Red Hat Summit 2009 - Session Abstracts

Red Hat Enterprise Linux Roadmap

Tim Burke — Senior Director, Red Hat Enterprise Linux Engineering

This presentation will highlight the major features delivered in Red Hat Enterprise Linux update releases, including 5.3, 5.4, and 4.8. Tim Burke, Red Hat's senior director of Red Hat Enterprise Linux engineering, will give an overview of the Red Hat Enterprise Linux product lifecycle, highlight the product's major features, and share a glimpse into Red Hat Enterprise Linux 5.4.

New Deployment Features in Red Hat Satellite

Michael DeHaan — Senior Software Engineer, Red Hat

Jesus Rodriguez — Senior Software Engineer, Red Hat Network

While virtualization offers many benefits to users, it also poses new management challenges. In order to reduce these challenges and virtualization complexity, it is necessary to better automate both physical and virtual deployments and change how provisioning configurations are managed.

The latest Red Hat Satellite Server contains an embedded version of the Cobbler install server for advanced deployment features. Cobbler, a next-generation installation server that abstracts out the differences between physical and virtual installations, allows for rapid setup of Linux deployments for physical and virtual systems.

In this presentation, Red Hat Satellite Server and Cobbler engineers will discuss deploying Linux systems using the newest version of Red Hat Satellite. Attendees will learn about new features available in Red Hat Satellite, an intuitive open source systems management platform, for physical and virtual machines.

Clustering Virtual Machines and Applications with Red Hat Enterprise Linux

Thomas Cameron — Enterprise Solutions Architect, Red Hat

Lon Hohberger — Senior Software Engineer, Red Hat

Red Hat's Thomas Cameron and Lon Hohberger will expand upon their presentation “Cluster Failover for any Application – Demonstration: A High Availability Cluster Installation,” which was given at last year's Red Hat Summit. At this year's presentation, Cameron and Hohberger will discuss the theory and best practice of clustering virtual machines and applications with Red Hat Enterprise Linux.

In addition Cameron and Hohberger will demonstrate the installation and configuration of a high-availability cluster of virtual machines using Red Hat Enterprise Linux Advanced Platform. They will give a live demonstration of Red Hat Global File System, Red Hat Cluster Suite functionality, and the Conga web-based cluster management tool. The demonstrations will include:

• Setting up an iSCSI target (server), iSCSI initiators (clients), and clustered logical volumes
• Deploying and clustering virtual machines
• Managing applications on the cluster

Open Source Video - A Guide to Free Self-hosted Online Video Apps

Leah Belsky — Director of Strategic Accounts & Assoc. General Counsel, Kaltura

Video lovers of the world unite. Leah Belsky, Director of Strategic Accounts & Assoc. General Counsel of Kaltura, will present the world's first full open source video solution stack (used by Wikipedia and 35,000 other publishers), and demo several self-hosted video applications. She'll walk through technicalities of setting up an online video platform, discuss pros and cons of self-hosted versus SaaS, and even dive into some code.

Red Hat Virtualization Leadership Series: Breaking Performance and Scalability Barriers

Sanjay Rao — Senior Software Engineer, Red Hat

D. John Shakshober (Shak) — Senior Consulting Engineer, Red Hat

Vijay Trehan — Solutions Architect, Red Hat

Virtualization is typically used for virtualizing small workloads such as development and test machines. Mission critical or enterprise class workloads are seldom virtualized due to limitations with the traditional virtualization solutions. In this session John Shakshober (Shak), Red Hat senior consulting engineer, and Vijay Trehan, a Red Hat solutions architect, will discuss the scalability features of the Red Hat Enterprise Virtualization platform allowing customers to deploy even the largest mission critical workloads without sacrificing performance.

This session will cover the performance and scaling of Enterprise Applications in a Red Hat Enterprise virtualization environment. Shak and Trehan will also cover:

• Deploying enterprise applications in Red Hat Enterprise virtualization environment
• Performance characterization of Oracle, SAP, LAMP, Exchange, IIS
• Virtualization efficiency when scaling-up (few large guests)
• Virtualization efficiency when scaling-out (many small guests)

Shak and Trehan will also cover the performance characteristics of Red Hat Enterprise Virtualization with the Linux/KVM hypervisor and demonstrate how enterprises can virtualize even their most I/O intensive workloads, such as databases and ERP systems. This session will cover performance tuning, best practices and competitive positioning of Red Hat Enterprise Linux virtualization technology. Shak and Trehan will also cover:

• Impact of network stack options, I/O options, VT-D, NUMA, large pages, etc.
• How to achieve best DB results
• How to achieve best messaging results
• Comparison with Hyper-V and VMware (PT work)
• Future performance enhancements with Red Hat Enterprise Linux 6

Building and Leveraging Compute Clouds with Red Hat Enterprise MRG

Bryan Che — Project Manager, Red Hat

Mike Santangelo — GPS Linux Consultant, Red Hat

Chad Tindel — Solutions Architect, Red Hat

Shawn Lower — Consultant, Red Hat

There are a wide variety of technologies, value propositions, and strategies for taking advantage of cloud computing. Red Hat Enterprise MRG provides a powerful and flexible platform for building a variety of internal and hybrid cloud solutions and for leveraging external clouds. MRG provides the capability to provision dynamic capacity at Amazon EC2, send work to EC2, retrieve results, and then tear down EC2 resources when finished—all from the same interface that executes local workloads and applications.

This presentation will provide an overview of MRG's cloud computing capabilities. Red Hat's Bryan Che, Mike Santangelo, and Chad Tindel will describe how to build internal and hybrid clouds and how to leverage external computing clouds with Red Hat Enterprise MRG. They will also detail solutions and techniques for clouds, provide a roadmap of upcoming cloud-related features in MRG, and present a customer use case focused on building clouds.

Birds of a Feather: Cloud Data Analytics

Tom Plunkett — Cloud Computing Architect, Serene Software

A new generation of data analytics tools will be needed to handle petabytes of cloud data. This Birds of a Feather discussion will highlight open source software projects for cloud computing and data analytics, including Apache Hadoop. Hadoop is built on the MapReduce algorithm invented by Google for large-scale, reliable, inexpensive analytics. Tom Plunkett, a cloud computing architect at Serene Software, will discuss the criteria for determining whether a domain problem is suitable to be solved using this technology and/or alternatives. Example use cases include Event Detection, Fraud Analysis, Relationship Maps, Relevance Ranking, Trend Analysis, Unstructured Data Analysis, and Video/Imagery Analysis.

Part I: Performance Analysis and Tuning of Red Hat Enterprise Linux

D. John Shakshober (Shak) — Senior Consulting Engineer, Red Hat

Larry Woodman — Consulting Software Engineer, Red Hat

This two-hour presentation, which is divided into two segments, will explore the system performance analysis and tuning necessary to maximize the performance of systems running RHEL5, RHEL4, and Red Hat Enterprise MRG.

This presentation from Doug Shakshober, a Red Hat senior consulting engineer, will focus on large systems (e.g., the latest x86 hardware platforms, quad/hex cores, >100GB memory, 10 Gb Ethernet) that run the most common applications (e.g., database servers, Internet servers, and various financial applications).

The two-hour presentation will:

• Provide information for determining the system capacity for various hardware resources
• Illustrate how to analyze the system performance using several performance analysis tools and techniques, which include OProfile and SystemTap
• Focus on the tuning of Red Hat Enterprise Linux and Red Hat Enterprise MRG
• Highlight the Red Hat Enterprise Linux tuning parameters while focusing on how these parameters affect the system running various workloads
• Contrast the Red Hat Enterprise Linux tuning parameters between different versions of Red Hat Enterprise Linux

Part II: Performance Analysis and Tuning of Red Hat Enterprise Linux

D. John Shakshober — Senior Consulting Engineer, Red Hat

Larry Woodman — Consulting Software Engineer, Red Hat

This two-hour presentation, which is divided into two segments, will explore the system performance analysis and tuning necessary to maximize the performance of systems running RHEL5, RHEL4, and Red Hat Enterprise MRG.

