[Main Conference] [clear filter]
Tuesday, October 27

11:15am JST

OpenStack at NTT Resonant: Lessons Learned in Web Infrastructure
NTT Resonant Inc., one of NTT group company, is an operator of the "goo" Japanese web portal and a leading provider of Internet services. NTT Resonant deployed and has been operating OpenStack as its service infrastructure since October 2014 in production. The infrastructure started with 400 hypervisors and now accommodates more than 80 services and over 1700 virtual servers. It processes most of 170 Million unique users per month and 1 Billion page views per month.
We will show our knowledge based on our experience. This talk will specifically cover the following areas:

  • This shortened the timeframe to provide virtual server to service developers.

  • This improved the efficiency of development cycle by making it possible for service developers to scrap and build virtual servers by themselves.

  • How we successfully introduced OpenStack under short timeframe from planning through migrating most of services from existing infrastructure.

  • How we designed our system using Nova, Glance, Swift, Keystone, Horizon and Neutron.

  • Which components we decided to integrate with OpenStack and not to integrate. OpenStack flexibly allows us to separately use appliances such as load balancer and firewall.

  • The distribution we chose was packages from RDO community of the Icehouse release.

  • How we modified the code to meet our requirement and our operation rules, mainly Horizon and some of APIs.

  • How we use Puppet, Zabbix and ticket system to manage and operate 400 hypervisors in efficient.

  • How we improve and automate our workflow including both of virtual server provisioning and web application configuration by using Puppet framework.

  • We show our whole actual process from the creation of virtual servers to cutting over a web service in connection with OpenStack, Puppet, Zabbix and a few manual configurations for network appliances.

  • How we monitor more than thousands virtual servers swarmed over OpenStack as a team responsibility of operation monitoring system in company.

  • The issues and future plan we have been addressing such as upgrade.

avatar for Tomoya Hashimoto

Tomoya Hashimoto

I have been working at NTT Resonant Inc. since 2001. I took a role of the design, deployment and operation for our services such as goo blog, goo QA site and so on. I'm in charge of the design, deployment and operation of the company-wide infrastructure services for most of company... Read More →
avatar for Toshikazu Ichikawa

Toshikazu Ichikawa

Senior Research Engineer, NTT Software Innovation Center
Toshikazu Ichikawa is a Senior Research Engineer for NTT Software Innovation Center of NTT Corporation in Tokyo, Japan. He has been working to design and develop the cloud infrastructure system using OpenStack for NTT Group companies more than one year. Prior to this job, he was... Read More →
avatar for Kazuhiro Tooriyama

Kazuhiro Tooriyama

Engineer, NTT Resonant
Kazuhiro Tooriyama has been working at NTT Resonant Inc. as an engineer. He mainly designs and administer operational workflows for a private cloud based on OpenStack in the company.

Tuesday October 27, 2015 11:15am - 11:55am JST

12:05pm JST

Know-how of Challlenging Deploy/Operation NTT Docomo's Mail Cloud System Powered by OpenStack Swift
DOCOMO MAIL is 24/7 cloud mail system which has accesses from over 20 million people. This mail system stores user's mail archive in OpenStack Swift with Peta Byte scale capacity deployed by NTT DATA.

We have been successfully operating this service since Sep 2014 without any downtime. In this session, we'll present the actual issues and challenges we have faced and conquered.

Here're some specific points we'd like to highlight.

 * No service degrade, no downtime.
 * Massive scale and still growing.
 * Hundreds of servers operated by few people.

avatar for Sosuke Kakehi

Sosuke Kakehi

Project Manager at DOCOMO MAIL project
Sosuke Kakehi is a systems architect at NTT DATA.He works to System designing, Construction and Project management. He has experienced a project management, system design and deployment of large-scale object storage (for Cloud Mail System) using OpenStack Swift, since 2013.

Ryosei Kasai

OpenStack engineer at NTT DATA

Masaaki Nakagawa

OpenStack Swift engineer at NTT DATA
Masaaki Nakagawa is a OpenStack Swift engineer at NTT DATA. He has experienced designing and deploying large scale OpenStack Swift for 5 years.

Tuesday October 27, 2015 12:05pm - 12:45pm JST

2:00pm JST

The Evolution of OpenStack Dev/Test Environment Setup at eBay
Setup OpenStack Dev/Test environment is a common daily task. But Devstack doesn't work for us because need our own openstack build and configuration management tool to deploy to mimic the real thing in production. This talk will go through a history of how we build a cloud for daily dev/test use. How we transitioned from baremetal to cloud over cloud environment, and eventually adopting Kolla to setup the test cloud. We'll go through the lessons learned during the process and changes we made to make it happen.


