount of information or are compatible across different systems. This article discusses the various options available for UNIX and how best to make them work. Articles 25 Jul 2006
System Administration Toolkit: Migrating and moving UNIX filesystems
Learn how to transfer an entire file system on a live system, including how to create, copy, and re-enable the new file system. If you have a UNIX(R) disk or system failure or simply fill up your file system, then you need to create a new partition and file system and copy over the contents. You might even need to mount the new partition in place to preserve the location of vital files and components. To add further complications, you need to do this on a live system, where you'd need to preserve file permissions, ownership, and possibly named pipes and other components. Effectively transferring these components and retaining all of this information is a vital part of the migration process. Articles 03 Jul 2006
System Administration Toolkit: Monitoring disk space and usage
Look at methods for determining disk usage across multiple UNIX(R) systems and how to create a simple warning system to alert you of potential problems. Keeping an eye on your file systems and ensuring they don't fill up is a trivial, but vital, process in the day-to-day management of your UNIX systems. In this article, you'll look at methods for keeping an eye on disk space, discovering which files, users, or applications are using up the most space, and how to make use of quotas and other solutions to find the information you need. Articles 13 Jun 2006
IBM DB2 Enterprise 9 performance with POWER5+ and AIX 5L multipage support
Learn how IBM DB2 9 automatically exploits the 64-kilobyte page support in AIX 5L to deliver high performance for database applications on this platform. This article talks about the DB2 9 for Linux, UNIX, and Windows exploitation of multiple page sizes, and shares some performance results as measured on IBM POWER5+ processor-based systems running IBM AIX 5L. Articles 08 Jun 2006
System Administration Toolkit: Monitoring a slow system
When your UNIX(R) system runs slow, it is vital that you discover what the problem is as quickly as possible so you can get your system back into the normal operating mode. There are many causes for a slow system, but actually identifying the problem can be exceedingly difficult. In this article, study examples of how to identify and diagnose the cause of your slow running UNIX system to get your machine running properly again. Articles 07 Jun 2006
Basic UNIX filesystem operations
Take advantage of the readdir() and stat() functions to run through the entries of a directory. Because of the plethora of files and directories on a UNIX(R) system, you're going to need to know how to process directory entries using the readdir() function and extract information about those entries using the stat() function. These basic file system operations can serve you well in your UNIX programming career, allowing you to easily discover and read files, directories, and symbolic links on your UNIX system. Articles 23 May 2006
Fun with strace and the GDB Debugger
Programming a UNIX system can be fun as well as educational. With the UNIX strace tool and GDB, the GNU Project Debugger, you can really dig deep into the functionality of your system and learn a lot about the various programs that comprise it. Using both tools in concert can be a rewarding experience as you look under the hood of your UNIX machine. Articles 11 May 2006
Performance tuning UNIX systems
Be nice to your computers and examine some general guidelines for tuning server performance. A computer is like an employee who does tasks for you -- it's a good idea to keep from overburdening them. One way to keep this from happening is to carefully tune the processes that run on it. This article provides some simple performance tuning steps using the UNIX nice commands. Articles 03 May 2006
System Administration Toolkit: Time and event management
Automating your administration tasks can save you a lot of time, and UNIX(R) incorporates systems to automate the process for you. In this article, you'll look at the best methods for using these systems, including how to monitor and track their execution and output and how to schedule complex events. Articles 02 May 2006
Make UNIX and Linux work together
Examine how to use the Network Information Service (NIS) to share core databases between Linux(R) and UNIX(R), and how to use the Network File System (NFS) to share file systems, both with direct links and through the automounter. Although UNIX and Linux are similar, there are some differences between the two that can complicate the process of integrating the two systems. Both, for example, share the same authentication system, but most systems are also standalone. Sharing this authentication information enables you to provide a single sign-on (SSO) functionality to any of the servers in your network. Articles 18 Apr 2006
Using ReiserFS with Linux