This presentation from Doug Shakshober, a Red Hat senior consulting engineer, will focus on large systems (e.g., the latest x86 hardware platforms, quad/hex cores, >100GB memory, 10 Gb Ethernet) that run the most common applications (e.g., database servers, Internet servers, and various financial applications).

The two-hour presentation will:

• Provide information for determining the system capacity for various hardware resources
• Illustrate how to analyze the system performance using several performance analysis tools and techniques, which include OProfile and SystemTap
• Focus on the tuning of Red Hat Enterprise Linux and Red Hat Enterprise MRG
• Highlight the Red Hat Enterprise Linux tuning parameters while focusing on how these parameters affect the system running various workloads
• Contrast the Red Hat Enterprise Linux tuning parameters between different versions of Red Hat Enterprise Linux

MRG Messaging: A Programmer's Overview

Jonathan Robie — Software Engineer, Red Hat

Red Hat Enterprise MRG is an open source messaging system based on Advanced Message Queuing Protocol (AMPQ), an open standard for enterprise messaging. Red Hat Enterprise MRG allows users to write applications for simple, high-speed messaging in multiple languages (Java JMS, C++, Python, Ruby) and it allows users to run the applications on multiple platforms (Linux, Unix, Macintosh, or Windows). Red Hat Enterprise MRG provides guaranteed delivery, speed, and security.

This presentation serves as a programmer's overview of Red Hat Enterprise MRG and is designed to teach programmers what they need to know to get started writing messaging applications. This presentation will feature:

• Overviews of Red Hat Enterprise MRG, Apache Qpid, C++, and Python Messaging APIs
• Discussion of the AMQP messaging standard and Apache Qpid
• API summary for C++ and Java
• Description of how to use Red Hat Enterprise MRG with Java JMS
• Demonstration of how to write applications based on common messaging paradigms (e.g., direct, fanout, publish-subscribe, request-response, and XML content-based routing)
• Demonstration of how to write Java JMS programs that use Red Hat Enterprise MRG
• Demonstration of how to implement persistence, messaging transactions, and failover for clients using a high-availability cluster
• Demonstration of the tools that can be used to manage the messaging server
• Review of the configuration and the state of messages in the system

Tuning Red Hat Enterprise Linux for Oracle and Oracle RAC

Scott Croft — Linux Consultant, Red Hat

While the default configuration for Red Hat Enterprise Linux systems, with respect to Oracle and Oracle RAC databases, is adequate for performance, adjustments are required as the database grows and more nodes are added to a cluster. Therefore, careful consideration of network, memory, filesystem layout, and kernel settings is necessary when tuning large configurations for maximum database efficiency and performance.

Scott Croft, a Red Hat Linux consultant, will detail these considerations so attendees understand how to maximize database efficiency and performance. He will discuss installation and kickstart configuration, filesystem layout, network configuration (including channel bonding), and kernel adjustments.

Secure Virtualization with SELinux - SVIRT

Daniel Walsh — Principal Software Engineer, Red Hat

Running multiple virtual operating systems on one server creates a range of new security risks. SELinux, when combined with libvirt, allows system administrators to control the isolation of virtual machines with the use of mandatory access control.

In this presentation Walsh will outline the security implications of virtualized environments and effective methods to deal with them.

Collaborative Innovation

Melanie Chernoff — Corporate Affairs Public Policy Manager, Red Hat

Linux grew from 10,000 lines of code in 1991 to 204 million lines of code in 2008. Similarly Wikipedia, which was launched in 2001, has rapidly grown to become one of the world's most popular websites, with over 10 million entries in 275 languages. The open source model of collaborative innovation is leading to some of the largest collaborations in the world.

In this presentation Melanie Chernoff, public policy manager of corporate affairs at Red Hat, will explore the underlying factors driving collaborative innovation. Chernoff will focus on how these factors can be applied to solving the challenges in education, e-government, medical research, and other areas.

SPICE: Enabling the Standard Desktop Remoting Protocol

Arnon Gilboa — Senior Software Engineer, Red Hat

Accessing a remote display of a virtual machine (VM) is not always a satisfying experience. Connecting a VM playing video (mpg, YouTube etc.) using existing solutions like VNC or RDP might be frustrating. This process is not only slow and far away from native performance, but it also requires high CPU and bandwidth consumption.

SPICE is a complete new open source solution for efficient remote display access. SPICE was born because VM users deserve a better solution. The goals of SPICE were to create a user experience as close as possible to a physical machine, which includes offering high quality video (and audio), for both LAN and WAN connections, without compromising the density on the host running the VMs.

In this presentation Arnon Gilboa, a Red Hat software developer from the SPICE team, will present SPICE's overall architecture, its key features and compare it with other available solutions.

IBM: Leading Enablement of Enterprise Cloud Computing

Dr. Robert S. Sutor — Vice President, Open Source and Linux, IBM Software Group

The world is smaller, flatter and smarter. Join the IBM Cloud leadership team to learn more about what IBM is doing to address IT pain points through accelerated adoption of Cloud Computing environments. Learn how IBM is creating new consumption and delivery models to drive new sourcing options and business flexibility, with new choices to deploy workloads that matter to enterprises with greater efficiency, productivity and control, especially those running on Red Hat Enterprise Linux®.

Simplifying Linux iSCSI Management with iSNS

Shyam Iyer — Development Engineer Senior Analyst, Dell

Mike Christie — Software Engineer, Red Hat

Storage provisioning is a cumbersome, time consuming and complex procedure in an environment of thousands of servers and tens of storage arrays.

Through standard protocols like iSCSI and iSNS, storage management can be highly simplified by creating a plug-and-play environment for iSCSI SANs.

This presentation seeks to help attendees understand simpler, faster ways to achieve storage LUN/volume provisioning through a single management console.

Introducing Red Hat Enterprise Virtualization

Navin Thadani — Senior Director, Virtualization Business, Red Hat

In this session, Navin Thadani will introduce the Red Hat Enterprise virtualization platform. Thadani will cover everything from the high-level product architecture of  Red Hat Enterprise Linux 5.4 and Red Hat's new Red Hat Enterprise Linux virtualization hypervisor to the rich Red Hat Enterprise Linux virtualization management platform with features such as high availability, live migration, clustering, image management and system scheduler. He will also outline Red Hat's roadmap for virtualization over the next 12 to 18 months.

Consolidation on HP ProLiant Servers - A Virtualization Reference Architecture

Thomas Sjolshagen — Linux Product Marketing Manager, HP

Being comfortable with virtualization on x86 platforms, you're looking to reduce as much cost as possible and consolidate a large number of systems onto as few, modern servers as your performance requirements permit. This session will try to answer, using the work completed by HP engineers as part of the HP ProLiant DL 785 with Red Hat Enterprise Linux 5.4 and KVM Virtualization Reference Architecture, some of the questions that you may have regarding how to optimally configure your system.

Thomas Sjolshagen, Linux Product Marketing Manager at HP will share tips and tricks that you can benefit from when tuning, managing and configuring an environment. Sjolshagen will also address the following questions:

• How many KVM guests can you successfully expect to run on a single HP ProLiant system?
• What are some of the workloads you can virtualize with a KVM-based system?
• What - if any - are the trade-offs you're making?
• How do you configure the environment to ensure availability?