The talk will be organized as follows:

  • Overview

  • A brief history of tools we used at eBay for OpenStack dev and test: Baremetal to VM to Docker using Kolla

  • Changes we made to run test cloud with Kolla (cell setup, NSX overlay network etc.)

  • Lessons learned from building a cloud for dev and test

  • Use cases: use Kolla for OpenStack upgrade test

  • Something about CI

  • Challenges


Tuesday October 27, 2015 2:00pm - 2:40pm JST

2:50pm JST

Multi-location Implementation of OpenStack
ConoHa supports multi locations which are Japan(Tokyo), USA(San Jose) and Singapore.  In this session, we will talk about the backend system operation when deploying OpenStack systems(multi locations) on single KeyStone, from a developer’s point of view.

This session will cover:
- Replication of KeyStone databases.
- In the case of the token table, Striping the table in order to reduce replication cost

About GMO Internet, Inc.
GMO Internet Group, headquartered in Tokyo, is a leading force in the Internet industry offering one of the most comperehensive ranges of Internet services worldwide.  We are providing a public cloud called “ConoHa” and “GMO APPs Cloud” as part of our services. Both are based on OpenStack.

avatar for Naoto Gohko

Naoto Gohko

Naoto-Gohko, GMO Internet, Inc.
our product by OpenStack. ConoHa public cloud; https//www.conoha.jp/en/ GMO AppsCloud; https://cloud.gmo.jp/en/
avatar for HYUNTAE PARK


Assistant Manager, GMO Internet, Inc.
Hyuntae Park, GMO Internet, Inc. hyuntae-park@gmo.jp https://www.linkedin.com/profile/preview?locale=ja_JP&trk=prof-0-sb-preview-primary-button

Tuesday October 27, 2015 2:50pm - 3:30pm JST

3:40pm JST

Challenge and Learnings of Running OpenStack Bare Metal at Yahoo
Alternate Title (Living the OpenStack Bare metal Dream).

This talk is a follow up of  the talk from Vancouver "Horizontal Migrations: How Yahoo is Moving it's Existing Hardware into a Bare-metal Cloud" by James Penick.

I will be sharing in more detail our current OpenStack bare metal infrastructure including:

  • Bare metal Hardware Lifecycle

  • Quota / Availability Zone Management

  • Random testing of bare metal hardware inventory

  • Challenges we faced

  • What we did to solve some of the problem

  • What we learned that we can apply to ironic

I will also share my day to day experience of running OpenStack bare metal cluster with 10's of thousands of nodes.

avatar for Simon Chung

Simon Chung

Principal Production Engineer, Yahoo!
Simon is a Principle Production Engineer at Yahoo on the Openstack operations team. He has been working on Openstack for the last 2 years operating one of the largest installation of VMs and Baremetal with experience from Grizzly -> Havana -> Icehouse -> Juno. Simon has been working... Read More →

Tuesday October 27, 2015 3:40pm - 4:20pm JST

4:40pm JST

Public Cloud Meeting High Security Standards Based on Openstack
Private High security customers and government customers are looking to use public clouds and pay as you go models.  The challenge they find is to meet tough regulatory demands but at the same time be flexible with infrastrucrure and resources in a public cloud model.  City Network is the first public cloud provider in Europe to offer a high security, Semi-private, pay as you go model based on Openstack Kilo release.  This is the use-case and story on how we got this offering in to production.

City Network is europes first public cloud provider running on OpenStack. Join us and discuss the challenges and benefits of Openstack in public clouds and how we can continue to improve it.

avatar for Kim Hindart

Kim Hindart

Chief Security Officer, City Network Hosting AB
Huge Open Source fan boy and developer with a background in mobility and especially Symbian and Android. I am a really geeky dude that was forced to handle security and compliance so the only coding I do is on my spare time. Luckily that's also what I like to do in my spre time, work... Read More →
avatar for Tobias Rydberg

Tobias Rydberg

Team Lead R&D and Senior Developer, City Network
Senior software developer at the OpenStack based public cloud City Network, based in Sweden but with locations around the globe. More than 10 years of industry experience.Tobias main focus is developing City Networks cloud management platform - used for all our OpenStack installations... Read More →

Tuesday October 27, 2015 4:40pm - 5:20pm JST

5:30pm JST

Unveiling CERN Cloud Architecture
CERN, the European Laboratory for particle physics, is running OpenStack for its private Could Infrastructure among other leading open source tools that helps thousands of scientists around the world to uncover the mysteries of the Universe. 

