November 5, 6, 7, 8
The OpenStack Summit is a four-day conference for developers, users, and administrators of OpenStack Cloud Software.
The OpenStack Summit
Hong Kong 2013
Jonathan Bryce,Dan Wilson,Chris Fischer,Guillaume AubuchonOpenStack Foundation executive director Jonathan Bryce will provide a community update, focusing on why OpenStack is quickly gaining adoption from a user perspective. You will also hear directly from OpenStack users at Concur, DigitalFilm Tree and Shutterstock.
Mark ShuttleworthOpenStack continues to improve with every release and many organisations now run Ubuntu OpenStack in production. But standalone clouds on their own do not deliver business value, this requires a cloud connected to business systems and able to run applications more flexibly and efficiently than before. Mark Shuttleworth will discuss how the advances in interoperability being made with Ubuntu OpenStack and a wide variety of technologies and applications is enabling organisations to get greater value from their Open Stack cloud faster and more easily than ever before.
Daniel SabbahIn a hyper-connected world with mobile, social and embedded intelligence, businesses face pressure to leverage technology to differentiate like never before. Winners will be based on how quick and responsive they are to customer needs and how rapidly they innovate amid pervasive change in a heterogeneous landscape. As an industry, we have an opportunity to be drivers of this next era of rapid innovation by supporting and strengthening interoperability. In his keynote address, Dr. Daniel Sabbah, IBM CTO and General Manager for Next Generation Platform walks the audience through the state of the industry and the imperative of an Open Cloud Architecture to enable rapid and constant innovation. In the last few decades, IBM has been at the forefront of open interoperability initiatives, now including OpenStack, and has helped unlock its potential for its clients. Dr. Sabbah will host one such client, Paul Lu CEO of Wuxi Lake Cloud Tai Cloud Computing, on the stage to talk about their experience with ease in integrating their solution based on OpenStack. IBM applauds OpenStack’s growth, relevance and role in an Open Cloud Architecture.
Uri Cohen,Yaron ParasolOpenStack Heat is gaining momentum as a DevOps tool to orchestrate the creation of OpenStack cloud environments. Heat is based on a DSL describing simple orchestration of cloud objects, but lacks better representation of the middleware and the application components as well as more complex deployment and post-deployment orchestration workflows. The Heat community has started discussing a higher level DSL that will support not just infrastructure components. This session will present a further extended suggestion for a DSL based on the TOSCA specification, which covers broader aspects of an application behavior and deployment such as the installation, configuration management, continuous deployment, auto-healing and scaling. We will also share some of our thoughts on how this DSL can interface with native OpenStack projects, such as Heat, Keystone and Ceilometer.
Everett Toews,Joshua McKentyThis lighthearted talk discusses the cultural divide between the OpenStack developers, system admins and business folks and urges both sides to show each other some love. I will review real use cases/situations and share lessons I've learned to overcome the artificial barriers so we can continue to have nice things like OpenStack.
Kenneth Hui,Dan Radez,Niki AcostaOpenStack is quickly becoming the defacto standard for Open Cloud platforms. But how can someone quickly get started with learning this exciting new technology? This workshop will walk participants through an overview of the OpenStack components and offer practical suggestions and resources for learning OpenStack. To demonstrate one way to get started, we will setup a multi-node OpenStack cloud using RDO and the Packstack utility. The installation will be performed on an RPM-based system.Participants will be introduced to a range of cloud functionalityincluding: Adding new users Adding an image to glance Defining networks in Neutron Starting a new virtual server Creating and attaching persistent storage volumes to virtual servers Storing objects in Swift Using the Horizon Dashboard user interface Instructions to prepare for the workshop can be found athttp://openstack.redhat.com/GettingStartedHavana_w_GRE.A must for OpenStack newbies!