Take a look at the ext2 (second extended file system), ext3 (third extended file system), and Reiser4 file systems and discover how to create your own Reiser4 file system. The most commonly used file system, ext2, is a traditional UNIX(R)-style file system that doesn't mix well with modern hard drive sizes. The ext3 file system adds journalling, but not much else. If you want something really advanced, you might want to check out the current Reiser4 file system. Articles 04 Apr 2006
Retool your Linux skills for commercial UNIX
Examine how to best migrate your Linux(R) skills to take advantage of AIX(R) and Solaris. Linux is all the rage, but what if you have experience in Linux and need to apply it to a commercial UNIX(R) environment? UNIX and Linux are similar, and many of the same principles exist; there is a shell, root is still all powerful, and many of the tools and applications are the same. But how do you cope with understanding the nuances and differences? Tutorial 30 Mar 2006
Advanced techniques for using the UNIX find command
Explore the vast terrain of the UNIX(R) file system with the find command. One of the most powerful and useful commands in the UNIX programmer's repertoire is find. All flavors of UNIX have file systems that can contain thousands of files of many different types. With so many choices, locating a specific file, or set of files, can be difficult. The find command makes this task easier in many ways. Articles 28 Mar 2006
Solve application problems with tracing
Peer into the behavior of an application with truss. When an application doesn't work as expected, you typically look at application and system logs as a first recourse. But when logs don't help, UNIX(R) provides a powerful set of tools that you can use to trace the application while it runs. Armed with these traces and a bit of knowledge about UNIX, you can easily solve your application problems. Articles 21 Mar 2006
Build UNIX software with Eclipse
Become more productive with your own code and others by utilizing Eclipse's syntax highlighting, code completion, and other amenities. Eclipse is an excellent open source IDE and has many helpful features. It runs on any UNIX(R) platform with a Java(TM) Runtime Environment (JRE) (Version 1.4 or newer) and an SWT port, such as Linux(R), Solaris, AIX(R), and HP-UX. It's easy enough to start a new project using Eclipse or to import an existing Eclipse project, but how do you bring existing code into the IDE? And what if you need to get an existing project compiling right away without modifying its existing makefile or configure script -- the most common ways of building UNIX software? Read along for the answer to all of these questions. Articles 14 Mar 2006
Differentiating UNIX and Linux
Investigate the areas where UNIX(R) and Linux(R) converge in terms of functionality, environment, usability, and also those areas where UNIX and Linux differ. Many refer to Linux as a UNIX-like operating system. It is an open source operating system that has many of the same principles and ideals as UNIX, but it is not a true UNIX operating system like Solaris, AIX(R), HP-UX, and others. This article covers a range of aspects, from the core technical elements, such as kernel and filesystem support, to application tools, availability, and the differences in how to administer them. Articles 14 Mar 2006
Effective management of system logs
Provide solid information resources to decision makers. Discover a simple, but useful, application of the combined processing capabilities of awk and XML that you can use to present UNIX(R) system data in a reader-friendly form suitable for posting to the company intranet or Internet. UNIX generates useful system performance, usage, cost, and related data that management and other interested stakeholders can use. Articles 07 Mar 2006
nmon performance: A free tool to analyze AIX and Linux performance
This free tool gives you a huge amount of information all on one screen. Even though IBM doesn't officially support the tool and you must use it at your own risk, you can get a wealth of performance statistics. Why use five or six tools when one free tool can give you everything you need? Articles 27 Feb 2006
IBM NFS/DFS Authentication Gateway
Take advantage of the new features Network File System Version 4 (NFS Version 4) now has to offer. With the ever-growing storage needs in large enterprises and NFS implementations offering more and more features, it makes business sense for enterprises to migrate to NFS Version 4. In this article, we discuss the need and various strategies for migrating from the IBM Distributed Computing Environment (DCE)/Distributed File System(TM) (DFS(TM)) infrastructure to NFS Version 4 on AIX(R) and Linux(R). Articles 26 Jan 2006
A comparison of security subsystems on AIX, Linux, and Solaris
Learn how to apply a strategy for implementing a single identification and authentication (I and A) framework across a heterogeneous, multi-platform environment. An I and A system provides a layer of abstraction between a user application and the implementation of any authentication or identification functions it needs to perform. Articles 13 Oct 2005
How to install AIX 5L
Get step-by-step instructions for a number of generic AIX(R) system setup procedures. These instructions should work for any of the AIX(TM) 5L releases. Variations, if any, should be minor. Articles 31 Aug 2005
Split mirror using suspended I/O in DB2 Universal Database