Red Hat Enterprise MRG (Messaging, Realtime, Grid) Update and Roadmap

Bryan Che — Project Manager, Red Hat

Red Hat launched the first version of Red Hat Enterprise MRG at last year's Red Hat Summit. Since then MRG has achieved rapid success with Red Hat's largest customers and gained exciting new capabilities. This presentation will detail some of these advances, which include:

• Messaging capabilities like RDMA drivers, Federation, and queue semantics, which customers have used for building everything from eventing systems to stock exchanges
• Native XML processing, LVQ, Ring Queues, and TTL, which provide capabilities that developers have typically had to custom-build for themselves
• Real-time advances in performance and tooling, which includes tools for evaluating and diagnosing hardware issues for real-time performance
• Grid capabilities that provide customers with exceptional management and control capabilities for traditional HPC workloads, and also for building and leveraging clouds with features such as integrated support for virtualization and Amazon EC2.

In addition to highlighting the many capabilities and successes of Red Hat Enterprise MRG, this presentation will also include a roadmap of Red Hat Enterprise MRG's upcoming features.

Cisco's Unified Computing System: ­ A Unified Data Center Infrastructure

Joe Vaccaro — Product Manager for UCS at Cisco

Cisco's Unified Computing System was developed to address critical data center challenges by integrating network, compute, virtualization and storage access into one cohesive system while simplifying the management. This presentation will provide an overview of the Unified Computing System and highlight the key elements and innovations including:

• Stateless Computing with Service Profiles
• Extended Memory Technology
• Fiber Channel over Ethernet
• Cisco's Virtual Interface Card

In this discussion, Joe Vaccaro, Product Manager for UCS at Cisco, will discuss these details and their benefits in a Red Hat Enterprise Linux environment.

Birds of a Feather: Transforming IT into a Line of Business

Chris Bauer — NetWeaver Solution Advisor, SAP

Business-driven IT.  What does that mean?  It means making sure that IT initiatives support business processes with miminal downtime, that your infrastructure capital is invested prudently and effectively, it means you met sustainability objectives, and provide meaniful, contextual data transparency in order to measure your success and alignment with organizational goals. SAP has transformed the way businesses manage their Enterprise, Supply Chains, Product lifecycles and Customer relationships - let us show you what we can do for IT. Gain insight into how SAP customers are transforming traditional IT into a Best Run Business.

Red Hat Enterprise Linux File Systems: Today and Tomorrow

David Egts — Principal Solutions Architect, Red Hat

The time-tested file systems available in Red Hat Enterprise Linux address a wide range of storage requirements. However file systems such as ext2 and ext3 are beginning to show their age as storage capacities continue to outstrip Moore's Law and outpace access time and interface bandwidth improvements.

This presentation will begin with a discussion about the state of the file systems that are currently available in Red Hat Enterprise Linux 5. David Egts, a Red Hat principal solutions architect, will also highlight the expected storage challenges and detail the advancing file systems (e.g., ext4 and Btrfs) that are being developed to address these challenges.

Continuing Story of IBM's Consolidation on the IBM System z® with Linux

Bill Reeder — STG Sales Linux and Virtualization Enterprise Systems Architecture, Strategy and Sales STG Platform Sales Specialist, IBM

IBM's project 'Big Green' continues to spur a global shift to Linux on the Mainframe. Bill Reeder will present the latest view of how IBM has consolidated and virtualized thousands of server images onto the IBM System z®. IBM's 'Big Green' project has recognized substantial savings in energy, software and systems support costs. This is an opportunity to learn the latest information on IBM's Enterprise Computing Model and how it relates to Linux® on the mainframe.

PHP and Platform Independence in the Cloud

Wil Sinclair — Manager, Advanced Technology Group, Zend

Cloud technologies have been getting a lot of attention lately. Despite their almost irresistible advantages, vendor lock-in has been a hard pill for PHP developers to swallow. In this presentation Wil Sinclair of Zend will focus on how these services can be generalized and accessed via a common interface to make your PHP applications portable in the cloud.

Red Hat Enterprise Linux 5 Virtualization in Real Life

Vinny Valdez — Senior Product Architect, Red Hat Consulting, Red Hat

Red Hat Enterprise Linux has been shipping with integrated virtualization capabilities for over two years and many customers have successfully deployed it for various workloads and use cases. With Red Hat Enterprise virtualization, customers can now deploy virtualized infrastructure solutions that cover the server as well as the client (VDI) use case.

With its latest products, Red Hat offers a very compelling architecture for true enterprise virtualization. Red Hat virtualization technology can be used for a wide range of solutions, and helps the customer implement a more agile and adaptable infrastructure.

This presentation will include use cases and case studies that detail deployment stories from a range of different customers (from those servicing anywhere from twenty to thousands of servers). Attendees will gain insight into how customers have used virtualization to solve obvious challenges (e.g., consolidation, multi-os, deployments, etc.) and how they have used virtualization for innovative deployments (hardware independence, dynamic resource assignment, flexible disaster recovery, on-demand infrastructure deployments, etc.). Detailed architecture overviews and implementations will be shared.

RHCE and RHCT Focus Chat Sessions

Randy Russell — Director of Certification, Red Hat

The Red Hat certification team held its first Red Hat Summit focus group for RHCEs (Red Hat Certified Engineers) last year. This year the popular focus group will be extended to RHCTs (Red Hat Certified Technicians).

This focus group will give attendees the opportunity to provide guidance on trends and practices so that Red Hat can keep its certification program valuable and relevant. In addition this focus group, which is exclusively for RHCEs and RHCTs, provides an opportunity for the certification team to meet with Red Hat's certified community to discuss its plans and directions.

Stop Underutilizing your Computer

Ulrich Drepper — Consulting Engineer, Red Hat

More and more programmers have removed themselves from the underlying hardware over the years, which results in code where the performance gap to the true potential of the hardware is increasing.

In this presentation, Ulrich Drepper, consulting engineer at Red Hat, will discuss performance problems, explain how to get some of the performance back for existing code, and describe how to write code without too much hassle and without too many knowledge requirements on the side of the application developer.

Maximizing AMD Six-Core Opteron™ Processor Performance with Red Hat Enterprise Linux

Bhavna Sarathy — Red Hat Technical Lead, Advanced Micro Devices (AMD)

Sanjay Rao — Senior Software Engineer, Red Hat

Bhavna Sarathy, AMD's ISV Technical Lead for Red Hat, and Red Hat's Sanjay Rao will lead a discussion about the six-core AMD Opteron Processor ("Istanbul") HW and SW features, and will discuss how each feature lends itself to superior performance, and scalability. They will also present the SW feature support for Istanbul in Red Hat Enterprise Linux.

Scaling data will be presented showing impressive performance of the Istanbul platform with a large single Oracle and Sybase database instance, multiple instances on bare metal and multiple database instances running in KVM guests. Data showing CPU / Memory bandwidth of the Istanbul will also be presented. Sarathy will demonstrate the efficiency and different deployment scenarios of the platform.

Red Hat Enterprise Linux as Furniture Backend

Claes Adamsson — Product Responsible/Solution Architect, IKEA IT AB

IKEA's, an international home products retailer with nearly 300 stores in 30 countries, IT department has gone from regular CD installs to fully standardize and automated network installations. In this presentation Claes Adamsson, a solution architect at IKEA, will provide an overview of IKEA's IT environment, and explain how they use their key IT components, which include: Red Hat Satellite, Red Hat Enterprise Linux, Red Hat Directory Server, and PXELINUX.

Understanding the Performance of Industry Standard Benchmarks on Red Hat Enterprise Linux

D. John Shakshober (Shak) — Solutions Architect, Red Hat

Vijay Trehan — Solutions Architect, Red Hat

Industry-standard benchmarks (e.g., TPC, SPEC, SAP, etc.) provide an agreed-upon definition of workload and performance metrics that seek to measure hardware, applications, and operating systems. Red Hat Enterprise Linux has received best-in-class results for a number of industry-standard categories, which include TPC-H and SPECweb.

In this presentation John Shakshober will detail the current performance measures of Red Hat Enterprise Linux. Shakshober will discuss industry-standard benchmarks in the database, OLTP, ERP, messaging, and virtualization space. He will analyze the strengths and weaknesses of benchmark results, as well as the mapping of benchmarks. This analysis will provide insight into how various factors, such as the actual customer environment, are determined.