Belmiro MoreiraCERN, the European Laboratory for particle physics, is for a long time running OpenStack test deployments among other leading open source tools to build its private Cloud Infrastructure that helps scientists around the world to uncover the mysteries of the Universe. During the Grizzly cycle CERN moved its production workload to its High Available Openstack Cloud Infrastructure that has multiple cells and is deployed across two computer centers (Geneva and Budapest) totalizing up to 6 MW of capacity. With this talk you will learn the history, architecture, tools and technical decisions behind the CERN Cloud Infrastructure along with tips and techniques for production deployment of OpenStack and the plans for the future.
Jonathan PickardA Strong Foundation was Built A modern, supported, and secure OS. Virtualization in production. A capable staff.Agility, Agility, Agility At a Personal Level An agile methodology is critical to the continued success of the Cloud team at PayPal, but its acceptance amongst “old school” infrastructure engineers was not always easy or expected. Each person took on a personal transformation to remain relevant and important. They accepted new roles that were not always visibly aligned with their previous capabilities. They embraced change on their own schedules, finding their own path, but ultimately embracing it. The team was strong. Now we are stronger and more productive (measurable by # of meaningful deliverables year-over-year).Finding Consistency in a Constantly Changing World These keep us sane: JIRA, Kanban, and Daily Stand-ups JIRA: Easy to track your work. Kanban: Easy to understand what you should be working on. Daily Stand-ups: Ensuring work is happening, and clearing any roadblocks.Get Ahead of the Game and Stay Ahead It's easy to fall behind. We did. Get ahead by being identifying where you haven't been smart. Stay ahead by automating the things that caused you to fall behind. Have each technology or service owner run a simple gap analysis by asking 6 things: Is it highly available and resilient? Is it able to be consistently deployed and managed? Does it utilize configuration management for installation and configuration? Is it documented? Is it secure? Is it monitored?If any of them are no, you've found a gap to address. Once you can answer yes to all of them, your next deployment or upgrade will require much less time. And before you know it, you're ahead.Lessons Learned Each person will transition on their own schedule based on their willingness to accept change. Don't expect an overnight turn-around in the boots on the ground. When it comes though, be prepared to deliver more products and projects than you ever thought possible, and have fun while doing it with cutting edge technologies. There is not one way to do agile. We started with a scrum approach and quickly realized that bi-weekly sprints really didn't work well for us. We found kanban and find it well-balanced for the type of work we are doing.I'm doing Support?! L1+L2+L3 Support is not ideal, but for an immature product, it is required. DevOps until it's documented well and supportable by L1+L2.
Mohammad Banikazemi,David Olshefski,John TraceyAs the core network abstractions provided by Neutron is getting widespread usage, newer abstractions for network services (such as service insertion and service chaining) are being worked out by the Neutron community. However, it remains unclear if the provided level of abstraction is suitable at higher layers of the stack. In particular, applications that are used for deploying various workloads may find a different and higher level network abstraction more useful. In this talk we first present the Neutron abstraction for networking resources including the newly developed network services. We then show how different levels of network abstraction can be more useful at higher layers and propose alternative workload-centric abstractions for network resources. In particular, we discuss Heat and TOSCA as they relate to specifying the network requirements of various real-world workloads.
Keith TobinThis presentation will focus on how we are working with the community to deployment a highly available and qualified OpenStack architecture that meets the requirements and solves the needs of enterprises today. We take a deep look at this highly available active/active architecture stack, how the architecture is constructed and deployed while keeping total cost of ownership low. Each Openstack component in the stack will be examined, we look at how we improve database stability, how we created an active/active OpenStack load balancer to solving this issues associated with using active/passive, and more
Scott CarlsonPayPal has adopted a hypervisor agnostic stance within our Openstack Grizzly cloud. This presentation will cover the details surrounding our grizzly implementation and integration of both KVM and ESX hypervisors under one management umbrella. Grizzly deployment details configuration details for ESX integration Reasons for execution of this strategy benefits and pitfalls of this plan This will be an audience modified presentation of one that I am giving at VMWorld 2013 in San Francisco in August 2013. PayPal is the faster, safer way to pay and get paid online, via a mobile device and in store. The service gives people simpler ways to send money without sharing financial information, and with the flexibility to pay using their account balances, bank accounts, credit cards or promotional financing. With 128 million active accounts in 193 markets and 25 currencies around the world, PayPal enables global commerce, processing more than 7.6 million payments every day. Because PayPal helps people transact anytime, anywhere and in any way, the company is a driving force behind the growth of mobile commerce and expects to process $20 billion in mobile payments in 2013. PayPal is an eBay (Nasdaq:EBAY) company and contributed 40 percent of eBay Inc.'s revenues in 2012. PayPal is headquartered in San Jose, Calif. and its international headquarters is located in Singapore. Scott Carlson is currently a Senior Engineer at PayPal and has been in the field of infrastructure and information security for 15 years within the online banking, education, and payment sectors.
