Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

Operations [clear filter]
Monday, September 16

10:35am CDT

Shouldn’t All Your Storage be Highly Available? - Gerry Smith, LSI
What is blocking that as a reality for your business? Is it Cost? Complexity? Lack of storage 
choices? Today’s reality is that business uptime is more critical than ever.  It translates directly to business viability and profitability.  However, we also know that businesses of all sizes are facing the double challenge of explosive data growth and IT budget pressure.  So, how do businesses achieve the goal of high availability in a simple, cost-effective way? Mr. Gerry Smith, LSI CTO, will discuss how the industry is looking at evolving DAS storage, making it sharable and scalable.  He will discuss how having high-availability in a DAS environment, combined with Linux support, will enable business benefits including uptime, ease-of-use, and reduced TCO.  Session attendees will learn from a technologist viewpoint, the significance of this technology transition, how it works, and how they can get started applying it to their businesses.

This session targets a technical audience, including developers and system administrators.  
Session attendees will learn from a technologist viewpoint, the significance of high availability 
direct attached storage, how it works in detail, and how they can apply it to solve their business needs.

avatar for Gerry Smith

Gerry Smith

VP Architecture, LSI
Gerry Smith is vice president of Architecture for the RAID Storage Division of LSI Corporation. In this role, he leads the architectural definition of future products, including silicon, firmware and software for LSI’s industry-leading I/O controller and MegaRAID product lines.Previously... Read More →

Monday September 16, 2013 10:35am - 11:25am CDT
Strand 11B

11:45am CDT

LXC, Docker, and the Future of Software Delivery - Jerome Petazzoni, dotCloud
LinuX Containers (LXC) have aroused interest for the past several years, as a lightweight, more efficient, and more flexible approach to virtualization. Containers  have a lower footprint, less overhead, and boot faster than conventional VMs However, they haven't been widely deployed yet. Recently, the emergence of complementary open-source technologies, such as AUFS BTTRFS, and Docker, have made it significantly easier to exploit and automate this powerful, Linux-based technology. It is now possible to take any application and its dependencies, encapsulate them in a portable, lightweight container, and run that same container on virtually any Linux server The same containerized application that runs successfully on a developer's laptop will run consistently and quickly on a baremetal server, an OpenStack cluster, public cloud, VM, etc. In this presentation, we will discuss how to use this powerful combination of technologies to enable large scale automated systems, such as web apps, databases, and backend services, without depending on a particular stack or provider. We will also discuss the application of these technologies to continuous integration and deployment and hybrid cloud.

avatar for Jerome Petazzoni

Jerome Petazzoni

Tinkerer Extraordinaire, Docker Inc.
Jerome is a senior engineer at Docker, where he helps others to containerize all the things. In another life he built and operated Xen clouds when EC2 was just the name of a plane, developed a GIS to deploy fiber interconnects through the French subway, managed commando deployments... Read More →

Monday September 16, 2013 11:45am - 12:35pm CDT
Strand 11B

3:10pm CDT

Resource Management with Systemd - Lennart Poettering, Red Hat
Resource management is highly relevant on both large servers (where many resources need to be distributed among an even higher number of consumers to get the maximum out of the invested infrastructure) and on small embedded devices (where few resources need to be distributed among many consumers). Newer systemd versions support resource management out-of-the-box, as first-class functionality of service management. In this talk we will show you how to make use of it. We will tell you how to dynamically assign resource limits to running services and virtual machines and how to reliably determine the current resource usage of your services and virtual machines. We will introduce you to new concepts such as "Slices" which you can use to create arbitrary resource partition hierarchies. And we will give you a bit of background on what actually happens on the lower level.


Lennart Poettering

Principal Software Engineer, Red Hat
Lennart works on systemd, for Red Hat.

Monday September 16, 2013 3:10pm - 4:00pm CDT
Strand 11B

4:10pm CDT

Reliability Availability and Serviceability (RAS) on Linux - Mauro Carvalho Chehab, Samsung
The largest are the systems and the IT infrastructure, the higher is the importance of Reliability Availability and Serviceability (RAS) monitoring. This speech will describe the recent changes on RAS monitoring that are available at the Linux Kernel 3.10. It will also describe the rasdaemon monitoring tool, with uses the special Kernel perf events generated by the Kernel to monitor fatal and non-fatal hardware errors that are detected by the CPU, by the memory controller and by the PCIe hardware.


Mauro Carvalho Chehab

OS Senior Engineer, Huawei Technologies Düsseldorf GmbH
Mauro is the maintainer of the media subsystem, and a large contributor for the Reliability Availability and Serviceability (RAS) subystems at the Linux Kernel. He has a broad experience in both Telecommunications (as he worked at IT, network and management areas at the Brazil's largest... Read More →

Monday September 16, 2013 4:10pm - 5:00pm CDT
Strand 11B

5:10pm CDT

Enterprise Identity Management with Open Source Tools - Dmitri Pal, Red Hat
The talk will cover the FreeIPA and partially SSSD projects introducing FreeIPA as the first fully functional open source centralized authentication and identity management solution comparable to Active Directory in its capabilities but focusing primarily on the identity management needs of the Linux and UNIX client systems.