Understanding Computer Performance with SystemTap Scripts

William Cohen — Performance Tools Engineer, Red Hat

SystemTap provides invaluable insight into the operation of computer systems by allowing developers and system administrators to instrument executing code without recompilation.

Various SystemTap scripts, which determine the root causes of performance problems, will be demonstrated at this presentation. William Cohen will demonstrate typical techniques, including associative arrays and filtering data, used in SystemTap scripts. In addition to SystemTap example scripts, Cohen will utilize SystemTap programming idioms to identify performance issues.

It's Not Your Father's NFS

Steve Dickson — Consulting Software Engineer, Red Hat

In this presentation Steve Dickson, a Red Hat consulting software engineer, will describe the upcoming improvements to the NFS file system that will be in Red Hat Enterprise Linux 6. Those improvements include changing the default protocol version to use NFSv4 to improve performance and scalability. Dickson will address the availability of NFSv4 minor version 1 (NFSv4.1), which is an enabler to the leading-edge Parallel NFS (pNFS) technology that is used in cluster environments. He will also discuss debugging techniques (including trace points and system tap scripts).

SELinux for Mere Mortals

Thomas Cameron — Enterprise Solutions Architect, Red Hat

While Security Enhanced Linux (SELinux) is an incredibly powerful tool for securing Linux servers, it has historically had a reputation for being difficult to configure, so many system administrators would simply turn it off. Fortunately the incredible amount of work done by the SELinux community in recent years has made SELinux much more system administrator-friendly.

Attendees of this presentation will learn the basics of SELinux, which include configuring, analyzing, and correcting SELinux errors, as well writing basic policies to enable non-SELinux aware applications to work on SELinux protected systems. Real world examples will be used to better demonstrate how to use SELinux.

Birds of a Feather: Red Hat Enterprise Linux 6 Deployments

Siddharth Nagar — Product Manager, Red Hat

Subhendu Ghosh — Product Manager, Red Hat

Red Hat Enterprise Linux 6 will be the most comprehensive technology platform to date. The purpose of this session is to discuss how best to structure the product composition to address customer use cases.

Red Hat Satellite Power User Tips and Tricks

Thomas Cameron — Enterprise Solutions Architect, Red Hat

Dave Russell — UKI - Solutions Architect, Red Hat

Red Hat Satellite Server is a powerful, flexible systems management solution for deploying, managing, monitoring, and redeploying Red Hat Enterprise Linux servers. Red Hat Satellite allows users to decrease administrative costs and increase time available for strategic tasks by performing single operations through the web user interface. These single operations can affect some or all of their servers.

This presentation, given by Red Hat's Thomas Cameron, explores the most useful Red Hat Satellite tips and tricks used by the Red Hat Systems Management subject matter expert team. Presentation attendees will learn how to use Red Hat Satellite in advanced, complex, and/or large-scale environments.

This presentation, which will be presented as a series of live demonstrations, is comprised of advanced usage topics, which include:

• Setting up a DHCP/PXE/tftp environment to enable automated installations
• Setting up kickstart profiles to perform automated installations
  • Example of advanced kickstart options like partitioning and activation keys
• Setting up configuration file management
  • Example of advanced use cases including configuration file macros
• Channel cloning
  • Includes creating channels "locked" at a certain update
• Custom channels for third party or internally developed content
• Using activation keys for highly flexible group assignment, configuration file deployment, and pre-installation and post-installation software provisioning
• Non-destructive re-provisioning of existing systems to new tasks using package profiles (e.g., converting a web server to a database server without having to reinstall the system)
• Subscription management
  • Includes provisioning new servers and decommissioning “end of life” servers
• Updating certificates for additional subscriptions and renewals

Upon completion of this presentation, attendees should understand the uses for all of these technologies and be able to perform all of these tasks.

Optimize Storage Performance with Red Hat Enterprise Linux

Mike Snitzer — Senior Software Engineer, Red Hat

Sanjay Rao — Senior Software Engineer, Red Hat

In this presentation Mike Snitzer, a senior software engineer at Red Hat, will describe how to tune the Linux storage subsystem to obtain the best performance based on a particular application and storage configuration. Snitzer will also detail how to configure and manage LVM for optimal performance. Presentation topics include:

• Capabilities and features
• Optimal I/O alignment
• I/O chunk size
• Read-ahead settings
• Multipath advantages

Unmatched Security is Manageable

Spencer Shimko — Senior Security Engineer, Tresys Technology

Security management, whether for one system or for thousands, is perhaps the most important aspect of security on Linux. Red Hat Enterprise Linux has an unmatched set of core security features, which include: netfilter, exec-shield, the audit subsystem, SELinux, and access-control lists. If an organization combines the breadth of security applications with those features, it has the ingredients to build solutions with unmatched security. Too often, however, organizations fail to use these features correctly and/or to their full potential because of the difficulty and cost of managing them.

This presentation will detail how organizations benefit from the implementation of open source system management tools to configure, monitor, and update the security configuration of Linux systems. The suggested techniques are based on real-world experience building secure solutions and appliances targeted at the most demanding environments. These techniques produce secure systems that can be easily configured and audited to comply with the most stringent security requirements and standards.

The Turtle and the Hare – A Tale of Two Kernels

Rik van Riel — Senior Software Engineer, Red Hat

The understanding that development of the Red Hat Enterprise Linux kernel is slow moving, and that upstream kernel development is fast, raises many questions, such as:

• What do “slow” and “fast” really mean? How slow is slow? How fast is fast?
• Is backporting less risky than following upstream?
• What does it all mean to me?

Rik van Riel, a senior software engineer at Red Hat, will discuss what happened in the upstream kernel since RHEL 5 was released. He will provide an overview of how much was backported to RHEL 5 and what features will only be available in RHEL 6. He will also detail the tradeoff between stability and features in Red Hat Enterprise Linux kernel updates, and provide insight into the backport decision-making process.

Beginner's Guide to Running Oracle on Red Hat Enterprise Linux

Brian Likosar — Solutions Architect, Red Hat

This presentation will cover everything a systems administrator should know in order to prepare for an Oracle installation on Red Hat Enterprise Linux. Brian Likosar, a Red Hat solutions architect, will conduct a demonstration of the main components of an Oracle installation and lead a discussion about further considerations for growth of the environment. In addition Likosar will:

• Offer advice for finding relevant information before beginning an installation
• Detail how to check/ensure that prerequisites for installation are met
• Discuss the “cultural issues” between a system administrator and database administrator
• Highlight some considerations for automating the installation
• Discuss Oracle's silent installation feature
• Compare the setup of raw devices, Automatic Storage Management (ASM), cooked filesystems, Oracle Cluster File System (OCFS), and Red Hat Global File System (GFS)
• Present working startup and shutdown scripts

User-Driven Grid Provisioning and Automated Power Management

Erich Morisse — Senior Solutions Architect

With an architecture based on the Condor workload scheduler from Red Hat Enterprise MRG and the systems management capabilities of Red Hat Satellite, organizations can quickly build a mechanism that can deploy systems without system administrator intervention. In addition organizations can track system usage for internal accounting and charge-backs and auto regulate power consumption in the grid.

In this presentation Erich Morisse, a Red Hat senior solutions architect, will demonstrate an architecture that allows end users to deploy their own systems, maintain charge-back information, and control power consumption in the grid. The presentation will address:

• The need for automation to minimize operating costs
• Architecture components
• Demand-based systems deployment
• Power state management within the grid

Themes including: automation, deployment, systems management, power consumption, and cost reduction will be addressed and Red Hat Enterprise MRG, Red Hat Satellite, and Red Hat Cluster Suite will likely be represented.