Martin CasadoNetwork Virtualization promises to impact the networking industry much like server virtualization impacted IT. What this means for you is that the networks in your OpenStack cloud can be architected and operated in a way that is drastically better than the manual and opaque networks of the traditional enterprise era. Specifically, network virtualization provides fully-automated provisioning of network connectivity and policy and opens up new opportunities for network visualization, troubleshooting, and security. Now that network virtualization has been around a few years and exists in many production OpenStack deployments, we can talk concretely about which of the promised benefits of network virtualization are already being realized today, and which remain as promising ideas yet to be fully realized. This view into the deployments of network virtualization pioneers, along with a view of key industry trends will help your organization navigate the transition to network virtualization.
Tim Bell,Toby OwenWith the growth and adoption of OpenStack, the real possibility of federating multiple clouds is emerging. But are the software and the community ready for this? What are the real use cases for a hybrid OpenStack cloud? Come and hear how CERN is working with Rackspace to develop solutions for connecting multiple OpenStack clouds, and how a hybrid cloud can help further CERN important research into the origins of the universe. Tim Bell, Manager of Infrastructure for CERN, and Toby Owen, Head of Technical Strategy for Rackspace, will discuss their joint Openlab research project into federation of OpenStack clouds. They will explain the problems they are aiming to solve with federation, potential architectures and blueprints that will guide the research project, current progress, and future plans.
Nick (Nicolas) Barcet,Mehdi AbaakoukOne of the elements that any software development groups envies OpenStack is the project ability to coordinate and maintain a working product based on several hundred contributors wihtout missing a single deadine. This is mostly due to the organisation of the OpenStack project and to the great work of the infrastructure team within the project. As more and more of our contacts are wondering about this miracle that happens every 6 months, we realized that many of them would like to use similar tooling to deliver their own project. For the past few months eNovance has been working on streamlining tooling to deliver this and started a new product offering called the “Software Factory”. This session will explain what changes and modules we had to make to Git, Gerrit, Jenkins and other tools to support OpenStack Nova, Swift, Cinder out of the box in order to provide and intergrated software factory for about any language.
Carmine RimiLearn how we built a business case for OpenStack adoption at Workday and where we are in our journey, some lessons learned, and how you might go from zero to hero. As a rising star in the enterprise software landscape, Workday supplies a software as a service that runs in our home-grown cloud. Adopting OpenStack required identifying and socializing benefits and key performance indicators, not to mention socializing a utopian vision of OpenStack goodness. Once we had buy-in, we began our adoption process with an actionable plan that included evaluating partners, operating systems, components and accessories. By identifying and aiming for early wins, we followed our roadmap to success, and you can too.
George ReeseThe suite of APIs that enable programmatic interaction with OpenStack environments are among the most elegant API suites in the cloud computing world. Currently, however, there's a disconnect between the design of the APIs and how they operate in the real world. This presentation provides an overview of the APIs that comprise the OpenStack API suite and then dives into how they function in the real world across different deployments and how they have changed as OpenStack has evolved. Topics covered in this presentation include: A quick tour of the OpenStack APIs Keystone and OpenStack API security The challenges in working with multiple OpenStack versions Nova volumes and Cinder Nova networks and Quantum Comparing the utility of the native OpenStack APIs to the EC2 APIs
Yusong TanAs we known, it is a big challenge to deploy and operate a large-scale Openstack cloud for tens of thousands of nodes. We've builded a 6,400 nodes openstack platform on TH-2 since a years ago, an enormous amount of work has gone into the deployment and optimization to create this large cloud. In this presentation, we will talk about the architectures for our cloud, the deployment strategies we have used, the experiencies we have gained, and the problems we have met.