The talk would target system administrators that are looking for efficient ways of centrally managing authentication, identities and related policies for Linux systems as a part of complex heterogeneous environments. The talk will suggest ways on how Linux systems can be most efficiently managed and integrated into AD environment. The goal is to make audience aware of the newest technologies available in the identity management space. The complexity is medium.

avatar for Dmitri Pal

Dmitri Pal

Director, Software Engineering, Red Hat
Dmitri Pal is a Director of Software Engineering at Red Hat responsible for the security and identity management projects and products in Red Hat Enterprise Linux and Red Hat portfolio. Dmitri has more than 20 years of the security and identity management-related software engineering... Read More →

Monday September 16, 2013 5:10pm - 6:00pm CDT
Strand 11B
Tuesday, September 17

10:45am CDT

SIPping from the Open Source Well - Matthew Bynum, Softchoice
This presentation will dive into where the Session Initiation Protocol (SIP) and eXtensible Messaging and Presence Protocol (XMPP) protocols are used on Linux platforms, and ultimately why SIP is the best thing to hit our communications infrastructure since the Central Office (CO).  I'll discuss DNS SRV in bind to enable SIP calls via domain lookups of User Resource Identifiers (URI).  I'll also go over Linux desktop and Android clients that  can register to various call control platforms.

This presentation is geared for Linux savvy Network Administrators who are interested in how to open up their communications environments, both by using Open Source and open standards.  They can expect to walk away with a roadmap that can help guide them into the next evolution of voice and video communication.

avatar for Matthew Bynum

Matthew Bynum

UC Practice Leader, Softchoice
Matt Bynum has been in IT for 12 years, and has been everything from a PC support tech to Systems Administrator to his current role, that of the Unified Communications Practice Leader at a company called Softchoice. Matt currently focuses in Unified Communications, but has had experience... Read More →

Tuesday September 17, 2013 10:45am - 11:35am CDT
Strand 11B

11:45am CDT

'Tickless' Kernel: Practical Experiences - Christoph Lameter & Fernando Garcia
A standard Linux kernel is typically configured to interrupt an application 1000 times per second to check for things that the kernel may have to do. These interrupts cause the application to experience seemingly random latencies and variations in runtime.  As of Linux 3.10 the kernel has the ability to switch the tick off in certain situations. However, there are numerous gotchas right now that have to be considered. The functionality is almost ready but very difficult to use. Here we will talk about experiences with this functionality, what other measures can be used to reduce OS noise and discuss how we think a tickless kernel should be working in the future. A new benchmark will be used to show what improvements are possible.  Suggestions are wanted as to how to make it easier to use a tickless kernel. The system configuration is rather complex at this point.


Fernando Garcia

Fernando has been involved for the last 5 years in various projects relating to high performance computing and low latency environments and is an expert in the configuration of systems for extremely noiseless operations.
avatar for Christoph Lameter

Christoph Lameter

R&D Team Lead, Jump Trading LLC
Christoph Lameter is working as a lead in research and development for Jump Trading LLC (an algorithmic trading company) in Chicago and maintains the slab allocators and the per cpu subsystems in the Linux Kernel. He contributed to a number of Linux projects since the initial kernel... Read More →

Tuesday September 17, 2013 11:45am - 12:35pm CDT
Strand 11B

3:10pm CDT

Prescriptive Topology Deamon - Pradosh Mohapatra, & Dinesh Dutt, Cumulus Networks
In data center topologies, right cabling is a time consuming endeavor and is error-prone. PTMD is a dynamic cabling verification tool to help eliminate such errors. It takes a graphviz-DOT specified network cabling plan (something many operators already generate) and couples it with runtime information derived from LLDP to verify that the cabling matches the specification. The check is performed on every link transition on each node in the network. PTMD runs as a Linux daemon. It has been tested on switches running Linux and by the time of the talk, the code will be hosted on github. The talk is primarily targetted for system administrators. The audience will get to know: Design and implementation of PTMD, Configuration examples, Use cases such as integration with Quagga-based routing protocol suite.

avatar for Dinesh Dutt

Dinesh Dutt

Chief Scientist, Cumulus Networks
Dinesh Dutt is the CSO of Cumulus Networks.

