Efforts to improve MySQL's performance have been paying off in the recent past. This talk covers the current state of MySQL performance, both in the server and the InnoDB storage engine, where much of the work has concentrated. We look at what is likely to come next: what is in progress, and problems we're aware of but haven't yet solved.

SANs, even for one datacenter, are usually expensive; when it comes to replicating your data to a remote site costs can easily go beyond scope. With LINBITs open source storage replication solution DRBD not only your on-site data is safe - even backup-sites are capable of having the same consitsent data based on the same free-of-charge technology. Immediate and affordable site-failover has become reality with DRBD.

In this lecture we will show how to build a complete iSCSI, DRBD and pacemaker SAN "drop-in" alternative that even over high-latency IP-routes allows prompt replication to your remote site. You will also learn how such an iSCSI target within seconds can be physically relocated in your data centre and can be brought online within minutes in your disaster recovery location.

This lecture implies basic SAN and iSCSI skills, knowledge of open source- and cluster management might be useful.

High availability is a basic requirement for the IT department nowadays. The best known solution in this area is Linux-HA. The first verison dates back to 1998, the second version evolved 2005 but prooved to be a temporary solution.

In 2007 it was replaced by a collection of other programs. The evolution of the software might be difficult to understand for a beginner. This talk tries to bring order into the several names of the programs, explains the setup of such clusters and tries to take a glance into the not too far future of the software.

This presentation will highlight interesting features such as ZFS Zones, resource management and Crossbow network virtualization. Moreover, it will compare the different systems and development philosophies. Note: the presentation will involve command line demonstrations.

War es bis vor einiger Zeit ein sehr teures Unterfangen, Lösungen für modernes Systems Management zu implementieren, so gibt es heute für alle wichtigen Prozesse und Funktionen des Service Management sehr gute Open Source Tools. Der Vortrag bietet einen überblick über die verschiedenen Lösungen in den wichtigsten Bereichen: Incident & Problem Management, Event Management, Operations Management, Service Desk und CMDB.

Storage is one of the most important part of a data center, the complexity to design, build and delivering 24/forever availability service continues to increase every year. For these problems one of the best solution is a distributed filesystem (DFS) This talk describes the basic architectures of DFS and comparison among different free software solutions in order to show what makes DFS suitable for large-scale distributed environments. We explain how to use, to deploy, advantages and disadvantages, performance and layout on each solutions. We also introduce some Case Studies on implementations based on openAFS, GlusterFS and Hadoop finalized to build your own Cloud Storage.

People are frustrated with slow web pages. At XING we know how important good performance is. Xing was founded in 2003 as OpenBC, and with just one server started a global business network. We enjoyed a growing and active user-base since then, and now have more than 300 servers providing a fault tolerant and scalable business social network for our customers.

Within our operation we mainly use open source solutions (Linux, Apache, MySQL, and Perl/Java/Ruby). To handle the high traffic we use many different scaling techniques which we want to share in this talk. Besides that we would like to talk about how great performance helps businesses succeed online.

Unter dem Schlagwort der Virtualisierung sind viele Techniken zu finden. Hier soll eine Virtualisierungstechnik vorgestellt werden, welche im Gegensatz zur Voll- und Paravirtualisierung nur das Betriebssystem virtualisiert und sich die Emulation von Hardware spart. Eine Idee, die auch in Solaris Zonen oder BSD Jails zu finden ist und verspricht sehr ressourcenschonend zu sein. Ein Umstand der wohl dafür verantwortlich ist, dass sich z.B. hinter den meisten (in Zeitschriften) beworbenen VServern OpenVZ oder das kommerzielle Pendant Virtuozzo befindet. Der Vortrag führt in die Administration von Containern (auf Basis von OpenVZ) ein. So werden Sie unter anderem erfahren wie Container erstellt, dynamisch von einem Rechner auf einen anderen migriert, Minima und Maxima von Ressourcen definiert, gemonitort und dynamisch zugewiesen werden. Auch werden die Vor- und Nachteile von OpenVZ (auch zu anderen Virtualisierungstechniken) beleuchtet.

When we speak of modern IT infrastructure changes, those made to capacity and usage present some of the greatest challenges. Often systems have to be adjusted to new load requirements at short intervals, and those responsible face many components, complex relationships and demands.

This talk will deal with precisely these issues and concerns in detail. Taking all involved system levels into account, best practices and strategies will be suggested for the prevention and elimination of bottlenecks. An overview of the current solutions and strategies at the network, application framework and database level will be presented in detail. Pros and cons of each solution will be covered in evaluating their potential in the modern data center.

Ensuring data survival through any disaster is equally important as ensuring its availability. The sixty minute session will provide an overview of a large variety of methods to create proper backups of MySQL databases in many environments and show in greater detail some of the most popular and effective ones.

Mit Smart ITSM werden großartige Open-Source Werkzeuge für die System- und Serviceverwaltung zu einer einzigartigen Lösung. Als isolierte Einzelkämpfer können diese Werkzeuge ihr volles Potential allerdings nicht ausschöpfen. Was fehlt ist eine integrierte Gesamtsicht auf das System, die es erlaubt die einzelnen Prozesse zentral zu verwalten. Basierend auf der Dokumentations- und CMDB-Lösung i-doit können bereits heute Prozesse zur Überwachung (Nagios®/Icinga), dem Helpdesk (Request Tracker) oder der Inventarisierung (OCS, hInventory) eine gemeinsame, konsolidierte Sicht auf die Unternehmens-IT liefern. Der nächste Schritt auf diesem Weg heißt [sic], der Smart ITSM Converter. Der Smart ITSM Converter soll es ermöglichen, dass Prozesse, die bislang isoliert agierten, Konfigurationsdaten untereinander austauschen und somit stets auf einer einheitlichen Datenbasis arbeiten.