Chris FischerShutterstock was founded in 2003 before AWS or the advent of consumer cloud technologies. Over the years Shutterstock has scaled significantly and needed to solve specific challenges of low latency, high volume deployment, the shift to services architecture, high performance databases transactions, unstructured data, and requiring PBs of storage. The infrastructure at Shutterstock is unique, embracing strong expertise in systems management and design as well as high levels of distributed networking pricipals (anycast, ospf, bgp). This talk will be focused on the design concepts of building a custom cloud with shared operational responsibility between your engineering and operations teams, as well as discussion on theory of building the right type of teams for modern infrastructure platforms.
Gregory ElkinbardOpenStack developed largely with KVM; ESXi and VMware are widely considered one and the same. Is it possible to mix and match, and why would you want to? What are the opportunities and challenges of running OpenStack orchestration on an ESXi hypervisor? In this talk, wel look at some real world examples of hypervisor compatibility with OpenStack, examine some performance data and sensitivity analysis, and identify practical approaches to configuring an ESXi and mixed hypervisor environment in your OpenStack cloud.
Shane Gibson,Brian ChongIn the search to find an automation framework to help manage the deployment of your OpenStack clusters, youl find several tool chains and frameworks to choose from. Symantec embarked on a Proof-of-Concept pilot to test various solution options. Symantec initial list of proposed features and capabilities included: ability to manage the bare metal provisioning enrollment of nodes in the cluster deployment of the OpenStack implementation on the cluster full lifecycle management of the node from birth, provisioning, operations, and decommissioning high availability configurations of OpenStack components RBAC user/group security mechanisms multiple physical network segments to segregate various traffic types for security multiple-cluster control capabilities multi-region management capabilities ability to scale to many thousands of nodes Symantec initial design posed significant challenges for all of the frameworks under consideration which required a high-touch professional services engagement with each vendor. This presentation will discuss the pros and cons of the various vendor solutions for deploying OpenStack. The vendor solutions under test were as follows: Mirantis Fuel http://www.mirantis.com/ Canonical JuJu/MaaS http://maas.ubuntu.com/ Dell Crowbar - http://dell.com/crowbar Triple-O - https://wiki.openstack.org/wiki/TripleO Foreman - http://www.theforeman.org/ RackSpace http://www.rackspace.com/ Did we find an implementation that hit all the requirements? Come find out ... !
Randy BiasRunning your own infrastructure *can* be as little as half the cost of running on AWS once you are at scale. OpenStack-based cloud systems can provide the same or similar economies of scale if you leverage the lessons of AWS and GCE when building your cloud. This talk will discuss the economic factors in designing a cost-efficient AWS + OpenStack hybrid cloud. Wel look at the issues involved in repatriating existing applications, and wel show a couple of real-world demonstration of tools that can assist in the repatriation process. Repatriation isn quite as simple as hitting the Easy button, but if you plan your deployment correctly, you can make it work, both technically and economically.
Somik Behera,JC Martin,Tristan Goode,Shintaro Mizuno,Simon Anderson,Yuriy BrodskiyThis session is a panel discussion of OpenStack users having experience deploying OpenStack Clouds in production environments. The panel will be asked to discuss specific Organizational challenges faced and transformation required to bring Cloud into their organization. The challenges of bringing together Server, Network & Storage teams. The rise of converged teams or virtual “Cloud” teams. What does the term “Cloud Architect” mean and is the Infrastructure guy, network guy, server guys, a mix? or something totally new?