Pradosh Mohapatra

Software Engineer, Cumulus Networks
Pradosh Mohapatra and Dinesh Dutt. One of the authors will present. Pradosh Mohapatra is a Linux software developer at Cumulus Networks. His expertise is in building routers and implementation of routing protocols. Until recently, he was a distinguished engineer at Cisco systems... Read More →

Tuesday September 17, 2013 3:10pm - 4:00pm CDT
Strand 11B

4:10pm CDT

(Tutorial) High Availability Solutions for MySQL and MariaDB - Max Mether, MySQL AB
Achieving High Availability with MySQL is a tricky thing; there are almost as many recipes as there are cooks out there. This talk will go through the most common solutions for achieving HA with MySQL and list the pros and cons for each of them. The solutions include standard replication, MariaDB multi-source replication, MHA, Continuent Tungsten, Galera, shared-disk based solutions and MySQL Cluster

This talk is aimed towards MySQL DBAs and Developers who are interested in setting up HA architecture for their data. After this talk they should have a clear overview of all the options available. 


Max Mether

Max Mether, a native of Finland received his M.Sc (Eng) in Physics and Maths from Helsinki University of Technology. Max joined MySQL AB in 2001 starting as a Consultant and an Instructor and ended up creating the MySQL training program and managing the curriculum under MySQL Ab and... Read More →

Tuesday September 17, 2013 4:10pm - 6:00pm CDT
Strand 11B
Wednesday, September 18

10:50am CDT

Architecting Block and Object Geo-Replication Solutions with Ceph - Sage Weil, Inktank
Ceph is a fully open source distributed object store, network block device, and file system designed for reliability, performance, and scalability from terabytes to exabytes. For many organizations, storage systems that span multiple data centers and geographies for disaster recovery or follow-the-sun purposes are an important requirement. This talk will give a brief overview of the Ceph architecture, and then focus on the design and implementation of asynchronous geo-replication and disaster recovery features for the RESTful object storage layer, the RBD block service, and Ceph's underlying distributed object store, RADOS. The fundamental requirements for a robust georeplication solution (like point in time consistency) and the differing requirements for each storage use-case and API and the implications for the asynchronous replication strategy will be discussed.

avatar for Sage Weil

Sage Weil

CTO, Inktank
Sage Weil is the creator of the Ceph project. Sage originally designed Ceph as part of his PhD research in Storage Systems at the University of California, Santa Cruz. Since graduating, he has continued to refine the system with the goal of providing a stable next generation distributed... Read More →

Wednesday September 18, 2013 10:50am - 11:40am CDT
Strand 10B

2:00pm CDT

Best practices for MySQL High Availability - Colin Charles, MariaDB
The MySQL world is full of tradeoffs and choosing a High Availability (HA) solution is no exception. This session aims to look at all of the alternatives in an unbiased nature. Preference is of course only given to opensource solutions. How do you choose between: MySQL replication, MHA, DRBD, Tungsten, Galera Cluster? Do you integrate Pacemaker and Heartbeat like Percona Replication Manager? Is NDB Cluster right for your workload? The cloud brings even more fun as geographical redundancy is rolled into this.

Audience: MySQL users who are growing with the product. Expect good discussion and worksheets to take back to perform the own evaluation. 

avatar for Colin Charles

Colin Charles

Consultant, GrokOpen
Colin Charles is the Managing Consultant at GrokOpen. Previously, Colin was on the founding team of MariaDB Server, worked at MySQL and Percona, and worked actively on the Fedora and OpenOffice.org projects. Colin has been a MySQL user since 2000. He’s well known within open source communities, enjoys building business and market entry in APAC and has spoken at many conferences... Read More →

Wednesday September 18, 2013 2:00pm - 2:50pm CDT
Strand 11B

3:00pm CDT

What is so Special About Mageia ? - Bruno Cornec, HP
Now ranked #2 on DistroWatch, this fully community driven FLOSS distribution is still largely unknown and its merits underestimated.  However, it provides unique features for both server or desktop users, based on a rich management, packaging and contribution set of tools, a friendly community eager to deliver the best Linux distribution. While young and born in 2010, Mageia is a full featured distribution inheriting from its Mandrake roots back in the 90's, but now fully managed by its community such as Debian, or Fedora. However, few people know what makes Mageia so appealing. This presentation will cover in detail deployment with auto_inst, security with msec, SELinux and Tomoyo, package managemen with urpmi and friends, and the numerous other tools (providing a real choice), or integration tasks or governance model that make it a unique distribution you should test next.

avatar for Bruno Cornec

Bruno Cornec

Open Source & Technology Strategist, HPE
Bruno Cornec has been managing various Unix systems since 1987 and Linux since 1993 (0.99pl14).Bruno first worked 8 years around Software Engineering and Configuration Management Systems in Unix environments.Since 1995, he is Open Source and Linux (OSL) Technology Strategist, Linux... Read More →

Wednesday September 18, 2013 3:00pm - 3:50pm CDT
Strand 12B