Simplifying New Server Deployment

Matt Domsch — Technology Strategist, Office of the CTO, Dell

Dell and Red Hat have worked together for over a decade to provide open source device drivers that are built into each OS version and work on most Dell PowerEdge Servers and Precision Workstations. Because there is a need for new or updated device drivers for the customer's existing operating systems, deployment of new-to-market servers is a challenge.

Dell and Red Hat are simplifying new server deployment using Red Hat Enterprise Linux's device driver update model and Dell's built-in systems management features. This collaborative solution ensures the right drivers are present for the customer's new server, when the customer needs it.

Running Red Hat Products at Red Hat - A Look into Red Hat Corporate IT Infrastructure

Nick Otto — Director, IT Technology Infrastructure, Red Hat

Red Hat has an approach to run mainly Red Hat and free and open source (F/OSS) products. As a result, the maintenance and support costs incurred at Red Hat are very attractive compared to infrastructures built on traditional providers. While some challenges are more difficult when accomplished without proprietary tools, the Red Hat infrastructure proves that you can run an IT Infrastructure on mostly F/OSS tools and meet/exceed the uptime requirements demanded by enterprise businesses.

This presentation will provide a glimpse into Red Hat's corporate IT Infrastructure with a discussion about the products, tools, and processes used, and the realities of running Red Hat products in the enterprise. It will cover the general benefits of running a mostly F/OSS infrastructure and include the financial and risk factors that should be considered.

Key elements that will be covered include:

• Infrastructure overview
• Virtualization strategy and implementation
• Server and storage standards
• Data replication strategy
• Application/Middleware strategy
• Configuration/build management, monitoring, administration
• Collaboration (e-mail, calendar) overview

Carving out Cost with the Mainframe

Andrew Hardy — Senior Solutions Architect, Red Hat Asia Pacific

Red Hat Enterprise Linux support for mainframe systems allows customers to deploy a common operating system environment across their entire IT infrastructure. This presentation will provide an overview of Red Hat Enterprise Linux on mainframe systems and describe how to carve out costs with one of the most expensive pieces of IT equipment available.

Specifically, this presentation touches on the business drivers and technical architecture of Red Hat Enterprise Linux on the mainframe, highlighting how a value proposition is formed with the use of Red Hat Satellite and JBoss. Red Hat customer success stories from within the least-impacted financial sector in the western world (Australia/New Zealand) will be highlighted.

How to Achieve TCO Savings of up to 89% Using a Red Hat, JBoss, and Alfresco Stack Instead of SharePoint

Ian Howells — CMO, Alfresco

With restricted IT budgets, businesses are under pressure to deliver more results with less resources. Open source allows companies to lower cost and total cost of ownership (TCO) by reusing existing hardware, software, and skills with no lock-in.

This presentation will provide a comparative TCO analysis of a Red Hat, JBoss, and Alfresco stack versus proprietary options like Microsoft SharePoint. Using publicly verifiable pricing to prove the true cost savings available with an open source stack, Dr. Ian Howells, CMO of Alfresco, will discuss how executives can fund IT projects with operating expenses as opposed to capital expenses.

Creating a Low Cost Red Hat Enterprise Linux Deployment

Andrew Hecox — Technical Account Manager, Red Hat

Red Hat Enterprise Linux provides a vast range of diagnostic utilities and features. This presentation will provide advice for taking advantage of its tools in order to create an easy to support and easy to manage Red Hat Enterprise Linux deployment.

Andrew Hecox, a technical account manager at Red Hat, will provide information to help attendees get started using:

• Debuginfo and GDB for user-space core analysis
• Kdump and Crash for kernel core analysis
• Strace and ltrace for user-space system and library tracing
• SystemTap for next generation kernel and user-space tracing
• Rsyslog, syslog, and sysstat for long-term data collection
• SNMP and IPMI for event monitoring
• Vmstat and iostat for interactive performance observation
• Systool and lspci for hardware inspection

In addition Hecox will provide a brief overview of subsystem specific tools like blktrace, debugfs, ethtool, netstat, schedstat, and slabtop. He will also highlight Red Hat support and entitlement options, such as the TAM service and EUS. While Hecox will provide use cases for each tool, he will present only the minimal amount of work necessary to provide a comprehensive support and management toolset.

How to Unlock and Free the Desktop and your Data Center with a Next Generation Email and Collaboration Platform

Andy Pflaum — VP of Business Development and Global Channels, Zimbra

Bruce Maas — CIO, University of Wisconsin-Milwaukee

Zimbra is an advanced email, calendaring, and collaboration suite that allows integration with other business applications and enables organizations to save money and avoid the lock-in of proprietary software such as Microsoft Exchange and Windows. Zimbra, a charter member of Red Hat Exchange and part of the Open Source Alliance, and Red Hat provide ongoing product updates, and support for this leading open source collaboration suite.

In February of this year, the Wall Street Journal covered the University of Wisconsin-Milwaukee's (UWM) successful deployment of Zimbra in a story titled "Business Solutions: Smart Ways to Cut Costs." Bruce Maas, CIO of UWM, will join Andy Pflaum, Zimbra's vice president of business development and global channels, in a discussion about Zimbra's cost-reducing benefits.

Maas and Pflaum will illustrate the cost reduction achieved on campus through the leveraging of Red Hat Enterprise Linux and Red Hat Cluster Suite. They will highlight how enterprises, educational institutions, and service providers can save up to 40% on collaboration costs for thousands of users.

Birds of a Feather: Open Source for America Campaign - Answering the Call for Open Government

Paul Smith — General Manager & Vice President – Government Sales Operations, Red Hat

Tom Rabon — EVP, Corporate Affairs, Red Hat

President Obama issued an executive order calling for increased openness, participation, transparency and collaboration in the government. To answer this call for a more open and transparent government, Red Hat, along with over 60 other companies, organizations, think tanks and academic institutions launched the Open Source for America Campaign www.opensourceforamerica.org. The primary mission of the campaign is to educate decision makers in the US Federal Government about the advantages of using free and open source software and to encourage the federal agencies to give equal priority to procuring open source software in all of their procurement decisions. Come listen and contribute to a conversation with Tom Rabon, EVP Corporate Affairs, and Paul Smith, VP & GM for Public Sector Operations, to hear how you can take advantage of the opportunity to rally members of the open source community to help the US government meet it goals of openness, participation, transparency and collaboration.

Cost Effective Business Intelligence with Red Hat and Open Source

Sherman Wood — Director of Business Intelligence, Jaspersoft

Business Intelligence (reporting and data analysis) has been at the top of IT's to do list for over 10 years, but has remained expensive to purchase and deliver. This session will show configurations of Red Hat, Jaspersoft's commercial open source BI suite and other open source technologies to provide business intelligence services for organizations and individual applications at a fraction of the cost of proprietary options.

Discount Tire Accelerates E-Commerce with Red Hat Satellite

Will Darton — Senior Server Administrator, Discount Tire

Discount Tire Company, the world's largest independent tire and wheel retailer, needed to support its rapidly growing e-commerce business while keeping costs contained and complying with strict PCI security standards. The firm ran its entire e-commerce operation on Red Hat Enterprise Linux, as well as numerous internal applications including its Lotus Domino server application. They needed a solution that would fully utilize the company's IT resources.

Will Darton, Discount Tire's senior sever administrator, will discuss the implementation of Red Hat Satellite, an easy-to-use systems management platform for Linux infrastructures. Darton will also discuss how Red Hat Satellite has provided an easily managed, secure, and utterly reliable platform for Discount Tire’s e-commerce business, while enabling compliance with PCI security standards, increasing IT staff productivity, and containing costs due to the ability to easily provision, maintain, and manage servers.

Cisco's Unified Computing Management and the Open XML-API

Roger Andersson — Manager, Technical Marketing, Cisco

The Cisco Unified Computing system is based on a full-featured open XML-API. This API enables developers to create unlimited use cases including inventory, monitoring, and configuration management that spans across compute, network, and storage resources in the UCS system.