In 2012, CERN started the deployment of its private Cloud Infrastructure using OpenStack. Since then we moved from few hundred cores to a multi-cell deployment spread between two data centres running on more than 130k cores with multiple hypervisor technologies.  After 3 years learning how deploy and manage OpenStack at scale, we now look back and discuss the problems and challenges that we overcome/face and give you some tips and tricks that can help you manage and scale your OpenStack Cloud.  

With this talk you will learn the history, architecture, tools and technical decisions behind the CERN Cloud Infrastructure.

avatar for Belmiro Moreira

Belmiro Moreira

Cloud Engineer, CERN
Belmiro Moreira is an enthusiastic software engineer passionate about the challenges and complexities of architecting and deploying Cloud Infrastructures in very large-scale environments. He works at CERN and during the last two years his main role was to design, develop and build... Read More →

Tuesday October 27, 2015 5:30pm - 6:10pm JST
Wednesday, October 28

11:15am JST

Can We Really Make an Operating System for Data Center as a Warehouse-scale Computer? The T-RON Operating System With OpenStack
We introduce our T-RON Operating System (T-ROS) project, which is the core platform of the SK Telecom’s Software-Defined Data Center (SDDC). The T-ROS project aims at realizing an operating system-like system software suite to manage and orchestrate resources in the virtualized datacenter coordinated with seamless physical infrastructure management. The T-ROS will work as an operating system of the Warehouse-Scale Computer (WSC), a data center as a big computer.

We discuss how OpenStack made the birth and the evolution of T-ROS possible and the values of OpenStack to T-ROS and the SK Telecom’s SDDC. We discuss how OpenStack is utilized inside T-ROS as the core resource orchestration engine and the abstraction layer for datacenter infrastructure. We introduce key concepts and technologies of T-ROS, look into some use cases and insights behind those concepts and technologies. The important features of T-ROS are:

  • Data center asset management

  • Automated bare-metal provisioning integrated with the asset management

  • Seamless monitoring of physical nodes and virtual machines

  • Virtual resource management with the OpenStack Horizon

  • Distributed, pluggable Event Bus and on-demand event module management with the Event Manager

  • Ceilometer-based metering and billing framework and metering and billing manager,

  • Advanced VM scheduling and management with the application-aware hierarchical scheduler

Finally, we also share and discuss our vision and future plan of T-ROS with OpenStack.

avatar for Jaesuk Ahn

Jaesuk Ahn

Dr., SKT
I have been involved in OpenStack from July 2010 when it is publically announced. At that time, I was leading a newly created research project team at KT R&D department focusing on Cloud. Since then, my life has been all about OpenStack. I established OpenStack Korea Community early... Read More →

Taeil Choi

Senior Research Scientist
I am a system engineer at SK Telecom, currently working on the SK telecom's next-generation Software-defind Data Center.   My research interest spans system engineering, deployment automation and cloud computing. Prior to SKT, I was a system engineer at KT Cloudware, where I was... Read More →
avatar for Jincheol Kim

Jincheol Kim

Ph. D. Principal Research Scientist, SK Telecom
Dr. Jincheol Kim is a principal research scientist working for SK Telecom currently. He is currently leading the T-RON Operating System (T-ROS) project as the chief software architect and founded the T-ROS project from early stages of the SKT’s Software-Defined Data Center project... Read More →

Wednesday October 28, 2015 11:15am - 11:55am JST

12:05pm JST

PayPal's Cloud Journey From Folsom to Kilo -- What We Learned in the Upgrade
At Paypal, we spent more than one year to upgrade all our Openstack deployments from folsom/grizzly to havana. In this talk, we would like to share the lessons we learned the hard way from the previous upgrade, and how we plan to change the process starting from kilo upgrade. Our final target is to bring Paypal's openstack cloud up to date with upstream and to continuously integrate with upstream.

The previous upgrade process was labor intensive, painful and risky, here is why:

  1. Paypal has one of the largest Openstack cloud in enterprise serving production workloads. The upgrade should not interrupt any of the business applications running in the cloud.

  2. We had mixed environment with folsom/grizzly and mixed networking platforms with nova-network/neutron. There was no standard way to upgrade from nova-network to neutron without interrupting running VMs.

  3. We had a lot custom changes for Paypal’s specific use cases. We also back ported some bug fixes from upstream. This made our code base diverged from upstream. For each new openstack release, we have to manually merge, modify or drop our own changes one by one with the upstream code and run thoroughly tests again the merged code.

To meeting above requirement, we developed a special side project called ‘upgrade-test’ which has all the custom migration code and testing code. The upgrade plan was carefully designed and a few dry runs were executed before the real upgrade.

