Loading…
Visit the OpenStack Summit page for the latest news, registration and hotels.

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

User Stories [clear filter]
Tuesday, November 4
 

11:15

Building the Cloud Ecosystem OpenStack, Docker, Kubernetes


Summary:



Openstack has slowly and steadily grown into a main stream IaaS cloud solution provider technology. At Symantec, we have primarily used Openstack to virtualize our datacenters and help drive infrastructure automation to ensure maximum use of our compute, network and storage capabilities. But many of our enterprise use-cases demand more than just virtualized environments. There is a need for certain applications to run on linux container like environments and make use of some of the advantages provided by these platforms.



In this presentation we will talk about how Openstack can be used along with other open source cloud technologies to build an ecosystem to solve all use cases and problems faced by today s enterprise world. We will try to present a performance based evaluation of the three most popular technologies out there Openstack, Docker and Google s Kubernetes and propose a solution that may help you achieve infrastructure automation in its complete sense.

 

The following topics will be covered in this presentation:




· Categorize enterprise use cases that can be solved using Openstack, Docker or Kubernetes.




· Performance analysis and benchmarking




· How can Openstack be used as the central scheduling and orchestration engine to integrate with other open-source cloud technologies?

 

· Can the ecosystem eventually be scalable, reliable, secure and truly multi-tenant?

 

Speakers
avatar for Shane Gibson

Shane Gibson

Sr. Principal Infrastructure Architect, Symantec, Corp.
Shane Gibson is the Sr. Principal Infrastructure Architect at Symantec for the Cloud Platform Engineering (CPE) team. He is responsible for the infrastructure design of the underlying platforms, operating systems, tools, and application stack that enables the OpenStack clusters within... Read More →
avatar for Chinmay Naik

Chinmay Naik

CEO, Shodat Inc
Chinmay Naik is a Sr. Principal Software Engineer in the Cloud Platform Engineering team at Symantec. He is currently working on Openstack development to design and implement features for solving real world production use cases to make IaaS cloud at Symantec highly available, reliable... Read More →


Tuesday November 4, 2014 11:15 - 11:55
Room 242AB

12:05

Seamless Migration from Nova-network to Neutron in eBay Production
Several datacenters were using folsom nova-network in the eBay production, with thousands VMs in service. We successfully upgraded to openstack Havana version, adopted neutron as network service instead of nova-network and replaced Linux bridge by Open vSwitch. One of the most important requirement to us was to avoid network service break for existing VM instances during nova-network to neutron migration, i.e. least downtime in the data plane.

In this session, we will share our story of the seamless migration from the folsom nova-network to havana neutron in the eBay production. The work covers both control plane and data plane. The control plane part includes neutron net/subnet/port management, and SDN controller integration, while the data plane part includes tap device migration from Linux bridge to openvswitch bridge, DHCP service, security group, and Libvirt/KVM configuration update.

Speakers
CL

Chengyuan Li

Sr MTS Software Engineer, eBay
Chengyuan Li is a member in eBay Kubernetes team, his focus area is host-runtime and storage in Kubernetes. Before joining Kubernetes project, he worked in computer and network area for eBay cloud.
avatar for Han Zhou

Han Zhou

Cloud Architect, Intel
Han Zhou is the cloud architect in Ebay Inc, and is now focusing SDN design for data centers.


Tuesday November 4, 2014 12:05 - 12:45
Room 242AB

14:00

Design and Operation of OpenStack Cloud on 100 Physical Servers (NTT DOCOMO)
You will face many problems when you start designing your OpenStack Cloud because of a lack of full design architecture information. For example, there are many Neutron plugins, but it is difficult to choose the best plugin and its configuration to get a high throughput of a Virtual Machine (VM) and achieve a High Availability (HA) of L3 Agent. Also, we couldn't find information for how much computing resource (CPU, Memory and HDD) is required for management and operation servers (e.g. API, RabbitMQ, MySQL and Monitoring etc.).

We built OpenStack Icehouse Cloud on 100 physical servers (1600 physical cores) without using commecial software, and did several performance and long-run tests to address these problems.

