April 15, 16, 17, 18
Thanks for attending! The OpenStack Summit is a four-day conference for developers, users, and administrators of OpenStack Cloud Software.
Note: Presentations are still being uploaded. If you do not see the presentation you are looking for, please check back soon.
Mark Collier, OpenStack FoundationThis is the 4 minute intro provided by Mark Collier of the OpenStack Foundation, discussing the impact of OpenStack beyond business to ambitious missions such as the NSA and the University of Victoria as part of the CERN project. Originally presented on April 17th, 2013, at the OpenStack Summit in Portland.
Randall Sobie, University of Victoria
This presentation describes the motivation for using cloud computing in high energy physics research. In particular, the talk focuses on the experiments at the Large Hadron Collider (LHC) located at the CERN LCEaboratory in Geneva which recently found evidence for the Higgs boson. We will highlight the challenge of analyzing the LHC data using computing resources distributed around the world. Randell Sobie shows that clouds can be used in a number of ways in high energy physics and describe our experiences in this rapidly changing field of computing.
Nathanael Burton, NSAWhat does "cloud" mean at NSA and a discussion of how OpenStack fits into the NSA ecosystem. How a small team drove massive process and efficiency change to become one of the NSA's largest hosting platforms. Fostering an environment where creativity and development risk are balanced within the bounds of existing enterprise processes and priorities. Methods for avoiding the "tragedy of the commons".
Brian Aker and Saar Gillai, HP
OpenStack has won support across the IT industry from users, developers, cloud providers, and vendors from around the world. The OpenStack community has delivered incredible innovation, and demand for more innovation, performance, and quality is accelerating. As the only vendor-neutral open source community for cloud, OpenStack is gaining rapid momentum and Hewlett Packard is contributing as one of the vendors enterprise solutions on top of OpenStack today. Join Saar Gillai, SVP and General Manager of HP Converged Cloud, to learn about HP's journey with OpenStack and how OpenStack could evolve to help accelerate cloud innovation.
Mark Shuttleworth, Ubuntu
OpenStack on Ubuntu is now a production cloud infrastructure in telco, service provider and enterprise settings. While the pace of innovation continues to accelerate at the edge, core functions have matured, and we can share insights and operational experience gained from standing up and supporting those platforms. Mark Shuttleworth will discuss OpenStack production challenges and solutions for telco's and service provider customers and chart the course for an expanding network of partners and solution providers who are collaborating around Ubuntu to accelerate cloud services.
Jose Antonio Coarasa Perez, CERNThe CMS and ATLAS online clusters consist of more than 3000 computers each. They have been exclusively used for the data acquisition that led to the Higgs particle discovery, handling 100Gbytes/s data flows and archiving 20Tbytes of data per day. An openstack cloud layer has been deployed on the newest part of the clusters (totalling 1300 hypervisors and more than 13000 cores in CMS alone) as a minimal overlay so as to leave the primary role of the computers untouched while allowing an opportunistic usage of the cluster. This presentation will show how to share resources with a minimal impact on the existing infrastructure.
Florian Otel, HP Cloud ServicesAt the last OpenStack summit we received a message from the Enterprise on the missing features in OpenStack. That message was received loud and clear. While some of the points are still valid today, there is also a clear answer back from OpenSatck to the Enterprise: ""Boldly go where few yet have gone so far...." In this session will be covering the lessons we have learned from positioning OpenStack in the Enterprise space.
Uri Budnik, RightScaleTake full advantage of the cloud value proposition. Emulate and improve upon the existing features we've used in our existing virtualized environments, mainly VMware, to an OpenStack platform using RightScale. See how traditional infrastructure features such as automation, HA and Vmotion are managed at the cloud and application layer.
Mark Muehl, ComcastThis session will feature Mark Muehl, SVP, Product Engineering at Comcast who will present an overview of their OpenStack strategy and deployment plans followed by a Q&A session with OpenStack Foundation Vice-chair Lew Tucker. Comcast Cable is the nation's largest video, high-speed Internet and phone provider to residential customers under the XFINITY brand and also provides these services to businesses. Comcast has invested in technology to build an advanced network that delivers among the fastest broadband speeds, and brings customers personalized video, communications and home management offerings.