In order to catch up with upstream release, we decided to skip icehouse/juno and directly upgrade to kilo from Havana. To make sure the future upgrade will be smooth without code merging issues, we did the follow changes:

  1. Refactoring our custom changes into a different project and avoid the dilemma between backport and upgrade. For example, the nova project in Paypal’s github is a fork from upstream nova project. All Paypal specific changes go into another project called “nova_pypl”. The build script will combine the two projects and generates a single nova tarball.

  2. Without ad-hoc change to the upstream code. Any custom features should be implemented through standard customization paths: Paste middleware, API extension, scheduler filters, and class extension.

  3. Actively contribute to the community. Reporting and fixing bugs in the community and sync the changes back to our fork repository continuously.

Kilo upgrade is a milelstone for Paypal's openstack cloud, we will no longer play catching up games and always stay on latest openstack release.

To keep up to date with upstream code base, we can get more help from the community with bugs fixes and new features.

We can also actively contribute to community in the following ways:

1. Reporting bugs we find in a large enterprise deployment.
2. Fixing bugs in the upstream.
3. Actively get involved in blueprints and design sessions for new features.

avatar for Wei Tian

Wei Tian

cloud performance Lead at Paypal
15 years of enterprise software development experience, including architecture design and technical leader experience. Over 10 years of experience on virtualization and cloud infrastructure implementation and deployment. Working in Paypal's cloud engineering team since 2013. Leading... Read More →

Wednesday October 28, 2015 12:05pm - 12:45pm JST

2:00pm JST

A Cloud Architect's Toolkit : What to Expect When Taking OpenStack From Zero to Production in a Fortune-500
War stories, learnings, and caveats to watch out for, as you and your team take OpenStack from introduction stage to mission critical production deployments in your Enterprise.  This session will have 2 or 3 speakers.

  • How best to introduce OpenStack within your Organization / Company.

  • Does it matter how the PoC (proof of concept) is done, for the OpenStack value-prop?

After working very hard, you, your team and your Director / VP have just earned a mandate from the executive leadership : "Bring OpenStack into production". Yay!  Now what? You are underwriting the private cloud story for your company. And it is a humbling realization.

  1. With the ever-elusive mandate in your hand, the bulk of the challenges may lay ahead.

  2. Starting small. Creating internal customer success story and then scaling up. White glove treatment.

  3. OpenStack Cloud distro of your choice is up and running. What's next?

  4. To hire or To build skills internally? Quickly.

  5. Integrating existing / legacy sub-systems like authentication, LDAP, AD, IPAM, DNS, DHCP.

  6. What cultural challenges to expect?

  7. How best to use real data to effectively anticipate and mitigate those challenges.

  8. Leveraging the vibrant community around OpenStack to maximize the chances of your success.

  9. Path to production. OpenStack as a unified control plane with Multiple hypervisors.

  10. Do we need to strategize around migration?

  11. Internal onboarding with Operations, Lab admins and support, and with the vendor partners.

avatar for Joseph Sandoval

Joseph Sandoval

Director, Cloud Platform Engineering, Lithium Technologies, Inc
My current role is Director, Cloud Platform at Lithium Technologies. With over 20 years experience designing and deploying in the datacenter and 7 years in public cloud.  I've got involved with Openstack since 2012 and operator since 2013 as the cloud platform of choice in the... Read More →
avatar for Amit Tank

Amit Tank

Sr. Principal Cloud Architect
Amit Tank (@amitTank) is a Sr. Principal Cloud Architect with DIRECTV. He spends his time designing and building large scale cloud infrastructure, and leading projects and teams of technologists working towards putting OpenStack in production - while contributing the learnings back... Read More →

Wednesday October 28, 2015 2:00pm - 2:40pm JST

2:50pm JST

Building a Private Cloud to Efficiently Handle 40 Billion Requests / Day
What do you do when your usual setup or turnkey solution isn’t suited for your workload? 

Most of the documentation and user feedback that you can find about OpenStack is written for the use-case of running a public facing cloud serving several external customers. When you want to host a single tenant with a single application the problem is completely different, you don't want publicly exposed APIs. You want to ensure optimal resource allocation to maximize your application performance. You want to leverage the fact that you own the infrastructure layer to optimize your instance placement strategy, and to get the best latency and to avoid creating SPOFs using affinity (or anti affinity rules). 

This talk will focus on what we learned during a two years journey; from getting OpenStack up and running reliably, to investigating performance bottlenecks, to maximizing the performance of our private cloud.


