DOCSLIB.ORG

Beyond Init: Systemd Linux.Conf.Au 2011

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Beyond Init: Systemd Linux.Conf.Au 2011 Beyond Init: systemd linux.conf.au 2011 Lennart Poettering January 2011 Lennart Poettering Beyond Init: systemd compatible with SysV and LSB init scripts. systemd provides aggressive parallelization capabilities, uses socket and D-Bus activation for starting services, offers on-demand starting of daemons, keeps track of processes using Linux cgroups, supports snapshotting and restoring of the system state, maintains mount and automount points and implements an elaborate transactional dependency-based service control logic. It can work as a drop-in replacement for sysvinit." \systemd is a system and session manager for Linux, Lennart Poettering Beyond Init: systemd systemd provides aggressive parallelization capabilities, uses socket and D-Bus activation for starting services, offers on-demand starting of daemons, keeps track of processes using Linux cgroups, supports snapshotting and restoring of the system state, maintains mount and automount points and implements an elaborate transactional dependency-based service control logic. It can work as a drop-in replacement for sysvinit." \systemd is a system and session manager for Linux, compatible with SysV and LSB init scripts. Lennart Poettering Beyond Init: systemd uses socket and D-Bus activation for starting services, offers on-demand starting of daemons, keeps track of processes using Linux cgroups, supports snapshotting and restoring of the system state, maintains mount and automount points and implements an elaborate transactional dependency-based service control logic. It can work as a drop-in replacement for sysvinit." \systemd is a system and session manager for Linux, compatible with SysV and LSB init scripts. systemd provides aggressive parallelization capabilities, Lennart Poettering Beyond Init: systemd offers on-demand starting of daemons, keeps track of processes using Linux cgroups, supports snapshotting and restoring of the system state, maintains mount and automount points and implements an elaborate transactional dependency-based service control logic. It can work as a drop-in replacement for sysvinit." \systemd is a system and session manager for Linux, compatible with SysV and LSB init scripts. systemd provides aggressive parallelization capabilities, uses socket and D-Bus activation for starting services, Lennart Poettering Beyond Init: systemd keeps track of processes using Linux cgroups, supports snapshotting and restoring of the system state, maintains mount and automount points and implements an elaborate transactional dependency-based service control logic. It can work as a drop-in replacement for sysvinit." \systemd is a system and session manager for Linux, compatible with SysV and LSB init scripts. systemd provides aggressive parallelization capabilities, uses socket and D-Bus activation for starting services, offers on-demand starting of daemons, Lennart Poettering Beyond Init: systemd supports snapshotting and restoring of the system state, maintains mount and automount points and implements an elaborate transactional dependency-based service control logic. It can work as a drop-in replacement for sysvinit." \systemd is a system and session manager for Linux, compatible with SysV and LSB init scripts. systemd provides aggressive parallelization capabilities, uses socket and D-Bus activation for starting services, offers on-demand starting of daemons, keeps track of processes using Linux cgroups, Lennart Poettering Beyond Init: systemd maintains mount and automount points and implements an elaborate transactional dependency-based service control logic. It can work as a drop-in replacement for sysvinit." \systemd is a system and session manager for Linux, compatible with SysV and LSB init scripts. systemd provides aggressive parallelization capabilities, uses socket and D-Bus activation for starting services, offers on-demand starting of daemons, keeps track of processes using Linux cgroups, supports snapshotting and restoring of the system state, Lennart Poettering Beyond Init: systemd and implements an elaborate transactional dependency-based service control logic. It can work as a drop-in replacement for sysvinit." \systemd is a system and session manager for Linux, compatible with SysV and LSB init scripts. systemd provides aggressive parallelization capabilities, uses socket and D-Bus activation for starting services, offers on-demand starting of daemons, keeps track of processes using Linux cgroups, supports snapshotting and restoring of the system state, maintains mount and automount points Lennart Poettering Beyond Init: systemd It can work as a drop-in replacement for sysvinit." \systemd is a system and session manager for Linux, compatible with SysV and LSB init scripts. systemd provides aggressive parallelization capabilities, uses socket and D-Bus activation for starting services, offers on-demand starting of daemons, keeps track of processes using Linux cgroups, supports snapshotting and