In this talk, we will present performance comparison of Neutron ML2 plugin implementations (Open vSwitch and Linux Bridge), tunnelling protocols (GRE and VXLAN) and physical network configurations (Network Interface Bonding and Server Side Equal Cost Multi Path) to achieve 10Gbps at a VM, and the L3 Agent HA we implemented. Also, we will present how much computing resource we used and each server loads to operate the cloud. Finaly, we will share our Ansible Based OpenStack deployment and management tool.

Key topics include:

- Performance comparison of OSS Neutron ML2 plugins (Open vSwitch and Linux Bridge) and tunneling protocols (GRE and VXLAN)

- Performance comparision of redundant network configurations (Network Interface Bonding and Server Side Equal Cost Multi Path)

- HA of L3 Agent (ACT/STBY) we implemented

- Ansible based deployment/operation tools

- Items we must watch for OpenStack operation

- Hardware specifications and resources we used to operate the Cloud

We will share a full design architecture and hardware sizing information for a large scale cloud and prove OSS based Neutron can handle a hundred servers.  

 

Speakers
avatar for Ken Igarashi

Ken Igarashi

Sr. Research Engineer, NTT DOCOMO, NTT docomo
Ken Igarashi is one of the first members of proposing OpenStack Bare Metal Provisioning (currently called "Ironic"). He is leading OpenStack based private cloud team as a developer and operator.
avatar for Hiromichi Itou

Hiromichi Itou

CTO of Virtualtech Japan Inc., Virtualtech Japan Inc.
Hiromichi Ito co-founded Virtualtech Japan Inc. in 2006 and he is currently working to providing consultation service for public and private cloud. He is one of the first members of proposing OpenStack Bare Metal Provisioning (currently called "Ironic").
AM

Akihiro Motoki

OpenStack Neutron/Horizon Core Developer
NEC Corporation


Tuesday November 4, 2014 14:00 - 14:40
Room 242AB

14:50

Building Telco Grade Cloud Services with OpenStack

The first challenge was to gain overall knowledge of OpenStack in order to deploy and manage it. The second, much bigger challenge was to orchestrate a private cloud at Orange with OpenStack as its middleware as well as various other tools and processes for a real world use case scenario. 

OpenStack is a set of modules which can be deployed in various ways. Which is the right way to deploy OpenStack was a key challenge.

Our aim was to deploy and to complement OpenStack with further, customized components in order to create a private cloud at Orange which would provide multiple, complex services to its internal customers: moving from IaaS middleware to Cloud services embracing infrastructure management, Cloud offers management and customer onboarding.

A telecom relies on Mirantis as a key partner partner to streamline our learning curve. Our collaboration included trainings, OpenStack Fuel as the deployment mechanism as well as development of a VxLAN plugin. Picking a partner who has in-depth expertise in deploying OpenStack in production at scale is of high importance for the success of a large enterprise grade OpenStack project.

Also, our collaboration with Mirantis helped us better understand how to collaborate with the OpenStack community and to contribute upstream - for example by providing manifests for OpenStack Fuel. Our goal is to share components we develop in order to make OpenStack better and more viable in coming years.


Speakers
avatar for Frank Weyns

Frank Weyns

Director, OpenStack International, Rackspace
Frank Weyns is the Director Sales Engineering EMEA at Mirantis. An authority on OpenStack, Linux and cloud computing, he has 20 years of experience in enterprise open source, with heavy concentration on Openstack. He is a trusted advisor to the financial and telecom world who lectures... Read More →


Tuesday November 4, 2014 14:50 - 15:30
Room 242AB

15:40

Docker Meets Swift: A Broadcaster's Experience
RAI (Radiotelevisione Italiana), Italy's national public broadcasting company, has to deal with the growing cost for new storage hungry media formats (e.g., 4K UHDTV), while also enabling its employees to collaborate and manage content across geographically separated sites throughout the media production cycle. To address these needs, we have integrated Docker with Swift via a generic mechanism we call "storlets". Integrating Docker with Swift keeps the storage cost low as well as co-locating compute to avoid costly network transfer for a wide range of media workflows. In this talk we will present the problem RAI is solving including its requirements, present the storlet mechanism, demonstrate how storlets can be used for media workflows and other scenarios, and place our work in the context of other efforts.