Pierre Gohon

Senior SRE at tubemogul
Speicalized in networking, Pierre is always interesed in testing new things. Over the last 20 years he's been involved in creating telecom operators in France (Cegetel, Free) and web hosting companies. Currently, Pierre is the leading expert in the design and implementation of a... Read More →

Wednesday October 28, 2015 2:50pm - 3:30pm JST

3:40pm JST

How Amadeus Data Processing is Using OpenStack to Build Global Travel Applications for Millions of Users
Amadeus provides the technology that keeps the travel sector in motion – from initial search to making a booking, from pricing to ticketing, from managing reservations to managing check-in and departure processes.

To be able to provide a faster programmatic access to infrastructure for the Amadeus developer community, and for later use as the basis of production systems of future applications, Amadeus chose OpenStack as the IaaS layer.

This talk will focus on a couple of key learnings that Amadeus made during their journey to OpenStack:

  • How OpenStack challenges the status quo of traditional operations of compute, networking and storage

  • What influence the needs of traditional applications had on the choice of the underlying compute, storage and networking layer

  • What challenges were faced when combining new design patterns of 3rd generation apps with best practices of traditional IT and virtualization

  • What OpenStack projects like Heat and Neutron were key to success, and what tooling was missing that needed to be externally provided

Wednesday October 28, 2015 3:40pm - 4:20pm JST

4:40pm JST

How OpenStack is implemented at GMO Public Cloud service offering
GMO Internet started using OpenStack in 2011for their internal deployment. Through this experience, they deployed OpenStack to start their public cloud services as the following.

  • Onamae VPS (Domain Name service) in 2012

  • ConoHa (Hosting Service) in 2013- GMO Application Cloud in 2014

  • Onamae.com Cloud (IaaS Cloud Server) in 2014

  • ConoHa (Renewal for Hosting Service) in 2015

Currently, GMO Internet is using Diablo, Grizzly, Havana, Juno.

In this session, we will explain how GMO Internet deployed OpenStack into our Public Cloud service for “ConoHa” and “GMO Application Cloud”. We will show the technical background, what components of OpenStack is actually used in those services and how those deployment brought the advantage to the service.

  • Why we chose OpenStack

  • Architecture of ConoHa back end infrastructure

  • 2 different deployment of OpenStack for those 2 public cloud services

avatar for Naoto Gohko

Naoto Gohko

Naoto-Gohko, GMO Internet, Inc.
our product by OpenStack. ConoHa public cloud; https//www.conoha.jp/en/ GMO AppsCloud; https://cloud.gmo.jp/en/
avatar for Hironobu Saitoh

Hironobu Saitoh

Technical Evangelist, GMO Internet, Inc.

Wednesday October 28, 2015 4:40pm - 5:20pm JST

4:40pm JST

OpenStack Consumption Models: Three User Perspectives
Deploying, managing, and maintaining an OpenStack powered cloud are a part of daily life for every operator.  In this panel, we will take a critical view of the various options that are available for building an OpenStack cloud.  Is DIY the best path or are there trade-offs?  Should you be concerned with vendor lock-in if leveraging a distribution?  What considerations should you take into account when deciding between the two options of DIY or Distribution?  Why, or why not, would you consider using a hosted OpenStack solution?  Join users operating OpenStack clouds to hear their take on these essential questions and as they discuss the good, the bad, and the ugly with each path to a building a successful OpenStack powered organization.

In this session, panelists will discuss:

  • Lessons learned in OpenStack deployment at scale

  • Pros and Cons for each approach

  • Tools and technologies for deploying and scaling applications and infrastructure

  • PaaS, Containers, Hybrid Applications: What comes next?

Moderator: Niki Acosta, Cisco Cloud Evangelist 

Panelists include:

  • Gerd Pruessmann, Deutsche Telekom
  • Subbu Allamaraju, Ebay
  • Edgar Magana, Workday


Niki Acosta

Director of Growth, Bitmob
Niki Acosta is an OpenStack Evangelist who came to Cisco by way of the Metacloud Acquistion in late 2014. Her OpenStack journey began when she joined Rackspace in 2008, eventually becoming an OpenStack Evangelist for the Rackspace Cloud, and later, the Private Cloud business.  As... Read More →
avatar for Subbu Allamaraju

Subbu Allamaraju

Vice President, Expedia Inc.
Subbu is the Chief Engineer of cloud at eBay Inc. His team builds and operates a multi-tenant geographically distributed OpenStack based private cloud. This cloud now serves 100% of PayPal web and mid tier workloads, significant parts of eBay front end and services, and thousands... Read More →
avatar for Edgar Magana

Edgar Magana

Software architect, Salesforce
Edgar is an emerging leader who has specialized in Cloud Computing, Network Virtualization, Software-defined Networking (SDN), Network Functions Virtualization (NFV), OpenFlow and OpenStack. He has developed excellent software development skills and outstanding customer and business... Read More →

Wednesday October 28, 2015 4:40pm - 5:20pm JST

5:30pm JST

Kirin User Story: Migrating Mission Critical Applications to OpenStack Private Cloud.
NTT Data is an IT service company.
Kirin is one of the largest beverages companies in Japan.