restoring of the system state, maintains mount and automount points and implements an elaborate transactional dependency-based service control logic. Lennart Poettering Beyond Init: systemd \systemd is a system and session manager for Linux, compatible with SysV and LSB init scripts. systemd provides aggressive parallelization capabilities, uses socket and D-Bus activation for starting services, offers on-demand starting of daemons, keeps track of processes using Linux cgroups, supports snapshotting and restoring of the system state, maintains mount and automount points and implements an elaborate transactional dependency-based service control logic. It can work as a drop-in replacement for sysvinit." Lennart Poettering Beyond Init: systemd init(8) Lennart Poettering Beyond Init: systemd Parallelization Lennart Poettering Beyond Init: systemd Lennart Poettering Beyond Init: systemd The kernel orders and buffers requests for us! Implicit dependencies! Patching daemons Socket-Based Activation Lennart Poettering Beyond Init: systemd Implicit dependencies! Patching daemons Socket-Based Activation The kernel orders and buffers requests for us! Lennart Poettering Beyond Init: systemd Patching daemons Socket-Based Activation The kernel orders and buffers requests for us! Implicit dependencies! Lennart Poettering Beyond Init: systemd Socket-Based Activation The kernel orders and buffers requests for us! Implicit dependencies! Patching daemons Lennart Poettering Beyond Init: systemd Bus-Based Activation Lennart Poettering Beyond Init: systemd Starting Less: On-Demand Loading Lennart Poettering Beyond Init: systemd autofs! Parallelizing File System Jobs Lennart Poettering Beyond Init: systemd Parallelizing File System Jobs autofs! Lennart Poettering Beyond Init: systemd Move to systemd, daemons, kernel, udev, . Provide proper debugging facilities Shell is evil Lennart Poettering Beyond Init: systemd Provide proper debugging facilities Shell is evil Move to systemd, daemons, kernel, udev, . Lennart Poettering Beyond Init: systemd Shell is evil Move to systemd, daemons, kernel, udev, . Provide proper debugging facilities Lennart Poettering Beyond Init: systemd Control Groups! The best babysitter. Lennart Poettering Beyond Init: systemd The best babysitter. Control Groups! Lennart Poettering Beyond Init: systemd Environment, resource limits, working directory, chroot(), umask, OOM adjustment, nice level, IO priority and class, CPU scheduler priority and policy/reset-on-fork, CPU affinity, timer slack, stdio to syslog/tty/null/kmsg, uid, gid, supplementary groups, file system namespacing (r/o file systems, inaccessible systems, mount propagation, private /tmp), capabilities (inherited set, bounding set, secure bits), . The best babysitter II Lennart Poettering Beyond Init: systemd The best babysitter II Environment, resource limits, working directory, chroot(), umask, OOM adjustment, nice level, IO priority and class, CPU scheduler priority and policy/reset-on-fork, CPU affinity, timer slack, stdio to syslog/tty/null/kmsg, uid, gid, supplementary groups, file system namespacing (r/o file systems, inaccessible systems, mount propagation, private /tmp), capabilities (inherited set, bounding set, secure bits), . Lennart Poettering Beyond Init: systemd Unit types: service, socket, device, mount, automount, target, snapshot, timer, swap, path Lennart Poettering Beyond Init: systemd Read SysV/LSB init script headers, read /etc/fstab, support traditional inetd modes, support /dev/initctl, utmp, wtmp, support double-fork()ing daemons. .desktop files. Don't reinvent the wheel: Lennart Poettering Beyond Init: systemd .desktop files. Don't reinvent the wheel: Read SysV/LSB init script headers, read /etc/fstab, support traditional inetd modes, support /dev/initctl, utmp, wtmp, support double-fork()ing daemons. Lennart Poettering Beyond Init: systemd Don't reinvent the wheel: Read SysV/LSB init script headers, read /etc/fstab, support traditional inetd modes, support /dev/initctl, utmp, wtmp, support double-fork()ing daemons. .desktop files. Lennart Poettering Beyond Init: systemd Snapshots Lennart Poettering Beyond Init: systemd Transaction System Lennart Poettering Beyond Init: systemd D-Bus! Lennart Poettering Beyond Init: systemd systemadm Lennart Poettering Beyond Init: systemd systemd for cross-distribution standardization systemd as basic OS building block Lennart Poettering Beyond Init: systemd systemd as basic OS building block systemd for cross-distribution standardization Lennart Poettering Beyond Init: systemd Substantial coverage of basic OS boot-up tasks, including fsck, mount, quota,
Load more

Recommended publications
  • Oracle® Linux Administrator's Solutions Guide for Release 6