In this session, Roger Andersson, Manager, Technical Marketing at Cisco, will discuss integration using the open XML-API and how they are applied to today's challenges in infrastructure management by focusing on the values of managing the infrastructure through policies and rules. The session will feature information about the open XML-API, a live demonstration of the UCS Manager, and samples of how to leverage the open XML-API.

Doing More with Less: The Economics of Open Source Database Adoption

Ed Boyajian — CEO, EnterpriseDB

Economics drive open source adoption. Thus, given the current state of the economy and the need for businesses to do more with less, many enterprises are considering a migration to open source. Because many IT staff have experience with open source operating systems and middleware and are familiar with best practices for migrating to open source, databases will be the next wave of open source adoption in the enterprise. (In fact a recent Forrester reported that the future of open source databases remains bright, and that every enterprise should consider open source databases as part of an overall DBMS strategy to deliver cost savings.)

In this presentation EnterpriseDB CEO Ed Boyajian will discuss the economics behind the adoption of open source. He will also discuss why and how databases will be the next wave of open source adoption in the enterprise.

CME Group Migration to Linux on Intel® Architecture: Challenges and Results

Vinod Kutty — Associate Director, CME Group

The proven value of open source software running on industry standard servers is driving adoption deep into the enterprise data center, enabling companies to replace aging UNIX*/RISC deployments with more powerful, flexible, and affordable solutions.

Vinod will describe CME's business environment; the migration requirements and challenges they faced; and the resulting TCO benefits of their transition on Linux on Intel® Architecture.

Desktop Virtualization: Achieving Real TCO Savings

Mike Ferris — Director of Product Strategy, Red Hat

Server-based, virtual desktop solutions are coming of age thanks in large part to the capability and adoption of server-based virtualization technologies. Many organizations are investigating deploying virtualized desktops with a specific focus on total cost of ownership as the primary criteria in the decision. However, the value of these deployments cannot always be captured in easily quantifiable cost savings. This session will explore both the hard and soft costs and advantages that a virtual desktop deployment can bring to a medium or large enterprise.

SAP Operating System and Data Base Migrations to Red Hat Enterprise Linux

Jan Beyen — Delivery Manager, REALTECH, Inc.

The first part of this presentation will discuss the trends from proprietary UNIX systems towards INTEL x64 and Linux systems. The second part of this presentation will discuss why customers decide to migrate to Linux and how SAP OS/DB migrations are performed.

Open Source Virtualization

Chris Wright — Senior Software Engineer, Red Hat

Andy Cathrow — Product Marketing Manager, Red Hat

In this session Red Hat will discuss the evolution of open source virtualization and some of the latest innovations in open source virtualization. Chris Wright and Andy Cathrow will cover the Red Hat Virtualization roadmap, introducing the new virtualization features in Red Hat Enterprise Linux 5.4 and preview some of the upcoming technologies.

The Great Migration

Bdale Garbee — Chief Technologist, Open Source & Linux, HP

The exodus of customers leaving legacy infrastructures behind to adopt Linux on industry standard servers with scalable low-cost virtualization is easy to see. Less obvious are the motivations of this migration. This session examines this theme from the vantage point of a long-standing contributor to open source communities, industry observer, and HP technology leader.

The Rising Open Source Ecosystem

Matt Asay — VP Business Development, Alfresco

Open source continues to grow, saving enterprises money and driving innovation, but it is not growing under the control and guidance of a single provider – in the way that classic proprietary software vendors operate. This session will explore how enterprises and Red Hat contribute to the health and governance of the open source ecosystem by shepherding resources around successful open source projects.

Delivering Innovation and Value through Customer-Centric Development

Tim Burke — Vice President, Linux Engineering, Red Hat

Red Hat's engineering development model is focused on delivering sustained, ongoing value to its customers. This value consists of balancing today's needs of performance and stability with tomorrow's needs of technical feature innovation. Putting customer needs first has dramatically influenced how Red Hat interacts with the broader open source development community, as well as how Red Hat internally structures its product releases.

This presentation will describe Red Hat's development model and explain how this powerful engine drives technical innovation in balance with customer centric enterprise value. Attendees will learn about the customer driven value methodology in Red Hat's development model and why it yields the most robust enterprise Linux offering.

IBM Dynamic Infrastructure®: Virtualize, Automate and Optimize IT

Connie Blauwkamp — Software Strategy, IBM Systems and Technology Group

Michael Day — Distinguished Engineer and Architect, Open Virtualization, IBM

As organizations plan for future infrastructure requirements and ways to save energy, they increasingly look to virtualization as one of the cornerstone technologies. Consolidation and virtualization are key initiatives for reducing costs in the data center. Virtualized, automated systems not only reduce costs, they also provide the server, storage and application flexibility required to improve service in a dynamic infrastructure. Over 40 years, IBM has developed unique capabilities – and the skills and experience that combine to provide the best solutions for our clients. As a primary maintainer of KVM, IBM is bringing a history of innovation in virtualization technology and implementation to the open source community.

This presentation will cover IBM's strategy for helping clients build a Dynamic Infrastructure, and will highlight the strategic role of KVM in IBM's Dynamic Infrastructure vision. Come and learn what IBM is bringing to KVM and why we think KVM can bring value to businesses through Dynamic Infrastructure.

The Red Hat-Fedora-OSS Development Model

Paul Frields — Fedora Project Leader, Red Hat

Red Hat's commitment to open source is about more than just the products, solutions, and services it delivers. It's also about the ways that the company engages community to come together and solve common problems. Red Hat does this in a number of ways, including participating in individual software projects called the upstream, and also by sponsoring and contributing to the Fedora Project. These engagements are central to Red Hat's mission, and key to the trust relationship it has with its customers and partners.

Fedora Project leader Paul W. Frields will provide an overview of the development and release models of Fedora and Red Hat Enterprise Linux, and the win-win benefits of this process to customers and community alike. Frields will include examples of how contributions have influenced new Red Hat Enterprise Linux features. He will also make the case that focusing on contribution produces a sustainable model for growth and innovation in open source technologies.

Birds of a Feather: Compliance Best Practices... Leveraging Red Hat Management Solutions for Success

Akash Chandrashekar — Red Hat Solutions Architect

Security, compliance, and auditing requirements can be daunting and frustrating for many organizations.  This discussion will explore the compliance requirements faced by companies today and the utilization of Best Practices leveraging Red Hat Management Solutions to comply with PCI and other compliance requirements.  Come join both Red Hat experts and fellow Red Hat customers to hear how they have overcome the compliance challenges in their organizations.

The Sky is the Limit. Building Your Own Cloud Infrastructure Made Easy

Jan Mark Holzer — Senior Consulting Engineer, Red Hat

Hugh Brock — Engineering Manager (oVirt Development), Red Hat

Virtualization has the promise of making IT environments more adaptive to change and business requirements. As virtualization is becoming mainstream technology, many IT departments and architects are expanding their interest into cloud computing and focusing on how it might influence their data center architecture.

With Red Hat virtualization technology, customers are able to manage their data centers in new, advanced ways. Thus new opportunities for deployment, consolidation, and provisioning that go beyond the traditional IT processes, are being implemented.

As cloud computing continues to demand attention, Red Hat is actively developing open source cloud infrastructure software as part of the oVirt project. The project's goal is to provide a cloud infrastructure software stack that allows a customer or partner to easily create and implement their own internal or external cloud deployment. In addition to the infrastructure software, this project aims to provide a “Cloud Starter Kit,” blueprint, and architecture documents.

This presentation will cover the current state of Red Hat's cloud efforts and provide technical information that details how to build a cloud infrastructure based on Red Hat's technologies and blueprints. The presentation will also provide an overview of the capabilities and features Red Hat Enterprise Linux virtualization provides and how virtualization can be used as an enterprise management platform.

Red Hat Security Advisory Secrets Revealed!