In this presentation, we will present the user story of migrating all applications from creaky infrastructure to OpenStack private cloud including actual challenges, know-hows and future prospects.

The key concept of this project is:

 * Mission Critical: Migrate all Kirin enterprise applications to OpenStack private cloud.
 * Think Big, Start Small: Start from small number of apps, and expand rapidly.
 * Agility and elasticity: Adopt a PaaS-like automation approach, targeting 50% less development cost and 40% less operational cost.

In order to achieve all items above, we have decided to use OpenStack IaaS, ICO, which is an automation product by IBM, serverspec for testing, and Hinemos for monitoring management.

Starting from Aug 2014, the project expects 100 VM / 100 TB storage as the first-stage migration by end of 2015. We're planning to migrate 500 VM / 300 TB by end of 2016 and 2000 VM / 1 PB finally.


Motoki Kakinuma

Platform engineer
Motoki Kakinuma is a platform engineer at NTT Data, who have worked on cloud infrastructure(OpenStack) and cloud storage(OpenStack Swift). He also works on automation technologies about IT platform like puppet/Kickstart for installing, serverspec for testing and so on. He constructed... Read More →
avatar for Atsushi Koga

Atsushi Koga

IT architect at NTT DATA
Atsushi Koga is an IT architect at NTT DATA.He has experienced integration projects and operating projects of mission critical system for large telco and manufacture customers.He has wide experience of database, Java framework and SOA (ESB) technology area.Today his role is project... Read More →

Wednesday October 28, 2015 5:30pm - 6:10pm JST
Thursday, October 29

9:00am JST

NTT's Journey With OpenStack
As of today, NTT Group, the largest telecommunications company in Japan, has a variety of OpenStack deployments, from R&D testbed to production in both public and private cloud services.

The journey, however, was not straight forward.  We had to struggle with issues in many different aspects before putting OpenStack in production.  Those include non-technical issues such as business customs especially in operation as well as strong quality requirements from "carrier grade" SLAs.

In this talk, we will be sharing experiences, both good and bad, from our four years of journey with OpenStack.  You will be hearing:

- How we tested OpenStack in early days when people were still skeptic about the hype of OpenStack
- How we dealt with 150 bugs that we found and the decision we had to make
- How we built our first "Folsom-based" OpenStack for production
- How we changed the architecture design based on the lessons learned from the first deployment
- How we changed our mindset to live with OpenStack in a sustainable manner
- What we did and didn't do to OpenStack against requirements from service/operation side
- Where OpenStack fit and still doesn't fit in NTT Group's broad requirements

It will be a user story, and will hopefully be a case study for others to learn from our experiences.

avatar for Shintaro Mizuno

Shintaro Mizuno

Senior Research Engineer, Supervisor, NTT Corporat
He has been involved in OpenStack Development project in NTT R&D for four years leading system/service design and community actitvity teams. He has also engaged with business devisions to discuss and setup strategy to introduce OpenStack in production.  

Takashi Natsume

Software Engineer, NTT, Nippon Telegraph and Telephone Corporation
Takashi Natsume has been working in the IT industry for about 15 years and has been working for NTT corporation since April, 2013.He is engaged in system design of public cloud systems based on OpenStack and functional verification of OpenStack.Prior to joining NTT corporation, He... Read More →

Thursday October 29, 2015 9:00am - 9:40am JST

9:50am JST

Huawei Public Cloud : Building a Large Scale Public Cloud on OpenStack
Huawei launched its public cloud serving the China market on 7/30/2015.. This is a large-scale public cloud implementation based on OpenStack. In this session, we will discuss our experiences in cloud platform implementation and post-launch operation; and share with the audience the lessons learned and future plans. 

avatar for Anni Lai

Anni Lai

Head of Global Business Development, VP of Strategy & Business Development, Huawei
Anni leads the Operations of Huawei’s Cloud Open Source Development Team responsible for OpenStack, Containers, Open Storage, AI/Deep Learning, and other Cloud-related open source projects. Anni currently sits on both OpenStack and CNCF Boards. In addition, Anni is part of Huawei’s... Read More →

Thursday October 29, 2015 9:50am - 10:30am JST

11:00am JST

OpenStack Practice @ Intel IT: Let's Accelerate the Design Process of Smart Devices
Most of our customers are from hardware design team, so how to connect our cloud with their requirements is a big challenge for us. We have investigated some new usage models of OpenStack Cloud to serve our hardware customers. In this talk, we will introduce one of typical scenario from our customers who is focus on the mainboard/mobile prototype validation. And we also want to discuss about some potential use cases of OpenStack to make it more practical.