David Lindquist,Haiying Wang,Lew Tucker,John EngatesHear directly from the executives who are building the future of cloud technology today.
Diane Mueller,Adrian Otto,Lew Tucker,Chris FerrisOpenStack has experienced a phenomenal growth, but, can it make the leap from 10s of thousands of community members to 10s of millions? Many large software ecosystems that have thrived recently have been powered by developers. So far, enterprise developers have remained largely in the periphery as far as OpenStack is concerned although there are serveral use cases like SDLC, Big Data, PaaS atop OpenStack, Hybrid cloud development, etc. that is aimed squarely at developers. Perhaps, IaaS will never be interesting to developers? Can some of the newer initiatives (like Heat) change this? Attend this panel to hear different perspectives of should OpenStack appeal to enterprise developers and if so how. Please be armed with the tough questions and opinions of what does OpenStack mean to enterprise developers. These stackers have worked with large ecosystem of developers in their past lives and represent different parts of the OpenStack ecosystem. More than anything they are not hesitant to speak their mind and so should you.
Kenneth Hui,Scott LoweAs OpenStack expands its footprint into the enterprise, more and more people who are familiar with VMware vSphere concepts and terminology will be exposed to OpenStack. In order for OpenStack to be successful in enterprise environments, these people who are familiar with VMware concepts and terminology will need to understand the concepts and terminology in OpenStack. In this session, VMware experts who have moved into the OpenStack space will talk about how to bridge the gap between vSphere concepts and terminology and OpenStack concepts and terminology, showing the value in OpenStack for VMware environments and the value of OpenStack to VMware administrators. Please note that this session does not an attempt to give attendees a detailed understanding of how vSphere-related integrations work or are configured, but instead focuses on educating VMware-savvy administrators and users on OpenStack. This session will be valuable for anyone who needs a better grasp of how to talk about both VMware and OpenStack within an enterprise context.
Joe Arnold,Sage Weil,John Griffith,Anand AvatiWhat is the industry doing to advance the architecture of Software-Defined Storage? SDS is going to be very big (>$10B over the next 4 years according to IDC), set to grow rapidly from the small fraction it represents in today market. Listen to what companies, both established and new alike, are doing to enable this data growth, and help transform how clouds consume and deliver storage as a service.
Tim Bell,Rob Hirschfeld,Keith Basil,Boris Renski,Joshua McKenty,Mark BakerChef, Crowbar, Puppet, Tripple-O, Piston, Nebula, eDeploy, etc... each offer more or less complementary techniques to deploy OpenStack, but which one should you chose if you start a new OpenStack project today? What about upgrading your OpenStack, what is the best method to keep your deployment up to date. The goal of this round table is to invite the main experts from the various projects to come and share with us their vision and try to explain why their method is better than the others.
Charlie Dai,James StatenWhile all the hype around cloud computing would suggest that everyone is already doing it, the reality is just a bit behind. There's a learning curve for most organizations that is holding back cloud adoption but in turn presents opportunities for the OpenStack community to accelerate them into the cloud era. In this session, Forrester Research analysts James Staten and Charlie Dai will detail the true picture of enterprise cloud adoption worldwide and here in Asia using fresh Forrsights survey data and qualitative analysis from enterprise client conversations. In this session you will learn: * The true pace of adoption of Infrastructure as a Service (both public and private clouds) * What challenges are holding enterprises back from wider adoption (and what you can do about it) * Where are the sweet spots for enterprise adoption (and how to best capitalize on them) About the speakers:
Ryan RichardSynopsis: OpenStack has been called the largest DevOps open source project in the world and as such there are a lot of lessons we can learn from the project. During this talk, I'll discuss some of the practices OpenStack follows to make the lives of Devs and Ops better as well as discussing where OpenStack can fit into your ToolChains. Overview: 1) Defining the ToolChain 2) Review of OpenStack Practices 3) Where OpenStack fits in 4) What's missing? 5) Q/A and Community Input
Boris RenskiMy job at Mirantis is to figure out how to co-opete in this crowded space. Yes, we are all in OpenStack to build a great open source community. But let face it, we are REALLY in it to make money So. Does Rackspace regret letting go of control? Will Red Hat OpenStack mean death for the start-up distros? And what is the game plan for IBM and HP? I spend my days analyzing every move anybody makes in the ecosystem and figuring what that means for the industry. In this talk Il shed some light on OpenStack coopetition politics, share subjective view on strategies of individual players, and offer some predictions on future OpenStack competition landscape.