Josh Bressers — Senior Engineering, Red Hat

Applying security updates in a timely and secure manner is a necessity at many organizations. Red Hat provides such updates, which keep systems as secure as possible, via Red Hat Security Advisories (RHSA). Every RHSA goes through a number of steps, from its inception to release, and contains a wealth of information for administrators to consume.

This presentation will explore:

• How bugs are determined to be security issues
• How bugs are organized in preparation for a security update
• When and how a security update is formed
• How to best understand the information contained within a security advisory
• What a Red Hat Security Advisory is, what it contains, how it was created, and why you should care

Green IT: A Red Hat Perspective

Chris Runge — Director of Solutions Architects, Red Hat

While green IT is pursued for a number of reasons (eco-responsibility, CAPEX/OPEX, facility constraints, etc.), a green IT strategy often has unforeseen benefits (increased manageability, security, business continuity, etc.).

This presentation will explore how Red Hat has and continues to contribute to the green IT movement. While there are many ways of going green, Chris Runge, a director of solutions architects at Red Hat, will discuss four of them:

1. UNIX-to-Linux migration: UNIX-to-Linux migrations facilitated green computing before the term “going green” was popular. As customers migrated from RISC systems to industry-standard systems running Linux, they were able to achieve higher performance with less hardware and lower power consumption.
2. Power management technologies: Runge will discuss Red Hat's involvement with power management technology, including LessWatts.org. Emerging improvements, which include: RHEL 5.3 and Nehalem, the tickless kernel, improved drivers, and improved coding techniques, will also be discussed.
3. Virtualization: Virtualization provides many benefits to a green IT strategy, from server consolidation to the increased benefits of Red Hat Enterprise virtualization's system scheduler and power-saver technologies. In addition to these benefits, Runge will discuss the benefits of emerging approaches, which include KVM's architectural approach (relative to power savings), cloud computing, and VDI.
4. The mainframe: Runge will highlight a migration to the mainframe as a path to green IT with a discussion of the Bank of New Zealand migration.

The Elephant in the Room: Free Software and Microsoft

Jeremy Allison — Co-creator of Samba

This session will cover Microsoft's inconsistent course on patents and FOSS

Recent Developments in the Law of Software Patents

Rob Tiller — Vice President and Assistant General Counsel, Red Hat

Steve Gardner —Co-Chair, Patent Litigation Group, Kilpatrick Stockton, LLP, Red Hat

Many open source developers and technology companies are concerned that software patents inhibit innovation and increase business risks. This session will help developers and business leaders understand the baseline issues of software patenting and will highlight important recent developments that affect the open source software community.

Topics will include the changing landscape for software patents, the Bilski case now pending before the Supreme Court, open source community defense measures, reform efforts, and recent case law developments.

Safeguarding your Data with Desktop Virtualization

Jim Brennan — Senior Product Marketing Manager, Red Hat

John Pirc — Product Line Executive at IBM Internet Security Systems and Security Thought Leader at The SANS Institute

The increased mobility of the modern workforce enables unprecedented levels of business agility and flexibility. This new found agility, however, introduces new challenges that threaten the very existence of today's enterprises. Insecure endpoints or stolen laptops can lead to the loss or compromise of both confidential and personal information. This, in turn, can result in lost competitive advantages as well as class-action law suits.

This session focuses on the use of desktop virtualization technology as a way to centralize sensitive information in your data center, as opposed to having it reside on your organization's laptops and desktops. This dramatically increases the security of this sensitive information, and allows you to meet many of today's demanding compliance regulations. Specifically, we will examine how Red Hat Enterprise Virtualization Manager for Desktops will provide this security while at the same time delivering an end user experience that is indistinguishable from that of a physical desktop.

The Challenge of Patent Trolls -- How to Contribute Prior Art

Adam Avrunin — Chief Patent Counsel, Red Hat

John Alemanni — Partner, Kilpatrick Stockton

Keith Bergelt — CEO, Open Invention Network

Although there have been few patent lawsuits against open source software, many in the open source community have concerns about the assertion of overly broad and obvious patents by patent trolls (entities that obtain and enforce patents while having no real business of their own) and others. Fortunately, the patent laws provide mechanisms for challenging and invalidating a patent when it covers technology that was not new and non-obvious over work that was previously done by others - the so-called "prior art." This session will provide interested persons, including technical managers and engineers, with the background that they need to be able to assist in the process of challenging such patents. Specifically, this session will explain how to read a patent and how to identify prior art useful in challenging a particular patent's validity. This session will also discuss the Linux Defender program, which has invited the public to put this information to use by providing a forum in which people can submit prior art and discuss its relevance to challenging the validity of patents of interest to the open source community.

Kickstart Tips and Tricks

Chip Shabazian — VP of Systems Engineering, Bank of America

Chip Shabazian, Bank of America's vice president of systems engineering, will detail the tips and tricks that his team learned while developing a global Kickstart infrastructure for thousands of distributed builds. Shabazian will focus primarily on using Kickstart, an automated method to build servers that utilize the Anaconda tools for system builds. This discussion of lessons learned will be valuable for anyone Kickstarting any deployment—from a large global deployment to a single server deployment.

Shabazian will also provide information about other technologies in development.

Managing Infrastructure as a Development Project

Michael Stahnke — UNIX/Linux Team, Caterpillar

Traditional system administrators are often stereotyped as anti-social hardware managers who execute repetitive tasks to maintain an availability of applications from which they are completely disconnected. This presentation seeks to break this stereotype by offering advice for adjusting the thinking and methodologies of system administrators and IT teams.

Michael Stahnke of Caterpillar's UNIX/Linux team will cover how applied software development practices in conjunction with system administration practices can be used to construct a high-performing team. He will address the following concepts:

• Using unit tests to prove out configuration changes
• Writing formal specification documents based on requirements
• Deploying a continuous integration system to measure defect rates on the production environment
• Managing bugs, enhancements, and releases in an infrastructure

Data Center on a Stick: The Reality of Data Center as a Service

Christopher Dale — Moderator, inCommon

Henry Robertson — President, Open Source Technologies, Inc.

This presentation serves as a follow-up to last year's "Implementing Grid-Enabled Services Infrastructure (GESI) with Red Hat, JBoss, and MetaMatrix" presentation. This year's topic takes a step forward into the realm of data center as a service. It will focus on deploying open source management technologies like Cobbler and Koan, Spacewalk and RHN Satellite Server in combination with JBoss middleware technologies.

Christopher Dale, a moderator at inCommon, will discuss how to effectively deploy a services-based data center from a central repository of predefined foundation components. With selected use cases, Dale will demonstrate the core technologies that enable the next generation grid-backed, service-centric data center.

SoftwareForge: Collaborative Development and Reuse of Open Source and DoD Community Source Software

Aaron Lippold — Project and Technical Lead, Open Source Solutions Group, Defense Information Systems Agency

Guy Martin — Forge.mil Community Manager, CollabNet, Inc.

The Defense Information Systems Agency established the Forge.mil project to enable the rapid development, testing, certification, deployment, operation, and acceptance of new products and services on the Global Information Grid (GIG). Breaking with traditional systems development approaches, Forge.mil will employ the principles of net-centricity (e.g., collaboration, information sharing, shared situational awareness, self-synchronization) to the Department of Defense (DoD) IT acquisition arena.

SoftwareForge, the collaborative development and reuse of open source and DoD community source software, promotes early and continuous collaboration among developers, material providers, testers, operators, and users throughout the development life-cycle. SoftwareForge provides the development community with application life-cycle management and collaboration tools.

From Tru64 Unix to Linux at Hilti Corp

Michael Hagmann — Head of Enterprise Server Technology, Hilti Corporation

In 2006 Red Hat Enterprise Linux was selected to become the Hilti Corporation's strategic operating system. This decision was reached mainly because of Red Hat's strong focus on enterprise systems, the matured Red Hat Global File System (GFS), and Red Hat's high-availability cluster framework.