Speakers
avatar for Alberto Messina

Alberto Messina

R&D Area Coordinator, RAI - Radiotelevisione Italiana
Alberto Messina (born 1971 in Torino, MSc. Electronic Engineering, PhD) began his collaboration as a research engineer with RAI in 1996. Since then, he has been involved in several internal and international research projects in digital archiving, automated documentation, and automated... Read More →
avatar for Eran Rom

Eran Rom

Research Staff Member, IBM
Eran Rom is a researcher in the IBM Haifa research lab focused on systems and storage. In recent years Eran has been mostly involved with object stores doing architecture research and development projects centered around object stores.


Tuesday November 4, 2014 15:40 - 16:20
Room 242AB

17:30

Building OpenStack Near the Users


Traditional cloud datacenters put all their eggs in a few really big baskets. But what if the applications need to be near the users?  Come see Sridhar Basam, Sean Collins and Anthony Veiga show you how they built a distributed multi-site environment that gets the data as close to the users as they can running OpenStack.

Traditional cloud datacenters put all their eggs in a few really big baskets. What if the applications need to be near the users?  Come see Sridhar Basam, Sean Collins and Anthony Veiga show you how they built a distributed multi-site environment that gets the data and applications as close to the users as they can running OpenStack.

 

Speakers
avatar for sridhar basam

sridhar basam

Principal Engineer II, Comcast
Sri is the Lead Engineer at Comcast's Openstack based private cloud. In addition to Openstack, Sri's interests include performance optimization, the kernel, networking.
SM

Sean M. Collins

Senior Software Engineer, Comcast
Sean Collins is a Senior Software Engineer at Comcast and is known for his work on Neutron. Sean has been hard at work introducing IPv6 support for OpenStack, landing his changes in the Juno release.
AV

Anthony Veiga

Network Engineer, Comcast, Comcast
A Network Engineer with Comcast since 2004, Anthony has worked in DOCSIS, IP routing and IPv6. With IETF experience, plublications with the SCTE, and work with the bufferbloat.net project, Anthony has worked to improve the state of the Internet and was involved with Comcast's IPv6... Read More →


Tuesday November 4, 2014 17:30 - 18:10
Salle Passy
 
Wednesday, November 5
 

09:00

Suning.com and Heat, Explore and Enable Elastic Cluster for Internet Applications

Suning Cloud Commerce is one of the largest privately owned retailers in China. Suning has more than 1600 stores covering over 700 cities of Mainland China, Hong Kong and Japan, and its e-commerce platform, Suning.com ranks among top three Chinese B2C companies. There are more than 180,000 employees, thousands of mixed power, x86, storage servers and tens of thousands of virtual machines from several large data center across China, HongKong and Japan. KVM, oVirt and virtualization technologies are widely used, and there are also very large server farm for VDI.

Suning starts OpenStack journery from Jun 2013, and in 2014, an OpenStack based production environment has been put in use, to server Android Mobile Application and Service, Search Engine, Retailer related business needs, plus critical code repository server. More than half are resource intensive and heavy loaded virtual machines Cinder multi-backend provides extra storage by leveraging LVM and Gluster. A light VDI solution to serve desktop users by leveraging open source spice. Within this deployment, network separated and openvswitch bonding is supported.