Shuquan HuangWee from Intel IT Engineer Computing, which is a group to provide Intel private cloud solution. Currently, we have a 36-node OpenStack Cluster (based on Grizzly) providing over 1000 VMs to our internal customers. Our OpenStack journey started from Essex. During this one year, we ran into a lot of problems and resolved them. It time for us to share with the community about our experience. (1) How to manage version update? I believe it very painful for each half year update. We use Continuous Integration to resolve this problem. We track the source code from github every day. Once the code is changed, our CI server will check out the code and deploy it to our test environment and run all the tests. We add our customized tests into tempest. By using CI locally, we can integrate the lasted code or tagged code daily and resolve the update issue in every dayswork. At the same time, we can track the progress of the community. (2) Tempest is not perfect. Wee using tempest in our CI environment. However, tempest will cause many issues to your environment after you run it. We have added a new module called tempest launcher to help us clean up the environment after running tempest. (3) More powerful and friendly portal. We develop a customized portal called Taurus which is an operation one-stop-shop center. Taurus not only covers all the functions Horizon have, but also integrate with nagios & ganglia. You can also use Taurus to trigger Tempest and view its result friendly. (4) Improve the speed of snapshot. Original snapshot is slow because of rebase and uploading each time. We did some improvement in this feature to reduce the snapshot time, which save our customers lots of time. (5) One click deployment. Based on dodai, we burn a live customized image into USB. And using this USB, it very easy to deploy an OpenStack Cluster. (6) BI in cloud data analysis. We launch Hadoop on our OpenStack Cluster to analysis this cluster log data nightly and archive the data into Ceph.
Steve CurryPrivate clouds have recently begun to gain momentum among enterprises--and with good reason. They're able to deliver the agility and ease-of-use that users have come to expect--without sacrificing any of the security or control that is critical to so many industries. But implemented incorrectly, there's no question that private clouds can be a costly and frustrating undertaking. In this talk, Steve will spell out the most common mistakes that he and his colleagues have seen in their experience working with companies that have attempted to implement private clouds. He'll talk about what has gone wrong for these firms, as well as what steps your company can take to avoid falling into similar traps as you pursue your own private cloud.
Joshua Harlow,Jeff BudzinskiIn this talk, Yahoo will discuss how they use OpenStack today, how they will be using it in the future, and some of the challenges they've encountered along the path to adoption.
Sandeep Singh Kohli,Didier Stolpe,Norifumi MatsuyaYahoo Japan IT infrastructure is a compound of massive, scale-out IT data centers that cover the Tokyo metropolitan area and East Japan. Two years ago, Yahoo Japan initiated a private cloud project. Given the heterogeneous nature of their equipment and future scaling requirements, an open and innovative architecture was paramount. Open source pioneers, Yahoo Japan decided on OpenStack for orchestration. For Server Load Balancer (SLB), the Brocade ADX was selected to ensure application delivery at the speed of the business. SLB Direct Server Return (DSR) is an effective method of deploying a load-balancer in terms of network configuration. The rate of change for this configuration is very high because directly correlated to workloads deployment; hence adding this step to the orchestration workflow is critical to Yahoo Japan modus operandi. DSR is a discussion topic for a design session at the OpenStack Summit in November 3 in Hong Kong, leaving Yahoo Japan today with a gap in the planning to full automation. This session will discuss how Brocade has shaped up LBaaS to address Yahoo Japan cloud necessities.