Looking for a high availability solution for DB2 UDB? This article provides a visual explanation of split mirror using suspended I/O for a DB2 UDB database and explains the details of various implementation scenarios. Learn how suspended I/O works and how it can be used to implement reliable high availability database solutions. Articles 25 Aug 2005
POWER5 virtualization: How to set up the IBM Virtual I/O Server
Get more information on the virtualization capabilities of IBM(R) POWER5(TM) servers. Follow along as Nigel Griffiths illustrates how to set up and use the IBM Virtual I/O Server (VIO Server). In his previous article, "POWER5 Virtualization: How to set up the SUSE Linux Virtual I/O Server", he described the benefits of the IBM POWER5 servers and provided examples on how to set up the environment for pSeries(R) p5 and eServer(TM) OpenPower(TM) systems. Articles 29 Jun 2005
Guide to porting from Solaris to Linux on x86
Solaris is considered one of the closest flavors of UNIX to Linux, but for migration purposes, there can be differences between the two in the areas of memory mapping, threading, or natural language support (to name just a few). This porting guide gives you advice on planning for the port to Linux/x86, and helps you understand the differences in the development environment and architecture. Articles 29 Apr 2005
Dual boot Linux and AIX
There may be times when you find it necessary to develop in both the Linux and AIX operating environments. This article describes dual booting Linux and AIX on the same IBM eServer pSeries (including eServer p5), eServer i5, or eServer OpenPower server. Articles 25 Apr 2005
Serving X from a Windows laptop
This article is an update to a previous article about working on UNIX(R) through your laptop. The author describes how to run the same environment from your laptop as you do when directly connected to a UNIX server's console terminal. The article discusses how to use X clients, installing uwin, and running X clients through a VPN. Articles 27 May 2004
Using Samba as a primary domain controller
Open source Samba turns a UNIX(R) or Linux(R) system into a file and print server for Microsoft(R) Windows(R) network clients. Tom Syroid dishes up a juicy tutorial that shows you how to configure Samba as the primary domain controller on an xSeries(R) server.
Maximizing the Performance of SUSE Linux Enterprise Real Time for Financial ... in Oracle9i Database Performance Tuning Guide and Reference Release 2 (9.2). ...
Over the past few years, Linux has made its way into the data centers of many corporations all over the globe. The Linux operating system has become accepted by both the scientific and enterprise user population. Today, Linux is by far the most versatile operating system. You can find Linux on embedded devices such as firewalls and cell phones and mainframes. Naturally, performance of the Linux operating system has become a hot topic for both scientific and enterprise users. However, calculating a global weather forecast and hosting a database impose different requirements on the operating system. Linux has to accommodate all possible usage scenarios with the most optimal performance. The consequence of this challenge is that most Linux distributions contain general tuning parameters to accommodate all users.
IBM® has embraced Linux, and it is recognized as an operating system suitable for enterprise-level applications running on IBM systems. Most enterprise applications are now available on Linux, including file and print servers, database servers, Web servers, and collaboration and mail servers.
With use of Linux in an enterprise-class server comes the need to monitor performance and, when necessary, tune the server to remove bottlenecks that affect users. This IBM Redpaper describes the methods you can use to tune Linux, tools that you can use to monitor and analyze server performance, and key tuning parameters for specific server applications. The purpose of this redpaper is to understand, analyze, and tune the Linux operating system to yield superior performance for any type of application you plan to run on these systems.
The tuning parameters, benchmark results, and monitoring tools used in our test environment were executed on Red Hat and Novell SUSE Linux kernel 2.6 systems running on IBM System x servers and IBM System z servers. However, the information in this redpaper should be helpful for all Linux hardware platforms.
Performance Tuning for Linux Servers provides the basic knowledge and skills needed to understand the performance of Linux servers, and contains examples based on popular enterprise Linux distributions (SUSE Linux Enterprise Server, and Red Hat Enterprise Linux) useful to Linux developers, administrators and architects of all experience levels. Read on to see if it this book might benefit you.
Kirk Coombs provides a great review of this book.
So, you have created that brand new killer Linux application for SUSE Linux and you're running it on the latest hardware. While testing, you hypothesize that your recently purchased server is capable of giving you a little more out of your application. How do you go about figuring out where you can gain more performance? Check out OProfile.