Jorge Castro and Mark Mims, CanonicalJuju is a terrific way to get an OpenStack cloud up and running. We've built a tool called Juju that makes deploying services on an OpenStack cloud very simple, and we've got over 110+ services ready to go. This workshop will be a technical overview of how Juju works, and how you can use it to make deploying services in OpenStack simpler for you; we'll also cover how to write your own orchestration scripts (that we call charms) for your self so that you can rapidly deploy your applications in OpenStack.
In this session, end users will be walked through installing and configuring OpenStack Cinder and Quantum. This interactive configuration session will provide detail in setting up:
Attendees should bring a laptop with Vagrant installed (vagrantup.com) as well as either Virtualbox or VMware Fusion.
Infrastructure-as-a-Service, as exemplified by the success of commercial cloud computing service providers, has clearly proven to be the fastest way to develop and deploy elastic web applications. With the introduction of OpenStack, SDN and API’s for programmatic control over both physical and virtual infrastructure, an opportunity emerges for a completely new way to think about “infrastructure as a service” for both tenants and system administrators alike. This talk will focus on how we and others in the community are working with OpenStack and extending the Quantum network service to explore new ways to think about the delivery of network services and infrastructure management. While still in it’s early formative stage, this approach promises to bring new meaning and possibilities for infrastructure-as-a-service limited only by our own imagination.
Das Kamhout & Don Dugger (Intel)
In this session Intel focuses on continued momentum for Trusted Compute Pools in OpenStack, a graphical SWIFT object store benchmarking tool, and enhancements across compute, networking, and storage targeted for future OpenStack releases. The session will conclude with the latest developments in Intel’s own deployment and use of OpenStack for their hybrid cloud.
By the summit, the NeCTAR Research Cloud will have been operating in production for more than a year. A federated Infrastructure-as-a-Service cloud for use in any discipline across the Australian research community has provided virtual machines and object storage to researchers around the country. Within the next eight months, it will expand to 30,000 cores over 8 sites distributed over an area comparable to the USA, using OpenStack Compute Cells.
Using this as a platform, the sector is moving to build an array of services and software solutions to enhance research ability and collaboration. This presentation builds on the existing case study (http://www.openstack.org/user-stories/nectar/collaborative-research/) to introducing the design, and address the challenges encountered in operating a multi-site cloud for researchers from Archaeology to Wine. For the technical: we'll dive into cells, list real hardware specifications, talk about HA, puppet, and more For the manager: ask us about how few people we need to run a cloud of this scale, and how we build up the stack.
While it's still an evolving area, the industry now has a a few years of virtual networking under its belt. And with many production deployments, and standard abstraction layers like Quantum, OpenStack is leading the way. In this talk, I'll draw from my experience of hundreds of customers visited, hundreds of thousands of miles flown, and dozens of deployments to describe use cases, what works, what doesn't, and where things seem to be going.
I will also touch on VMware's progress on vSphere in OpenStack and our plans for VMware NSX, the industry's first unified network virtualization platform out later this year.
On a beefy machine (24 cores, 96 GB RAM, SSD), booting a single instance (from "nova boot"" to ACTIVE) takes seconds. However, when you try booting 20 instances in parallel, the last instance might not be ACTIVE for minutes! While you're waiting, you notice that the host's CPUs and disk are mostly idle and there's plenty of free RAM. While your instances are BUILDING, you wonder what's going on -- why's this taking so long?
Ryan Floyd, Richard Shank, Anders Tjernlund, Scott Sanchez
Networking challenges in data center and cloud environments have received significant attention by industry and standards organizations. The data center environment is dominated by the presence of software networking components (vswitches) in server hypervisors, which may outnumber by an order of magnitude the physical networking nodes. Bridging the gap between server based networking and existing network services is a significant challenge, since the ultimate goal is the design of end-to-end network services. When it comes to advanced L3 services and interoperability with existing managed VPN services, existing solutions rely on static routing and/or centralized routing mechanisms that cannot meet the requirements for resiliency and dynamic networking.