The successful story leads to large scale OpenStack deployment, and heat's auto-scaling pretty fits the scenarios for Suning's popular Internet applications. Here we use Heat to assiste 2 kinds of applications:


  1. Eleastic Web computation. The web cluster that we are using is IHS + WAS, it is one of major web applications that serve tens of million visitors, and the visits changes sharply. We need to adjust the cluster scale based on visits and also CPU load;



  2. Real-time image thumbnail processing. This requires to generate image and document preview, plus real-time video-transcoding. The distributed scheduling framework is used, Beyond this, we also leverage OpenStack Heat, to dynanically create or destroy workers based on requests. For example, there are concurrent image thumbnail requests, at minimum, there are less than 10 works per seconds to serve the requests. However during peak time, there might be thousands of workers to be created. This demand the clusters to be expand or shrink fully automated and fast response. Once the request volume drop down, the resources should be released immediately to be reused by other kind of jobs.




We will share our story, experience and practice to enable such kind of big cloud and how to support them, best practise and lesson learn.

Speakers
LJ

Long Jin

Director, Cloud Infrastructure
avatar for Qiming Teng

Qiming Teng

Research Scientist, IBM
I have been working on virtualization technologies since 2005, though this is the first time I get myself involved in an open community. It has been an exciting experience so far where I can work with so many talents around the globe, all with different backgrounds, interests, expertise... Read More →


Wednesday November 5, 2014 09:00 - 09:40
Salle Passy

09:50

Bringing Research to the Cloud: the NeCTAR Research Cloud
Australian researchers are reaping the benefits of access to an accessible, collaborative and scalable OpenStack-based cloud computing infrastructure provided through the NeCTAR project.

Modern research is increasingly collaborative in nature and dependent on access to large scale research data, sophisticated computational models and research domain-oriented software tools and services. The NeCTAR Research Cloud is providing a national computational infrastructure supporting the computational and collaboration needs of a rich diversity of research domains and scientific disciplines in Australia: from the hard sciences to the humanities and cultural studies.

This talk will tell some of those stories: of the research communities, large and small, who are using the NeCTAR Research Cloud as a platform to provide collaborative access to research data, scientific models and research computing tools. Stories to be covered will be drawn from Astronomy, Marine Science, Geophysics, Biology, Genomics, Environmental Science and the Humanities. We will show how cloud computing is supporting improved innovation, efficiency and productivity in research.

We will also discuss our interest and first steps in supporting interoperability and collaborative access with similar initiatives around the world.

The NeCTAR Research Cloud has supported Australian research since January 2012 and has helped to bring more and more Australian research into the cloud . The NeCTAR Research Cloud comprises 8 regional sites (nodes) operating as a federation of Cells within a single national OpenStack cloud fabric. The Research Cloud currently comprises over 15,000 CPU cores, and will grow to 30,000 CPU cores before the end of 2014.

The National eResearch Collaboration Tools and Resources (NeCTAR) Project is an Australian Government project funded under the Super Science initiative.

Speakers
GM

Glenn Moloney

Director, NeCTAR, NeCTAR
NeCTAR is partnering with Australian research organisations to build a nationally federated research cloud platform to support collaborative research. We also fund programs to deploy research-domain oriented collaborative environments on the NeCTAR cloud. My own personal history... Read More →


Wednesday November 5, 2014 09:50 - 10:30
Salle Passy

11:00

How Tapjoy Uses OpenStack to Deliver Billions of Daily Requests Globally
Founded in 2007, Tapjoy is a mobile performance-based advertising platform that drives deep engagement and monetization opportunities for app publishers, while delivering valuable, engaged consumers to some for the world s biggest brand advertisers. With a monthly reach of hundreds of millions of mobile users and billions of requests, Tapjoy needed an environment that would allow them to accelerate the development and progress of their Big Data algorithms. They chose OpenStack.

Join Wes Jossey, Head of Operations at Tapjoy, who will discuss why and how Tapjoy is augmenting their AWS footprint with OpenStack in a DevOps model, and what they hope to see from OpenStack in the future.  


Speakers
WJ

Weston Jossey

Head of Operations, Tapjoy
Weston Jossey is an engineer, DevOps junkie, and aspiring entrepreneur.