    Oracle® Linux Administrator's Solutions Guide for Release 6 E37355-64 August 2017 Oracle Legal Notices Copyright © 2012, 2017, Oracle and/or its affiliates. All rights reserved. This software and related documentation are provided under a license agreement containing restrictions on use and disclosure and are protected by intellectual property laws. Except as expressly permitted in your license agreement or allowed by law, you may not use, copy, reproduce, translate, broadcast, modify, license, transmit, distribute, exhibit, perform, publish, or display any part, in any form, or by any means. Reverse engineering, disassembly, or decompilation of this software, unless required by law for interoperability, is prohibited. The information contained herein is subject to change without notice and is not warranted to be error-free. If you find any errors, please report them to us in writing. If this is software or related documentation that is delivered to the U.S. Government or anyone licensing it on behalf of the U.S. Government, then the following notice is applicable: U.S. GOVERNMENT END USERS: Oracle programs, including any operating system, integrated software, any programs installed on the hardware, and/or documentation, delivered to U.S. Government end users are "commercial computer software" pursuant to the applicable Federal Acquisition Regulation and agency-specific supplemental regulations. As such, use, duplication, disclosure, modification, and adaptation of the programs, including any operating system, integrated software, any programs installed on the hardware, and/or documentation, shall be subject to license terms and license restrictions applicable to the programs. No other rights are granted to the U.S.
    [Show full text]
  • The GNOME Census: Who Writes GNOME?
    The GNOME Census: Who writes GNOME? Dave Neary & Vanessa David, Neary Consulting © Neary Consulting 2010: Some rights reserved Table of Contents Introduction.........................................................................................3 What is GNOME?.............................................................................3 Project governance...........................................................................3 Why survey GNOME?.......................................................................4 Scope and methodology...................................................................5 Tools and Observations on Data Quality..........................................7 Results and analysis...........................................................................10 GNOME Project size.......................................................................10 The Long Tail..................................................................................11 Effects of commercialisation..........................................................14 Who does the work?.......................................................................15 Who maintains GNOME?................................................................17 Conclusions........................................................................................22 References.........................................................................................24 Appendix 1: Modules included in survey...........................................25 2 Introduction What
    [Show full text]
  • State of Linux Audio in 2009 Linux Plumbers Conference 2009
    State of Linux Audio in 2009 Linux Plumbers Conference 2009 Lennart Poettering [email protected] September 2009 Lennart Poettering State of Linux Audio in 2009 Who Am I? Software Engineer at Red Hat, Inc. Developer of PulseAudio, Avahi and a few other Free Software projects http://0pointer.de/lennart/ [email protected] IRC: mezcalero Lennart Poettering State of Linux Audio in 2009 Perspective Lennart Poettering State of Linux Audio in 2009 So, what happened since last LPC? Lennart Poettering State of Linux Audio in 2009 RIP: EsounD is officially gone. Lennart Poettering State of Linux Audio in 2009 (at least on Fedora) RIP: OSS is officially gone. Lennart Poettering State of Linux Audio in 2009 RIP: OSS is officially gone. (at least on Fedora) Lennart Poettering State of Linux Audio in 2009 Audio API Guide http://0pointer.de/blog/projects/guide-to-sound-apis Lennart Poettering State of Linux Audio in 2009 We also make use of high-resolution timers on the desktop by default. We now use realtime scheduling on the desktop by default. Lennart Poettering State of Linux Audio in 2009 We now use realtime scheduling on the desktop by default. We also make use of high-resolution timers on the desktop by default. Lennart Poettering State of Linux Audio in 2009 2s Buffers Lennart Poettering State of Linux Audio in 2009 Mixer abstraction? Due to user-friendliness, i18n, meta data (icons, ...) We moved a couple of things into the audio server: Timer-based audio scheduling; mixing; flat volume/volume range and granularity extension; integration of volume sliders; mixer abstraction; monitoring Lennart Poettering State of Linux Audio in 2009 We moved a couple of things into the audio server: Timer-based audio scheduling; mixing; flat volume/volume range and granularity extension; integration of volume sliders; mixer abstraction; monitoring Mixer abstraction? Due to user-friendliness, i18n, meta data (icons, ...) Lennart Poettering State of Linux Audio in 2009 udev integration: meta data, by-path/by-id/..