This talk will discuss a simple approach for this problem, which combines traditional control plane and routing protocol approaches with the flexibility of SDN architectures and Openflow. From a deployment perspective, we will discuss how this approach can be easily integrated within the Openstack Quantum framework by requiring minimal modifications. The presentation will conclude with a live demo of the solution.
Devananda van der Veen
OpenStack currently enables the provisioning and operation of applications on cloud VMs, but what if you want to provision onto real hardware instead of VMs? Wouldn't it be nice to use OpenStack for this, too?
Going one step further, what if you could easily use OpenStack itself to orchestrate the deployment and scaling of an OpenStack cloud onto new machine nodes?
At HP Cloud Services, we think that the possibility of a single suite of tools and a unified API for managing both physical and virtual resources is very exciting. Working with others in the community, we have added the "baremetal"" compute driver to the Grizzly release of OpenStack to provide a common framework for deploying images to real hardware. We will present the current design of the driver, some limitations it currently has, how to simulate an environment for dev/test purposes, and cover some of the quirks in deploying it. After all, this isn't your average hypervisor.
Nati Ueno, Mariano Cunietti, Jurgen Thoelen, Mauricio Rojas
Whether it's regulatory issues, early cloud market questions, licensing uncertainty, or competitive pressures, cloud service providers around the world are facing unique challenges in their markets to build a business and gain traction. In this session, we'll hear from service providers across four continents that have chosen OpenStack as their cloud platform. We'll learn the unique technology and market drivers that led them to OpenStack, how OpenStack is enabling them to compete in the global IaaS market and where they think OpenStack should go next.
Mike Cohen, Dan Mihai Dumitriu, Nnamdi Orakwue, Sam Greenblatt
Software Defined Networking (SDN) is a hotbed of activity with established players as well as well-funded start-ups tackling the largely unsolved problem of large-scale network virtualization in cloud deployments. OpenStack Quantum provides the foundational interfaces to bring in SDN technologies to the OpenStack environment. The ultimate goal is to finally free applications from being aware of specific networking details (like ports, IP addresses etc) and at the same time reducing the operational costs of managing the switching fabric in the era of cloud and mobile computing. Companies have a variety of approaches and solutions to this. In this panel discussion, Dell will moderate a discussion with experts from Big Switch, Midokura Dell, and others, to talk about the evolution of this exciting new space and its relevance within the OpenStack context.
Companies large and small are utilizing the agility and ease of use found in the public cloud. However for large enterprises the public cloud is expensive at scale and availability is not guaranteed For smaller SME's experiencing rapid growth, every day is one step closer to a call from the CFO regarding another rapid growth... cost.
In my presentation I will go over two use cases, one will feature a very large Fortune 100 company experiencing an increasing internal customer base moving to Amazon AWS for rapid provisioning of new infrastructure and how we used OpenStack to provide feature parity, ease of use and self-provisioning to move those customers back into an on-premise private cloud. My second use case will cover a rapidly growing and large consumer of Amazon AWS who was feeling the pain/costs associated with running entirely in the public cloud and found economies of scale in an OpenStack based private cloud.
Both cases illustrate big wins for OpenStack, and the ability to provide massive value to both Fortune 1000's and rapid growth SME's.
Can we build an OpenStack HA solution that allows the same type of automation as Amazon AWS? This session compares an OpenStack HA reference architecture to Amazon to see how OpenStack stacks up.
As OpenStack goes real world production environment, the real world problems like reliability, high availability need to be addressed. In this session, we will explore various time-tested techniques for improving availability of your applications on OpenStack private cloud.
We'll be joined by RIghtScale customer, Samsung SDS, to discuss their revolutionary OpenStack project. Samsung SDS has devised a multi-cloud architecture that leverages OpenStack and AWS to enable cloud-bursting while eliminating latency and security issues. Samsung SDS will review the architecture and technologies they are levering to make this solution possible.
Key take-aways from the presentation would be:
How much for an Openstack Cloud please?