Wednesday November 5, 2014 11:00 - 11:40
Salle Passy

11:50

How SAP SE Transformed a Productive Environment from Virtualization to Cloud Computing with OpenStack
Today SAP SE (formerly SAP AG) is running an OpenStack environment with several thousand virtual machines. The SAP project lead and B1 Systems will share the initial situation at SAP and what it took to integrate OpenStack with the existing environments. B1 Systems GmbH was involved in this process right from the start being responsible for design, implementation, operation and support of SAP's largest OpenStack landscape. Back in 2011 SAP SE had been using a virtualization platform which was partly automated. This environment did not scale fast enough to meet the increasing needs of SAP. In 2012 B1 Systems started the design of a cloud platform with the OpenStack Folsom release. The OpenStack environment was designed to replace the regular virtualization landscape. By using Puppet for configuration management, it was possible to provide a fully-automated and fast-growing environment to consolidate the needs of multiple customers and projects. 

SAP's current OpenStack landscape is based on the Icehouse release of OpenStack and encompasses nearly all available OpenStack components.

In July 2014 SAP SE officially committed to OpenStack. SAP's and B1 Systems' collaboration has furthered the collaboration with the OpenStack community with regards to feature requests and bug fixes. B1 Systems is part of the OpenStack Community since 2011 and has contributed to nearly all parts of the dynamic development. Right now B1 Systems is preparing the implementation of the upcoming Juno release at SAP.

Speakers
CC

Claus-Henning Cappell

Senior Architect, Cloud Infrastructure Delivery, SAP SE
avatar for Ralph Dehner

Ralph Dehner

CEO, B1 Systems GmbH
Ralph Dehner is the founder and CEO of B1 Systems GmbH , a German provider of Linux/Open Source consulting, training, development and support. B1 specialize in tailor-made solutions based on virtualization (XEN, KVM), high availability (Pacemaker) and cloud computing (OpenStack... Read More →


Wednesday November 5, 2014 11:50 - 12:30
Salle Passy

14:40

OVH/RunAbove: Going Beyond the Performance Limits of Public Cloud
Discover how the OVH Group brought exceptional computing, storage, and network performance to our cloud computing users with RunAbove.

Our mission: Maximize VM performance, surpassing traditional bare metal servers. 

Our weapons:       

A 1 VM per host model    

Latest generation IBM Power 8 and Intel Xeon CPU’s 

Different choices of VM disk storage for perfect high-availability and performance combinations  

OVH's first class network integrated into OpenStack 

Speakers
avatar for Maxime Hurtrel

Maxime Hurtrel

Product Manager, RunAbove
Product Manager for RunAbove & controls the incubation of  startups for OVH group. Focused on user experience, he has expertise in both B2B and B2C web products and managing small teams dedicated to design and innovation. Former Product Owner (PO) in the healthcare industry and for... Read More →
avatar for Nicolas Le Manchet

Nicolas Le Manchet

DevOp, OVH
DevOp working with RunAbove assisting customers in successfully migrating their services to the cloud. He strongly believes in the benefits of Free Software which provides cutting edge features and improved security. With passion and knowledge his duties vary from managing fleets... Read More →


Wednesday November 5, 2014 14:40 - 15:20
Room 243

14:40

Scaling Embedded Software Engineering Exercising Continuous Development and Integration on OpenStack
In this presentation from practice, BMW Car IT and Intel tell the story how they used home-grown OpenStack clusters as build farms for their automotive software development. The combination of Jenkins and OpenStack made it possible to scale up the number of build targets into the hundreds while keeping commit cycles in line with CI expectations. This talk will focus on lessons learned and detail the effects of performance gained by scaling core numbers and advancing CPU generations.

Speakers
TH

Tobias Henkel

BMW Car IT representative
avatar for Markus Leberecht

Markus Leberecht

Solution Architect, Intel Corporation
Markus Leberecht works as a data center and cloud architect in the Intel's Technical Business Development. In this position he specializes in energy efficiency at the data center, latest server technologies and cloud computing.