Here is the detail of the story:
Our customer’s daily work is testing a large number of hardware devices. The workflow is using one PC to take the role of controller node for connecting the several test instruments and one DUT ( Device Under Test ) via USB cables. Now IT helps them to migrate the validation platform to the OpenStack. Replacing amounts of PCs with VMs and leverage the USB over network solution & terminal client to meet the requirement, which test instruments & DUTS need to be automatically connect with the VMs after devices reboot/soft-interrupt. And not only that, it is independent of VM guest OS or console connection. User could easily to attach USB devices to target VM like operate the volume attachment. Based on this retrofit, we not only uncoupled software test platform with physical devices, but also place the physical devices over the tridimensional space, such as high-density rack mounted model. Benefit is quickly deployment the test environment on demand and saving the Lab space & controller nodes cost.


litrin Jiang

developer engineer
Developer @ Intel

Chen Meng

System engineer
System Engineer @ Intel
avatar for Eli Qiao

Eli Qiao

OpenStack Senior Development Engineer
Qiao, LiYong(Eli)is an OpenStack engineer. He was an virtulazition(KVM/libvirt)  Engineer before, and worked for IBM, now is working for Intel Corp. He Joined OpenStack community since Juno release and continue contributing to Nova project till now. He helped to doing nova-api enhancement... Read More →

Feng Shaohe

software engineer
Shaohe  Feng, a senior software engineer at Intel, has been working on OpenStack since the Kilo release, with contributions to Nova for object versioning, Nova API enhancements, and V3 API. He also actively contributes to Heat and Magnum. Prior to Intel, he worked at IBM on virtualization... Read More →

lin tan

developer engineer
developer engineer @ Intel

Thursday October 29, 2015 11:00am - 11:40am JST

11:50am JST

Hadoop on OpenStack: Scaling Hadoop-SwiftFS for Big Data
Elastically scalable big data clusters that can respond to varying workload demands, while efficiently utilizing and sharing cloud resources, is a reality that is attainable with Hadoop on OpenStack.  To achieve that reality requires seperating cluster compute from cluster storage in order to enable scaling compute independently of data.  In this session we discuss how OpenStack Swift can serve as the basis for an elastically scalable Hadoop cluster on OpenStack and detail the challenges faced when using Swift as the primary data store for big data.  We describe the cluster storage design and enhancements to the Hadoop Swift file system implementation that are necessary to achieve performance at big data scale.

Successful approaches to a number of the challenges are presented:

  • Storage architecture design addressing object, block, and transient storage

  • Hadoop SwiftFS enhancements to handle tens of thousands to millions of objects

  • Vendor specific support for Swift API implementations (CEPH)

  • Tool ecosystem interoperability

avatar for Andrew Leamon

Andrew Leamon

Director, Engineering Analysis, Comcast
Drew Leamon started his career at Microsoft while studying Computer Science at Princeton University.   In his studies, he delved into Computer Graphics, Artificial Intelligence and Computational Neurobiology.  At Microsoft, he collaborated with Microsoft Research on one of the first... Read More →
avatar for Chris Power

Chris Power

Principal Engineer, Comcast
At Comcast I work on many things cloud native including cloud architecture, operational visibility and data platforms.

Thursday October 29, 2015 11:50am - 12:30pm JST

1:50pm JST

Challenges in Planning, Building, and Operating Large Scale Infrastructure
Every company’s IT dream is to build the most efficient and reliable datacenter to run their applications.  In this panel discussion, hear from some of the most successful companies using OpenStack today.  What challenges do OpenStack operators still face?  How can operators and developers collaborate better?  What toolsets would IT managers like to have to get a better ROI on their cluster?  All of these topics and more will be covered in this free-flowing Q&A session.

avatar for James Downs

James Downs

Cloud Strategy and Architecture, WalmartLabs
James brought OpenStack and other cloud technologies to WalmartLabs. He championed the adoption of Openstack within Walmart E-Commerce, leading efforts behind initial installs and production pilots. Starting with Essex, he guided the cloud program from an engineering project to an... Read More →
avatar for Anant Kumar