That's it! You're ready! You know everything there is to know about OpenStack.
It's architecture, features and a bunch of other cool stuff.
But finally: How much does an OpenStack Cloud cost?
During this talk, we'll work on two real production use cases to provide you with a detailed, yet simple, financial analysis that will help you budget your future cloud projects. We’ll present the case of a basic simple infrastructure and a second example will treat of a much more complex, high end platform.
Two years ago, Rackspace set out on a quest to build a public cloud in the open, powered by OpenStack. Building the Open Cloud from an idea to a viable deployment of thousands of servers has come with lots of growing pains. We've also gained a little bit of wisdom along the way and are ready to keep pushing for great things. In this case study, we will talk about the various deployment strategies we have used, our experience with them, and how what worked for 200 nodes didn't always work for 1,000 nodes. We'll end by look at how we will get to the first 10,000+ node deployment and stay continuous.
OpenStack high availability has seen a lot of progress since it was defined as an overarching design goal during the Folsom cycle. Not only is there now a reference architecture for highly available OpenStack infrastructure services, but OpenStack is also gradually growing native high availability features. In this session, we are giving an overview of the current state of high availability technology in OpenStack, and an outlook as we enter the Havana cycle.
In particular, this session cover Infrastructure HA, High Availability features in OpenStack Compute (Nova), and High availability improvements in OpenStack Networking (Quantum).
Attendees should have a good general understanding of OpenStack components. High availability experience is a plus, but not required.
Pete Chadwick, Cameron Seader
SUSE®, a pioneer in open source software, provides reliable, interoperable Linux and cloud infrastructure solutions that give enterprises greater control and flexibility. More than 20 years of engineering excellence, exceptional service and an unrivaled partner ecosystem power the cloud infrastructure solutions that help our customers improve resource utilization and speed the delivery of IT services to meet changing market needs. Whether the enterprise is deploying services in a public, private, or hybrid cloud environment, SUSE has a secure, compliant and fully supported solution. Come learn how SUSE Cloud will help you meet the demands of cloud computing in your organization.
Jun Park, Mike Wilson
EIG/Bluehost is one of the largest hosting companies in the world. The scale of our systems at Bluehost have been rapidly increasing. We are challenged with respect to how to manage such large server farms distributed across multiple data centers that are geographically distant. Although the main stream products at BlueHost still reside on traditional hosting platforms that have their own idiosyncratic requirements, we have been seeking an ideal cloud-based solution for an efficient management system in a highly scalable environment. We chose to leverage Openstack in provisioning tens of thousands of servers while meeting customers’ availability requirements.
In this talk, we will present our use case of Openstack in a traditional hosting environment. Although it can be considered odd or appear to be a conflict between 'traditional hosting' and 'cloud,' Openstack has been a good decision with which we could successfully launch our cloud-based traditional hosting products for dedicated servers and virtual private servers (VPS). We also plan on utilizing other benefits that exist in Cloud technologies so as to improve our hosting products as much as possible. We have learned many useful lessons from our unique experience, including the scalability issues that we faced while operating tens of thousands of servers and the stability issues that each OpenStack subcomponent currently has, as well as the architectural design considerations while using OpenStack. We are interested in sharing with the OpenStack community our experiences and code for some new features that we have added to OpenStack. We would like to obtain feedback from the community and also contribute our experiences back to the community.
In this talk we will explain how to make some OpenStack services highly available in active / active mode, or active / passive for some of them. We will also approach the subject of shared storage and how OpenStack is using it. We will share our experience on production environments where High Availabity is one of our top priority.
Aaron Rosen, Eric Lopez, Dan Wendlandt, Salvatore Orlando, Somik Behera (Nicira/VMware)Users will get access to a live OpenStack + Quantum setup and be able to walk through key quantum deployment use cases, with members of the Quantum core development team available to provide guidence and answer questions. At the OpenStack conference 6 months ago we presented a similar Quantum hands on lab and it was standing room only. We'd like to run another session this time, incorporating lessons learned from the previous sesion and also including new Quantum capabilities introduced in the Grizzly release.