    [Show full text]
  • Developer Conference 2012 Schedule
    Developer Conference 2012 Schedule Friday D1 D2 D3 Vlastimil Eliáš Jakub Hrozek, Jan Zeleny František Řezníček Centralized Identity Management 9:00 - 9:45 FreeIPA/SSSD + Intro to MIT Kerberos [Security] Towards Unified Messaging [CoreOS] And SSO for jboss.org Community Systems [JBoss] Alexander Bokovoy, Andreas Schneider Kamil Dudka, Ondřej Vašík Christian Sadilek 9:50 - 10:35 Cross-realm Trusts in FreeIPA v3.0 [Security] Common C/C++ Error Patterns & Static Analysis [CoreOS] JBoss Errai Framework [JBoss] Eduard Beneš, Miroslav Grepl Adam Tkáč Martin Kouba 10:40 - 11:25 SELinux News in Fedora 16 [Security] DNSSEC in Fedora 17 [CoreOS] CDI for Seam2 [JBoss] Dmitri Pal Marcela Maslanova, Jindrich Novy Peter Skopek 11:30 - 12:15 Identity Management Roadmap + MIT and Kerberos [Security] Dynamic Software Collections [CoreOS] PicketBox in AS7 [JBoss] Quick break Peter Vrabec Bryn Reeves Karel Piwko 12:30 - 13:15 Security Compliance Check in Fedora [Security] Supporting the Open Source Enterprise [Misc] Arquillian Drone Helping Ike Get Rid of the Bugs [JBoss] Jan Hutar Miloslav Trmac Software Robot Competition Around the World and Our Way Pavol Pitoňák 13:20 - 14:05 Concise Overview of Security [Security] [Misc, CZ] RichFaces: Testing on Mobile Devices [JBoss] Steve Grubb Stanislav Kozina Geoffrey De Smet 14:10 - 14:55 Government Security [Security] What Can Linux Learn from the Others [Misc] What Are Drools, Guvnor and Planner? [JBoss] Lennart Poettering, Kay Sievers Phil Knirsch Pavel Tisnovsky 15:00 - 15:45 Do's and Don'ts when Writing System
    [Show full text]
  • Oracle® Linux 7 Working with LXC
    Oracle® Linux 7 Working With LXC F32445-02 October 2020 Oracle Legal Notices Copyright © 2020, Oracle and/or its affiliates. This software and related documentation are provided under a license agreement containing restrictions on use and disclosure and are protected by intellectual property laws. Except as expressly permitted in your license agreement or allowed by law, you may not use, copy, reproduce, translate, broadcast, modify, license, transmit, distribute, exhibit, perform, publish, or display any part, in any form, or by any means. Reverse engineering, disassembly, or decompilation of this software, unless required by law for interoperability, is prohibited. The information contained herein is subject to change without notice and is not warranted to be error-free. If you find any errors, please report them to us in writing. If this is software or related documentation that is delivered to the U.S. Government or anyone licensing it on behalf of the U.S. Government, then the following notice is applicable: U.S. GOVERNMENT END USERS: Oracle programs (including any operating system, integrated software, any programs embedded, installed or activated on delivered hardware, and modifications of such programs) and Oracle computer documentation or other Oracle data delivered to or accessed by U.S. Government end users are "commercial computer software" or "commercial computer software documentation" pursuant to the applicable Federal Acquisition Regulation and agency-specific supplemental regulations. As such, the use, reproduction, duplication, release, display, disclosure, modification, preparation of derivative works, and/or adaptation of i) Oracle programs (including any operating system, integrated software, any programs embedded, installed or activated on delivered hardware, and modifications of such programs), ii) Oracle computer documentation and/or iii) other Oracle data, is subject to the rights and limitations specified in the license contained in the applicable contract.