Michael Hagmann, head of enterprise server technology at Hilti Corporation, will discuss the company's migration from HP Tru64 Unix on Alpha Servers to Red Hat Enterprise Linux 4 on standard servers. Hagmann will discuss the company's migration process from the decision-making to the successful implementation of Red Hat Enterprise Linux. Throughout this discussion, Hagmann will highlight the technology and strategic challenges that his team faced, and the steps that were taken to overcome those challenges.

Bird's of a Feather: Applying Open Source Principles to the Federal Government

Gunnar Hellekson — Lead Architect, Red Hat

The current administration talks the talk in terms of its adoption of new technology solutions, access to information, and the call for transparency and increased citizen participation. But can it walk the walk? Gunnar Hellekson, a lead architect for Red Hat Government, will lead a discussion that addresses how open source advocates can help the Federal Government unlock the innovative potential of the open source development model. The session will also highlight new initiatives to bring government and community together to create a technology infrastructure that will increase the government’s responsiveness to its citizens.

Get Connected – Using Open Source Technologies on Facebook

Deb Woods, — VP of Product Management, Ingres

Binesh Gummadi, — Senior Software Engineer, Rivet Logic

Rivet Logic, a consulting and systems integration firm that helps organizations better engage with customers, recently delivered a next generation application on Facebook using:

• JBoss for the application server workload
• Alfresco for content management
• Ingres Database for the repository
• Red Hat Enterprise Linux as the operating system

This presentation will detail the delivery of this cost-effective solution, which was quick to integrate, easy to use, and robust enough to handle a 24x7 application.

Creating a Virtualized, Secure Blade Cluster Using Red Hat Enterprise Linux 5

Lynn Riggs — Senior Economist, US Bureau of the Census

The Census Bureau manages a network of nine Research Data Centers (RDC) around the United States. The Research Data Centers are secure Census Bureau facilities where users on approved projects access specific, project-related information. Therefore user and group controls are critical to the operation.

The previous RDC Network computing model had essentially one server per RDC, with additional servers for specific projects and Census Bureau staff. There were two main problems with this method: 1) Census Bureau staff needed to manage users and applications across all of these different servers; and 2) some RDCs are busier than others and have different use patterns, especially across different time zones. If one RDC, and hence the RDC's server, was busy, there was no way to move users to less busy servers. Therefore there was much redundancy under this model.

Clustering six IBM blade servers using Red Hat Enterprise Linux 5 with LDAP solves these problems by allowing users to move to underutilized nodes of the cluster trivially and by greatly simplifying management of the RDC Network. Census Bureau staff no longer have to manage 15 different accounts (just for the RDC Network) or log into the correct server to fix a user problem. The blade cluster allows for incremental expansion of the computing power of the RDC network to accommodate an increased (but unknown) number of users with a minimal amount of effort on the part of Census Bureau staff.

In this presentation Lynn Riggs, senior economist at the US Bureau of the Census, will discuss both the benefits (e.g., management, performance, ease of transition) and the challenges (e.g., removing residual processes and files, correctly transferring user settings and permissions) of clustering IBM blade servers using Red Hat Enterprise Linux 5. Riggs will also highlight the Bureau's plans to potentially open five new RDCs over the next two years.

RPM-ifying Third Party Software

Paul Waterman, — Linux Technical Lead, Motorola

Many system administrators sigh in frustration when the installation instructions for third party software alerts them to execute install scripts. If they do not use RPM packages to install software, they lose many of the benefits that RPMs provide. For example there is no easy, one-stop place to find everything that is installed on the system, there is no ability to quickly and easily see what has been changed on the system, software interdependency issues become more likely, and it is often more difficult to automate software deployment, especially during kickstarts. Therefore, it takes more system administrator time and effort to install and manage unpackaged software.

Presentation attendees will learn how to eliminate all of these problems and more by turning third party software installs, even binary-only installs, into RPM packages. The presenter, Motorola's Linux technical lead Paul Waterman, will provide real-world examples from his experience turning IBM Rational ClearCase, a complicated product with kernel dependencies, into RPM packages.

Specifically, Waterman will:

• Provide a high-level overview of the steps involved:
  • Prepping a system
  • Identify what is on the system
  • Installing the software
  • Identifying what changed
  • Creating a package
• Provide an example of how attendees can use these steps to create a simple, single package with very strict dependencies
• Highlight ways that attendees can streamline the process using advanced concepts, such as logical volume snapshots and virtual machines
• Explore how attendees can make RPM packages more flexible by splitting them into application and configuration packages
• Examine how attendees can minimize specific dependencies by using pre-install and post-install scriptlets to handle different system conditions on install
• Briefly examine how attendees can go beyond what the software vendor provides by using trigger scriptlets to allow software to continue to work without any operator intervention when dependencies on the system change

Solving the Threat of "Dirty Devices"

Spencer Shimko, — Senior Security Engineer, Tresys Technology

Defense and intelligence communities face a constant barrage of inventive attacks against computer and communications equipment and networks. Given the pace and diversity of these attacks, it is critical that companies leverage agile, open technologies to rapidly build and deploy the latest generation of security solutions. Similarly, defense and intelligence agencies deal with unique security challenges - in mission-driven and fast-paced environments - where it is imperative to have advanced systems in place for information assurance and data integrity. By leveraging the right technologies, often customized to meet varying agency needs, organizations can solve the most sensitive security missions through agile security.

The combination of Tresys Technology's security expertise and Red Hat Enterprise Linux enabled a solution that was created from initial requirements to working prototype in less than 10 days, and from requirements to field-ready in 90 days. This approach validates the advantages of integrating strong government-off-the-shelf (GOTS) and commercial-off-the-shelf (COTS) security on an open source platform.

In this presentation Tresys Technology's senior security engineer, Spencer Shimko, will detail how the combination of open source tools and industry-leading Red Hat Enterprise Linux enabled Tresys Technology's rapid response to the recent "dirty device" security threat.

Providing an Effective Intranet Knowledge Base using Linux and Open Source Software

Clay Wells, — Senior System Programmer, University of Pennsylvania

Lance Barbour, — Systems Administrator, University of Florida

By applying concepts of a Knowledge Management System (KMS) to the intranet, an effective knowledge base for users can be created. On the other hand, inadequate management of documentation produces an office environment in which users are unable to obtain pertinent information in a timely and effective manner. This presentation will outline the use of Red Hat Enterprise Linux, Apache, MySQL, PHP, and MediaWiki as the foundation for an efficient and cost effective KMS.

Understanding the Cost Savings Associated with Consolidating SAP Environments on IBM Power Systems™

Bob Friske — IBM Power Systems Linux Offering Manager

Many customers have been managing large numbers of distributed Linux® x86 systems running SAP modules. By reducing the total number of server footprints, customers can achieve savings in operational costs. Learn how, through consolidation projects on IBM Power Systems™, a smaller number of servers can support the entire SAP environment, reducing the organizations' operational costs for power/cooling and systems management, while providing more scalability to the SAP applications – and ensuring high levels of availability for end-users.

Automation and Management of the Application Stack

Graham Gear — Senior Solution Architect, Red Hat

As the IT environment moves to a utility computing model, workload automation becomes vital to the success of these models. When workloads increase and more capacity needs to be provisioned, you need tools that can help you deploy, control, monitor and manage your solutions in automated ways.

In this interactive presentation, Red Hat UK's Graham Gear, senior solution architect, will demonstrate how Red Hat Satellite combined with JBoss Operations Network can help meet this need quickly and efficiently. This demonstration will include:

• Automated provisioning management of workloads (virtualized guests) in a Red Hat Enterprise Linux host
• Automated provisioning and deployment of JBoss Middleware containers in those workloads
• Self management capabilities of the solution to ensure up-time and performance SLAs of the workloads