Sean DagueWhenever you submit a patch to OpenStack it is run through an extensive set of automated tests designed to ensure we always have a working upstream OpenStack. If any of these fail on your patch submission, it will be blocked from merging, and will become a low priority for anyone to review. Debugging these failures is often not straight forward, especially for a new contributor to OpenStack. This talk will look at all those tests in detail, and how to debug and address a failure at each level. What logs to look for, how to understand the interactions between services, and how to handle scenarios that you can't seem to replicate locally, but that are hit in the OpenStack gate. Sean Dague is the OpenStack QA Program Project Technical Lead, and a Nova Core member. He's contributed to OpenStack since the Essex release.
Robert CollinsUsing diskimage-builder your application deployments can be tested easily and deployed quickly. If you are interested in learning how to create custom golden disk images and use them in your CI system and then deployments, you should come to this talk. In OpenStack Deployment (tripleo) we have built a lightweight golden disk image creator called diskimage-builderwhich we use to make testing and deploying OpenStack itself easy and fast. The same tool can be used to prepare golden disk images of any Linux application. I will cover the architecture of diskimage-builder, how to extend it with your own elements and use the resulting images for deploying within an OpenStack cloud.
Hui ChengChina is often regarded as a huge but mysterious market. In the market of consumer Internet, Google, Amazon, Paypal and a long list of U.S Internet companies are almost all lost in China, leaving the huge market to their Chinese counterparts. By contrast, the traditional IT giants, such as IBM, HP, VMware and etc. have dominated the Chinese IT hardware, software and services business for a long time while the local enterprise technology and software companies have few chances to win. What exactly the background reasons behind these facts? How about the cloud business in China? Which type of cloud business is most suitable for the market? How do the Chinese government policies influence the cloud business? Since the cloud ecosystem is still in flux, but rapidly involving, the market pattern will be formed in a few years, and WHO will be the last cloud winner in China, the Internet cloud companies like Amazon, Google, or the traditional IT giants in China, or the Chinese local companies?As the founder and CEO of UnitedStack, the first dedicated OpenStack company in China, Hui Cheng has deep insight of Chinese cloud business and strategies. In this topic, he will try to answer the above questions, and also will deeply dive into the cloud status quo, as well as OpenStack opportunities in China.
Florian Haas,Sebastian KachelThis presentation covers practical experience from building an OpenStack private cloud with a specific emphasis on high availability. Based on a production deployment of OpenStack Grizzly at Pixelpark (a hosting company based in Berlin, Germany), we explore and explain highly-available data storage with Ceph, OpenStack service high availability with Pacemaker, virtual machine high availability in OpenStack Compute, and high availability considerations for OpenStack Networking. We also outline the lessons a previously conventional hosting company has learned from moving to a private cloud deployment, both from a technical and organizational perspective. About the speakers: Sebastian Kachel is an all-around devops guy at Pixelpark, and has been involved in the OpenStack deployment project since its inception. Florian Haas has provided consulting, training and guidance to Pixelpark over the course of the project.
Koby Holzer,Liran CohenLiveperson Infrastructure in a nutshell Requirements and goals achieved! OpenStack is running the Core business of Liveperson The diffrent openstack components in use and why 1 year in Production and Growing like Crazy Infrastructure as a building block Tips to get you started Future plans to continue with Openstack and with bleeding core edge projects.
Alejandro Comisario,Leandro ReoxThis time the Mercadolibre Cloudbuilders are going to explain their Openstack multidisk Swift Architecture, the journey that they traveled to reach the their current deployment, configuration and tuning hints to achieve constant 300.000rpm and persist the initial performance in time, common mistakes, common failure scenarios and how to go trough them. - Learn how the biggest e-commerce company on latin-america stores all the images, of all their items on Openstack Swift - How we moved from a trillions of dollars storage schema to an OpenSource , hardware comodity based object storage system - How we can handle +300.000 rpm and +85 millions of users, with no performance degradation , key caching and black magic - In depth technical explanation of our Swift multidisk architecture - Lessons learned - How we planned to scale up and out Dont be afraid to Swift up your storage!