    [Show full text]
  • Readahead: Time-Travel Techniques for Desktop and Embedded Systems
    Readahead: time-travel techniques for desktop and embedded systems Michael Opdenacker Free Electrons [email protected] Abstract 1.2 Reading ahead Readahead techniques have successfully been used to The idea of reading ahead is to speed up the access to reduce boot time in recent GNU/Linux distributions like a file by preloading at least parts of its contents in page Fedora Core or Ubuntu. However, in embedded sys- cache ahead of time. This can be done when spare I/O tems with scarce RAM, starting a parallel thread read- resources are available, typically when tasks keep the ing ahead all the files used in system startup is no longer processor busy. Of course, this requires the ability to appropriate. The cached pages could be reclaimed even predict the future! before accessing the corresponding files. Fortunately, the systems we are dealing with are pre- This paper will first guide you through the heuristics dictable or even totally predictable in some situations! implemented in kernelspace, as well as through the userspace interface for preloading files or just announc- ing file access patterns. Desktop implementations will • Predictions by watching file read patterns. If pages be explained and benchmarked. We will then detail Free are read from a file in a sequential manner, it makes Electrons’ attempts to implement an easy to integrate sense to go on reading the next blocks in the file, helper program reading ahead files at the most appropri- even before these blocks are actually requested. ate time in the execution flow. • System startup. The system init sequence doesn’t This paper and the corresponding presentation target change.
    [Show full text]
  • Beyond Init: Systemd Linux Plumbers Conference 2010
    Beyond Init: systemd Linux Plumbers Conference 2010 Kay Sievers Lennart Poettering November 2010 Kay Sievers, Lennart Poettering Beyond Init: systemd Triggers: Boot, Socket, Bus, Device, Path, Timers, More Kay Sievers, Lennart Poettering Beyond Init: systemd Kay Sievers, Lennart Poettering Beyond Init: systemd Substantial coverage of basic OS boot-up tasks, including fsck, mount, quota, hwclock, readahead, tmpfiles, random-seed, console, static module loading, early syslog, plymouth, shutdown, kexec, SELinux, initrd+initrd-less boots. Status: almost made Fedora 14. Kay Sievers, Lennart Poettering Beyond Init: systemd including fsck, mount, quota, hwclock, readahead, tmpfiles, random-seed, console, static module loading, early syslog, plymouth, shutdown, kexec, SELinux, initrd+initrd-less boots. Status: almost made Fedora 14. Substantial coverage of basic OS boot-up tasks, Kay Sievers, Lennart Poettering Beyond Init: systemd mount, quota, hwclock, readahead, tmpfiles, random-seed, console, static module loading, early syslog, plymouth, shutdown, kexec, SELinux, initrd+initrd-less boots. Status: almost made Fedora 14. Substantial coverage of basic OS boot-up tasks, including fsck, Kay Sievers, Lennart Poettering Beyond Init: systemd quota, hwclock, readahead, tmpfiles, random-seed, console, static module loading, early syslog, plymouth, shutdown, kexec, SELinux, initrd+initrd-less boots. Status: almost made Fedora 14. Substantial coverage of basic OS boot-up tasks, including fsck, mount, Kay Sievers, Lennart Poettering Beyond Init: systemd hwclock, readahead, tmpfiles, random-seed, console, static module loading, early syslog, plymouth, shutdown, kexec, SELinux, initrd+initrd-less boots. Status: almost made Fedora 14. Substantial coverage of basic OS boot-up tasks, including fsck, mount, quota, Kay Sievers, Lennart Poettering Beyond Init: systemd readahead, tmpfiles, random-seed, console, static module loading, early syslog, plymouth, shutdown, kexec, SELinux, initrd+initrd-less boots.