Check out this nice array of questions and answers about the SUSE kernel.
For Suse 11 Simply type "yum install iotop". Iotop is licensed under the terms of the GNU GPL. The latest version is Iotop 0.3.2 (NEWS), available here : iotop-0.3.2.tar.bz2 or iotop-0.3.2.tar.gz.
Freshmeat project page to stay informed: http://freshmeat.net/projects/iotop.
QUOTE(Static @ Apr 6 2007, 11:52 AM)I uninstalled Zenworks, but I'm still getting this massive load on my cpu whenever I log in. I don't like it and I want this parse-metdata process gone.
Also remove packages beagle and kerry. They may cause substantial CPU load after booting.
QUOTE(D0M1N8R @ Apr 9 2007, 01:25 PM)Try zypper (opensuseupdater). It is a selection in YaST, software management, filter on patterns. At the same time you install 'OpenSuSE Software Management', deinstall the 'Enterprise Software Management'. Reboot. If you do not see a blue icon in the taskbar, start it with 'opensuseupdater' , no quotes.
Experiencing same thing..
I setup this box probably about 2 months ago
just decided to toss my games on it
Installed nVidia driver
startup glxgears and get 15500+ FPS for a few seconds and then drops way down to 500 FPS. sometimes down to 10FPS
95% parse-metadata thread..
then 95% load with update-status
back to 15500 FPS
so How do I remove meta all together? the search feature is not worth this.
Google matched content
Chapter 1. Tuning the operating system
Chapter 2. Tuning tools
Chapter 3. Analyzing performance bottlenecks
Chapter 4. Tuning Apache
Chapter 5. Tuning database servers
Chapter 6. Tuning Samba for file and print
Chapter 7. Tuning LDAP
Chapter 8. Tuning Lotus Domino
Server Oriented System Tuning Info
Configuring SUSE Linux on POWER5 to maximize performance
Tuning Red Hat for maximum performance
Tom Syroid, Author and Linux administrator
21 Aug 2002
This tutorial details the ins and outs of transforming a stock, "out of the box" Red Hat installation into a finely tuned, stable system customized to individual needs and tastes. The material presented here is based on Red Hat 7.3, although many of the techniques and procedures discussed are equally applicable to other mainstream Linux distributions. And while the title uses the phrase "performance tuning," you'll soon discover that performance and security often go hand in hand.
Slashdot Performance Tuning for Linux Servers
Groupthink : Two Party System as Polyarchy : Corruption of Regulators : Bureaucracies : Understanding Micromanagers and Control Freaks : Toxic Managers : Harvard Mafia : Diplomatic Communication : Surviving a Bad Performance Review : Insufficient Retirement Funds as Immanent Problem of Neoliberal Regime : PseudoScience : Who Rules America : Neoliberalism : The Iron Law of Oligarchy : Libertarian Philosophy
War and Peace : Skeptical Finance : John Kenneth Galbraith :Talleyrand : Oscar Wilde : Otto Von Bismarck : Keynes : George Carlin : Skeptics : Propaganda : SE quotes : Language Design and Programming Quotes : Random IT-related quotes : Somerset Maugham : Marcus Aurelius : Kurt Vonnegut : Eric Hoffer : Winston Churchill : Napoleon Bonaparte : Ambrose Bierce : Bernard Shaw : Mark Twain Quotes
Vol 25, No.12 (December, 2013) Rational Fools vs. Efficient Crooks The efficient markets hypothesis : Political Skeptic Bulletin, 2013 : Unemployment Bulletin, 2010 : Vol 23, No.10 (October, 2011) An observation about corporate security departments : Slightly Skeptical Euromaydan Chronicles, June 2014 : Greenspan legacy bulletin, 2008 : Vol 25, No.10 (October, 2013) Cryptolocker Trojan (Win32/Crilock.A) : Vol 25, No.08 (August, 2013) Cloud providers as intelligence collection hubs : Financial Humor Bulletin, 2010 : Inequality Bulletin, 2009 : Financial Humor Bulletin, 2008 : Copyleft Problems Bulletin, 2004 : Financial Humor Bulletin, 2011 : Energy Bulletin, 2010 : Malware Protection Bulletin, 2010 : Vol 26, No.1 (January, 2013) Object-Oriented Cult : Political Skeptic Bulletin, 2011 : Vol 23, No.11 (November, 2011) Softpanorama classification of sysadmin horror stories : Vol 25, No.05 (May, 2013) Corporate bullshit as a communication method : Vol 25, No.06 (June, 2013) A Note on the Relationship of Brooks Law and Conway Law
