The Xen Hypervisor was built for the Cloud from the outset: when Xen was designed, we anticipated a world, which today is known as cloud computing. Today, Xen powers the largest clouds in production. This talk explores success criteria, architecture, trade-offs and challenges for cloudy hypervisors.
It is intended for users and developers and starts with a brief introduction to Xen and XCP, their architecture, shine some light on common challenges for KVM and Xen, such as the NUMA performance tax and securing the cloud. It will introduce the concept of domain disaggregation as an approach to increase security, robustness and scalability: all important factors for building clouds at scale. The talk will conclude with an update on Xen support in Linux, Xen for ARM servers and other exciting developments in the Xen community and their implications for building open source clouds.
This session discusses different types of virtual machine snapshots with QCOW2 disk images. A snapshot is a view of a virtual machine (its OS and all its applications) at a given point in time, giving an ability to revert to a known sane state in case of a failure. The session also includes a brief demonstration of the different types of snapshots under discussion (Internal snapshots, External snapshots, System Checkpoint, Online/Offline snapshots) and their use-cases.
Audience would include Linux system administrators familiar with KVM virtualization. The take-away would be an overall understanding of different snapshotting capabilities using QCOW2 disk images, and some practical examples which could be applied while managing virtual machines and snapshots. Finally, a glimpse of upcoming developments in this area.
OpenNebula.org is an open-source project developing the industry standard solution for building and managing virtualized data centers and cloud infrastructures. The presentation will describe the unique innovative features provided by OpenNebula and its integration capabilities that allow to build a cloud within any data center environment.
The target audience are integrators and IT administrators interested in deploying a private cloud solution or in the integration of OpenNebula with other projects. The talk will be useful for both people with experience or without prior knowledge of OpenNebula, as it will start by introducing the project and its main features, along with a quick demonstration. By the end of the presentation attendees will have a comprehensive idea of the integration and customization capabilities of OpenNebula.
Adding Linux support for industrial hardware entails a number of problems not usually found in consumer devices. In most cases they are installed in critical environments with limited availability, where fixing bugs, developing new features and foreseeing possible fault conditions is often a hard task. Using the support of industrial devices in the Linux kernel, the presentation will show how to use QEMU to solve these issues, improve the robustness of the device driver and speed up its development.
People interested in ways to write and test software and hardware using virtualization technologies will find this talk useful, in particular hackers working in testing, device drivers and hardware development.
KVM is gaining market awareness and adoption amongst clients as the Open Alternative to costly hypervisor solutions. Virtualization is a foundational component of the Cloud, and decisions made at the hypervisor layer have significant ramifications in terms of cost and flexibility as IT departments journey towards the Cloud. Supported by industry heavyweights such as IBM, HP, and Red Hat, KVM and Open Virtualization is at the heart of delivering flexibility, performance and cost savings that free customers from single-source hypervisor solutions that lock you in.
This session is targeted at IT leaders, LOB and C-level professionals who have overall responsibility for budget, planning and strategies of their IT departments. The required technical knowledge is low, as this presentation will focus on the business benefits of KVM and strategic planning as related to Cloud.
Do you dream of spinning up ten, twenty, or a thousand virtual machines in an instant? Discover and repair bottlenecks without moving a finger? Dodge the loss of an entire storage array with no-one noticing? During this tutorial, we will demonstrate how to leverage FOSS and Linux tools to build a powerful, scalable cloud that easily competes with proprietary solutions.The tutorial is aimed at those interested in building clouds and uses Xen and CloudStack as examples to get started. You will leave with a collection of pre-made tools that you can use right out of the box or modify to your liking. You will also leave with immediately useful knowledge on best practices and common pitfalls covering areas such as security, multi-tenancy and others. Tools discussed include Xen, XCP. Open vSwitch, OpenStack, CloudStack and DevOps tools such as Chef, Puppet and Juju.
Do you dream of spinning up ten, twenty, or a thousand virtual machines in an instant? Discover and repair bottlenecks without moving a finger? Dodge the loss of an entire storage array with no-one noticing? During this tutorial, we will demonstrate how to leverage FOSS and Linux tools to build a powerful, scalable cloud that easily competes with proprietary solutions.The tutorial is aimed at those interested in building clouds and uses Xen and CloudStack as examples to get started. You will leave with a collection of pre-made tools that you can use right out of the box or modify to your liking. You will also leave with immediately useful knowledge on best practices and common pitfalls covering areas such as security, multi-tenancy and others. Tools discussed include Xen, XCP. Open vSwitch, OpenStack, CloudStack and DevOps tools such as Chef, Puppet and Juju.
Every change to OpenStack is automatically tested with a complete installation of OpenStack on a cloud server. This is a fairly intensive process which includes a good deal of installation and configuration changes to the server on which its tested. To ensure that each such test happens in a clean environment and is correct, the OpenStack CI team has developed a process to quickly and reliably provide single-use cloud servers for Jenkins to use in integration tests.