    [Show full text]
  • Using the Linux NFS Client with IBM System Storage N Series
    Front cover Using the Linux NFS Client with IBM System Storage N series Kernel releases and mount options Tuning the Linux client for performance and reliability Utilities to support advanced NFS features Alex Osuna Eva Ho Chuck Lever ibm.com/redbooks International Technical Support Organization Using the Linux NFS Client with IBM System Storage N series May 2007 SG24-7462-00 Note: Before using this information and the product it supports, read the information in “Notices” on page v. First Edition (May 2007) This edition applies to Version 7.1 of Data ONTAP and later. © Copyright International Business Machines Corporation 2007. All rights reserved. Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp. Contents Notices . .v Trademarks . vi Preface . vii The team that wrote this book . vii Become a published author . vii Comments welcome. viii Chapter 1. Linux NSF client overview and solution design considerations . 1 1.1 Introduction . 2 1.2 Deciding which Linux NFS client is right for you . 2 1.2.1 Identifying kernel releases . 3 1.2.2 Current Linux distributions . 4 1.2.3 The NFS client in the 2.4 kernel . 5 1.2.4 The NFS client in the 2.6 kernel . 6 1.3 Mount options for Linux NFS clients . 7 1.4 Choosing a network transport protocol . 12 1.5 Capping the size of read and write operations . 16 Chapter 2. Mount options. 19 2.1 Special mount options. 20 2.2 Tuning NFS client cache behavior . 21 2.3 Mounting with NFS version 4 .
    [Show full text]
  • Troubleshooting Made Easy in Linux
    Expert Reference Series of White Papers Troubleshooting Made Easy In Linux 1-800-COURSES www.globalknowledge.com Troubleshooting Made Easy In Linux David Egan, RHCX Introduction Part 1: Troubleshooting Overview Part 2: Getting Started Part 3: Getting Services Started Part 1: Troubleshooting Overview Every operating system portrays itself as powerful, able to leap tall buildings and applications in single strides, able to weather the storms of prolonged heavy usage, resilient to weathering, and so much more! One of the necessary evils, albeit less often spoken of, in case we jinx something, is the topic of troubleshooting. What could possibly go wrong? Doesn’t everyone believe in fair play, honorable intentions, and being openly honest and straightforward in all things virtual? Where does one start when the operating system does not start, the application does not load or run correctly, the network hangs or a network service does not listen or perform? A common practice is to reload the machine, either from the installation media and then patch it, or from some standardized image. Either way, the problem has only been gotten around, not solved. What if you want to solve the problem itself? What if you need to solve the problem to move ahead? Linux is noted for being rock solid, stable, and easy to manage but not as user friendly ... ‘Where do I click?’ is usually not the answer. There is an ancient, and maybe black, dark, or mysterious ‘art’ form known as troubleshooting. The idea of troubleshooting problems when they occur, whether with the system bootup or setup, networking or network services, or applications in memory, is a daunting task for anyone.
    [Show full text]
  • Evolving Ext4 for Shingled Disks Abutalib Aghayev Theodore Ts’O Garth Gibson Peter Desnoyers Carnegie Mellon University Google, Inc
    Evolving Ext4 for Shingled Disks Abutalib Aghayev Theodore Ts’o Garth Gibson Peter Desnoyers Carnegie Mellon University Google, Inc. Carnegie Mellon University Northeastern University Abstract ST5000AS0011 Drive-Managed SMR (Shingled Magnetic Recording) disks 30 ST8000AS0002 10 ST4000LM016 offer a plug-compatible higher-capacity replacement for WD40NMZW conventional disks. For non-sequential workloads, these 3 WD5000YS disks show bimodal behavior: After a short period of high 1 throughput they enter a continuous period of low throughput. 0.3 We introduce ext4-lazy1, a small change to the Linux 0.1 Throughput (MiB/s) ext4 file system that significantly improves the throughput 0.03 in both modes. We present benchmarks on four different 0.01 drive-managed SMR disks from two vendors, showing that 0 100 200 300 400 500 ext4-lazy achieves 1.7-5.4× improvement over ext4 on a Time (s) metadata-light file server benchmark. On metadata-heavy Figure 1: Throughput of CMR and DM-SMR disks from Table 1 under benchmarks it achieves 2-13× improvement over ext4 on 4 KiB random write traffic. CMR disk has a stable but low throughput under drive-managed SMR disks as well as on conventional disks. random writes. DM-SMR disks, on the other hand, have a short period of high throughput followed by a continuous period of ultra-low throughput. 1 Introduction Type Vendor Model Capacity Form Factor Over 90% of all data in the world has been generated over the DM-SMR Seagate ST8000AS0002 8 TB 3.5 inch last two years [14]. To cope with the exponential growth of DM-SMR Seagate ST5000AS0011 5 TB 3.5 inch data, as well as to stay competitive with NAND flash-based DM-SMR Seagate ST4000LM016 4 TB 2.5 inch solid state drives (SSDs), hard disk vendors are researching DM-SMR Western Digital WD40NMZW 4 TB 2.5 inch capacity-increasing technologies like Shingled Magnetic CMR Western Digital WD5000YS 500 GB 3.5 inch Recording (SMR) [20,60], Heat Assisted Magnetic Record- ing (HAMR) [29], and Bit-Patterned Magnetic Recording Table 1: CMR and DM-SMR disks from two vendors used for evaluation.