Fifty glorious years (1950-2000): the triumph of the US computer engineering : Donald Knuth : TAoCP and its Influence of Computer Science : Richard Stallman : Linus Torvalds : Larry Wall : John K. Ousterhout : CTSS : Multix OS Unix History : Unix shell history : VI editor : History of pipes concept : Solaris : MS DOS : Programming Languages History : PL/1 : Simula 67 : C : History of GCC development : Scripting Languages : Perl history : OS History : Mail : DNS : SSH : CPU Instruction Sets : SPARC systems 1987-2006 : Norton Commander : Norton Utilities : Norton Ghost : Frontpage history : Malware Defense History : GNU Screen : OSS early history
The Peter Principle : Parkinson Law : 1984 : The Mythical Man-Month : How to Solve It by George Polya : The Art of Computer Programming : The Elements of Programming Style : The Unix Hater’s Handbook : The Jargon file : The True Believer : Programming Pearls : The Good Soldier Svejk : The Power Elite
Most popular humor pages:
Manifest of the Softpanorama IT Slacker Society : Ten Commandments of the IT Slackers Society : Computer Humor Collection : BSD Logo Story : The Cuckoo's Egg : IT Slang : C++ Humor : ARE YOU A BBS ADDICT? : The Perl Purity Test : Object oriented programmers of all nations : Financial Humor : Financial Humor Bulletin, 2008 : Financial Humor Bulletin, 2010 : The Most Comprehensive Collection of Editor-related Humor : Programming Language Humor : Goldman Sachs related humor : Greenspan humor : C Humor : Scripting Humor : Real Programmers Humor : Web Humor : GPL-related Humor : OFM Humor : Politically Incorrect Humor : IDS Humor : "Linux Sucks" Humor : Russian Musical Humor : Best Russian Programmer Humor : Microsoft plans to buy Catholic Church : Richard Stallman Related Humor : Admin Humor : Perl-related Humor : Linus Torvalds Related humor : PseudoScience Related Humor : Networking Humor : Shell Humor : Financial Humor Bulletin, 2011 : Financial Humor Bulletin, 2012 : Financial Humor Bulletin, 2013 : Java Humor : Software Engineering Humor : Sun Solaris Related Humor : Education Humor : IBM Humor : Assembler-related Humor : VIM Humor : Computer Viruses Humor : Bright tomorrow is rescheduled to a day after tomorrow : Classic Computer Humor
The Last but not Least Technology is dominated by two types of people: those who understand what they do not manage and those who manage what they do not understand ~Archibald Putt. Ph.D
Copyright © 1996-2018 by Dr. Nikolai Bezroukov. www.softpanorama.org was initially created as a service to the (now defunct) UN Sustainable Development Networking Programme (SDNP) in the author free time and without any remuneration. This document is an industrial compilation designed and created exclusively for educational use and is distributed under the Softpanorama Content License. Original materials copyright belong to respective owners. Quotes are made for educational purposes only in compliance with the fair use doctrine.
FAIR USE NOTICE This site contains copyrighted material the use of which has not always been specifically authorized by the copyright owner. We are making such material available to advance understanding of computer science, IT technology, economic, scientific, and social issues. We believe this constitutes a 'fair use' of any such copyrighted material as provided by section 107 of the US Copyright Law according to which such material can be distributed without profit exclusively for research and educational purposes.
This is a Spartan WHYFF (We Help You For Free) site written by people for whom English is not a native language. Grammar and spelling errors should be expected. The site contain some broken links as it develops like a living tree...
|You can use PayPal to make a contribution, supporting development of this site and speed up access. In case softpanorama.org is down you can use the at softpanorama.info|
The statements, views and opinions presented on this web page are those of the author (or referenced source) and are not endorsed by, nor do they necessarily reflect, the opinions of the author present and former employers, SDNP or any other organization the author may be associated with. We do not warrant the correctness of the information provided or its fitness for any purpose.
Last modified: March 12, 2019