In this presentation, developers and test engineers will see how the OpenStack project uses OpenStack based clouds to perform intensive testing on single-use computing resources, and how Jenkins and JClouds can be configured to provide a similar scalabale testing environment for any project.
The oVirt Project is an open virtualization project providing a feature-rich server and desktop virtualization management platform with advanced capabilities for hosts and guests, including high availability, live migration, storage management, system scheduler, and more.
oVirt provides an integration point for several open source virtualization technologies, including kvm, libvirt, spice and oVirt node. oVirt was launched in November 2011 as a fully open source project, based on assets from Red Hat Enterprise Virtualization Manager platform. The project has an open governance model, and initial board has members from IBM, Canonical, Cisco, Netapp, Red Hat and SUSE.
This session will examine the OpenStack Project, which is focused on developing a cloud computing platform for private and public clouds. The session will include the projectâs history, contribution and licensing models, the formation of the OpenStack Foundation, the changes associated with moving the project to the OpenStack Foundation, and a more general discussion about moving an open source project to a foundation.
Eileen is representing HP on the OpenStack Board of Directors and was a member of the Drafting Committee for the OpenStack Foundation formation and governance documents. She has spoken on various open source topics at conferences in the United States and in Europe and with the European Commission, Members of the European Union Parliament and the European Union National Competition Authorities.
Service providers and service users are rushing to embrace cloud offerings, many of them powered by open source software. Patent holders are not far behind them. I would like to discuss patents and how they apply to cloud computing, and how to address these issues in cloud initiatives. My presentation is important to the Linux ecosystem and this event in particular because many cloud offerings, both public and private, heavily use open source software, especially Linux, in their infrastructure.
The audience includes cloud service providers, open source providers with cloud offerings, and their customers. They can expect a brief and lively overview of patent law and how it applies to cloud computing, followed by an in-depth discussion of specific issues that need to be addressed when negotiating business arrangements involving cloud offerings. No technical expertise is required.
The last years saw a rapid growth of virtualization technologies. Each virtual machine (VM) has one or more virtual interfaces with its unique characteristics such as MAC addresses. Typically, each VM network interface is connected to a virtual bridge on the host. Traffic between several VMs on one host is routed by a virtual bridge and stays within the host. However, some environments prefer to route this host internal traffic to an external switch so that network administrators are able to apply access and security rules between VMs as well. The IEEE 802.1Qbg standard, finalized July 5th 2012, addresses this requirement. It provides a frame relay service between the VMs and the adjacent bridge and defines methods and protocols to register VMs with external switches. This is achieved without modifying any network packet originated from a VM. Network administrators are now able to monitor VM traffic and apply access restrictions and bandwidth limitations. It provides a single point for management, network control and security and reduces complexity. The discussion describes the current state and the future work to be done to support more advanced data center specific hardware, for example SR-IOV cards. Discussions will include reliability features (bonding).
With the cloud hype the distributed file systems became quite hip too. This topic has been seriously picked by commercial Linux vendors too. Hence, distributed file systems like CephFS and/or Glusterfs will come to the majority of data centers and/or companies around the world. Distributed file systems change quite a bit how storage has to be managed and operated. Using CephFS and GlusterFS this talk will describe their key features and will show the commonalities and differences of them from a datacenter operations point of view. Some basic file system and storage knowledge is useful.
Ceph is one of the most interesting new technologies to recently emerge in the Linux storage space. Based on the RADOS object store, the Ceph stack boasts massive scalability and high availability commercial, off-the shelf hardware and free and open source software. Ceph includes a massively distributed filesystem (Ceph FS), a striped, replicated, highly available block device (RBD), S3 and Swift object storage capability through the RESTful RADOS Gateway, and a simple, well-documented native API withlanguage bindings for C, C++ and Python.
This hands-on tutorial will walk you through the initial setup of a Ceph cluster, highlight its most important features and identify current shortcomings, discuss performance considerations, and identify common Ceph failure modes and recovery.
Good Linux sysadmin/devops background recommended for attendees. Distributed storage knowledge is a plus.
Ceph is one of the most interesting new technologies to recently emerge in the Linux storage space. Based on the RADOS object store, the Ceph stack boasts massive scalability and high availability commercial, off-the shelf hardware and free and open source software. Ceph includes a massively distributed filesystem (Ceph FS), a striped, replicated, highly available block device (RBD), S3 and Swift object storage capability through the RESTful RADOS Gateway, and a simple, well-documented native API withlanguage bindings for C, C++ and Python.
This hands-on tutorial will walk you through the initial setup of a Ceph cluster, highlight its most important features and identify current shortcomings, discuss performance considerations, and identify common Ceph failure modes and recovery.
Good Linux sysadmin/devops background recommended for attendees. Distributed storage knowledge is a plus.