    [Show full text]
  • SUSE Linux Enterprise Server 12 SP4 System Analysis and Tuning Guide System Analysis and Tuning Guide SUSE Linux Enterprise Server 12 SP4
    SUSE Linux Enterprise Server 12 SP4 System Analysis and Tuning Guide System Analysis and Tuning Guide SUSE Linux Enterprise Server 12 SP4 An administrator's guide for problem detection, resolution and optimization. Find how to inspect and optimize your system by means of monitoring tools and how to eciently manage resources. Also contains an overview of common problems and solutions and of additional help and documentation resources. Publication Date: September 24, 2021 SUSE LLC 1800 South Novell Place Provo, UT 84606 USA https://documentation.suse.com Copyright © 2006– 2021 SUSE LLC and contributors. All rights reserved. Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or (at your option) version 1.3; with the Invariant Section being this copyright notice and license. A copy of the license version 1.2 is included in the section entitled “GNU Free Documentation License”. For SUSE trademarks, see https://www.suse.com/company/legal/ . All other third-party trademarks are the property of their respective owners. Trademark symbols (®, ™ etc.) denote trademarks of SUSE and its aliates. Asterisks (*) denote third-party trademarks. All information found in this book has been compiled with utmost attention to detail. However, this does not guarantee complete accuracy. Neither SUSE LLC, its aliates, the authors nor the translators shall be held liable for possible errors or the consequences thereof. Contents About This Guide xii 1 Available Documentation xiii
    [Show full text]
  • Containers Are Linux, Run/Optimized and Tuned Just Like Linux
    Performance Analysis and Tuning – Part I Containers are Linux, run/optimized and tuned just like Linux. D. John Shakshober (Shak) – Tech Director Performance Engineering Larry Woodman - Senior Consulting Engineer Bill Gray – Senior Principal Performance Engineer Joe Mario - Senior Principal Performance Engineer Agenda: Performance Analysis Tuning Part I • Part I - Containers are Linux, run/optimized and tuned just like Linux. • RHEL Evolution 5->6->7 – Hybrid Clouds Atomic / OSE / RHOP • System Performance/Tools Tuned profiles • NonUniform Memory Access (NUMA) • What is NUMA, RHEL Architecture, Auto-NUMA-Balance • Network Performance – noHZ, Throughput vs Latency-performance • Tuna – IRQ pinning, alter priorities, monitor • NFV w/ DPDK fastdata path • Perf advanced features, BW monitoring, Cache-line tears C-2-C •“Performance + Scale Experts” - 205C - 5:30-7 PM • Free - Soda/Beer/Wine Red Hat Enterprise Linux Performance Evolution • RHEL5 • 1000 Hz, CFQ IO elevator, ktune to change to deadline • Numactl, taskset affinity, static hugepages, IRQ balance, oprofile • RHEL 6 • Tickless scheduler CFS, islocpus, userspace NUMAD tool • Transparent hugepages (THP), numa-IRQ balance, cGroups • Tuna, Perf, PCP ship w/ OS • RHEL 7 • NoHZ_full for CFQ, islocpu, realtime ship same RHEL7 kernel • AutoNuma balance, THP, systemd – Atomic containers RHEL Performance Workload Coverage (bare metal, KVM virt w/ RHEV and/or OSP, LXC Kube/OSEand Industry Standard Benchmarks) Benchmarks – code path coverage Application Performance ● CPU – linpack, lmbench ● Linpack
    [Show full text]