Wednesday November 5, 2014 14:40 - 15:20
Salle Passy

15:30

Meet The Jock of College Sports Backend Infrastructure - Pac-12
This session hosts the OpenStack Swift infrastructure mastermind behind both the Big Ten s pioneering partnership with News Corp and the current innovations being employed by the Pac-12 s system architecture and technology director, Scott Adametz. He presents his strategy for rethinking the possibilities of live sports content, the data storage distribution challenges, and feeding our voracious sports content appetite flawlessly over web and mobile. He s joined on stage by open-sourced storage expert Joe Arnold in examining the behind-the-field action of media streaming. Together, they re reimagining digital content not just for the world of college athletics but for the future of all live, digital content.

Speakers
SA

Scott Adametz

Director, System Architecture & Technology, Pac-12 Networks
Scott Adametz is Director of System Architecture & Technology for Pac-12 Networks. Headquartered in San Francisco s SOMA district, Pac-12 Networks is the new content and multiplatform media company of the Pac-12 Conference, a leader in collegiate athletics that is made up of 12 of... Read More →


Wednesday November 5, 2014 15:30 - 16:10
Salle Passy

15:30

Overstock.com's Journey to the Cloud: Our Trials and Triumphs

Overstock.com's cloud architect Mike Smith discusses the company's adoption of OpenStack. Topics discussed include the challenges encountered using OpenStack for traditional workloads, a showcase of Overstock's home-grown orchestration layer based around the OpenStack python APIs, and some of the recent OpenStack-powered services the company has launched including the Overstock Pet Adoption program which connects visitors to animal rescue shelters around the country.


Speakers
avatar for Mike Smith

Mike Smith

Lead Cloud Systems Architect, Overstock


Wednesday November 5, 2014 15:30 - 16:10
Room 243

16:30

BBVA Bank on OpenStack
BBVA Bank on OpenStack

Due to unproven scalability and security concerns, enterprises take a wait and see approach to Open Source deployments much less OpenStack. Yet, not only are these deployments feasible but also can yield substantial multi tenant efficiency, agility, speed, dynamic and security advantages over legacy frameworks. While a hybrid cloud approach is quite popular for agile services delivery, for some enterprise segments a private cloud is essential in order to comply with regulations.

In this session, we will explore how Banco Bilbao Vizcaya Argentaria SA (BBVA), a Spain-based global financial group, banks on OpenStack. BBVA has designed an automated, multi tenant service Cloud that provides:



  • Efficient, granular security: Via a global policy framework from Nuage Networks


  • Agility: Via utilization of KVM as a virtualization hypervisor


  • Speed: Provisioning and delivery of services in near real-time via the RedHat OpenStack distribution



Moreover, we show the integration of Neutron based on external SDN overlay solutions in order to improve the networking and security functionalities.


This will be an eye-opening session you can bank on it! (Seguro que si!)

 

Speakers
DC

Daniel Chavero Gaspar

Innovation Team, BBVA
New technologies.
avatar for Jose Luis Lucas

Jose Luis Lucas

Innovation Engineer, BBVA
CS

Chema Sanjose

Global Head of Innovation in Technology
Jose-Maria (Chema) Sanjose is the Global Head of Innovation in Technology, for Digital Bank division at BBVA (http://www.bbva.com). His role is integrate in Digital Bank every innovation which gives BBVA Digital Bank advantage offering better service to BBVA's customers. He and his team covers cloud strategy, cloud platforms, cloud development and devops practice, designing and deploying infrastructures for... Read More →


Wednesday November 5, 2014 16:30 - 17:10
Salle Passy

17:20

OpenStack on OpenCompute with Ironic
OpenStack has been opening up the stack from the cloud down, while OpenCompute has been opening up the stack from the metal on up.  Come hear how we used Ironic to tie the two together, the pain points we hit in the process, and some of the extra bits of managing an OpenCompute deployment outside of Ironic.

Speakers
PM

Peter Martini

DevOps
Peter Martini is a part of the team working on Data Center Operations and OpenStack at Bloomberg. He started in the industry supporting big iron systems in the field for Sun Microsystems, then moved to Bloomberg supporting general data center operations, and has alternated between... Read More →


Wednesday November 5, 2014 17:20 - 18:00
Salle Passy