Anant Kumar

Product Manager, PayPal
Anant is currently responsible for the Paypal Cloud. Spent over a decade working in Cisco Engineering, another 2 running a technology driven Fashion eCommerce business. Have since worked on wireless Location at Arba Networks and currently responsible for driving requirements for... Read More →
avatar for Edgar Magana

Edgar Magana

Software architect, Salesforce
Edgar is an emerging leader who has specialized in Cloud Computing, Network Virtualization, Software-defined Networking (SDN), Network Functions Virtualization (NFV), OpenFlow and OpenStack. He has developed excellent software development skills and outstanding customer and business... Read More →
avatar for Joseph Sandoval

Joseph Sandoval

Director, Cloud Platform Engineering, Lithium Technologies, Inc
My current role is Director, Cloud Platform at Lithium Technologies. With over 20 years experience designing and deploying in the datacenter and 7 years in public cloud.  I've got involved with Openstack since 2012 and operator since 2013 as the cloud platform of choice in the... Read More →
avatar for Sumeet Singh

Sumeet Singh

CEO at AppFormix, AppFormix Inc.
Sumeet Singh is the Founder & CEO of AppFormix. AppFormix optimizes enterprise clouds by providing sophisticated analytics and control of how virtualized infrastructure interacts with applications. Previously, Sumeet was the Principal lead for Hybrid Cloud Networking at Microsoft... Read More →

Thursday October 29, 2015 1:50pm - 2:30pm JST

2:40pm JST

Ancestry's OpenStack, Docker and Kubernetes Journey
avatar for Jordan Neilsen

Jordan Neilsen

Cloud Architect, Ancestry.com
avatar for Al Walter

Al Walter

Infrastructure Architect, Ancestry

Thursday October 29, 2015 2:40pm - 3:20pm JST

3:30pm JST

Motoring Ahead: OpenStack in the BMW Group's Enterprise Environment
BMW has set high standards for innovation and efficiency in the datacenter. We had rolled out optimized automation and virtualization environments by 2011 that were best in class in the auto industry. We saw the opportunity that cloud provided and initially rolled out our own implementation. We then saw the power of a community-supported environment like OpenStack in 2014, and in the space of one year we have deployed a self-service private cloud environment with 350 cores and 40TB of storage.

In this session, we will talk about the strategies employed to roll out the OpenStack-based service and the experiences with 4 departments who have deployed some of their applications. We’ll talk about some of the things we did to make it easy for the developers (like a Jenkins continuous-build environment), and the challenges that still lie ahead especially with on-boarding of legacy applications. Attendees may also benefit from our experience with managing user expectations and training them appropriately for a self-service environment.

avatar for Andreas Pöschl

Andreas Pöschl

Head of Edge Computing, Container Runtime, BMW AG

Thursday October 29, 2015 3:30pm - 4:10pm JST

4:30pm JST

Running LBaaS at Scale for Multiple Virtual Private Clouds in PayPal
avatar for Prabhakhar Kaliyamurthy

Prabhakhar Kaliyamurthy

Team Lead, Cloud Platform Engineering
avatar for Maharajan Nachiappan

Maharajan Nachiappan

Sr. Cloud Engineer, PayPal
Sr. Cloud Engineer in PayPal, Managing LBaaS.
avatar for Anand Palanisamy

Anand Palanisamy

Architect, PayPal
Leading PayPal's Cloud Engineering team.PayPal is running one of the largest private Cloud in the world. Also, we run SDN  (with overlay) in prodcution along with OpenStack LBaaS.  Our team provides the standardized scalable Cloud Compute, Storage, Network and Identitu APIs for... Read More →
avatar for Bharath Venkatakrishnan

Bharath Venkatakrishnan

Cloud Software Engineer LBaaS, PayPal Inc
Bharath is a member of PayPal Cloud team and focusses mostly on LBaaS component within it. His main focus has been on adopting/enhancing LBaaS within PayPal cloud. Prior to that, Bharath worked in Oracle with specific focus on Database Manageability, performance tuning etc.

Thursday October 29, 2015 4:30pm - 5:10pm JST

5:20pm JST

Operations and Management of a Hyper-Converged Multi-Tenant Platform
avatar for Chris Ferraro

Chris Ferraro

Cloud Engineering – Sr Engineer, FICO
avatar for Nick Gerasimatos

Nick Gerasimatos

Director of Cloud Services Engineering, FICO
avatar for Asvin Selvaraj

Asvin Selvaraj

Cloud Engineering – Lead Engineer, FICO

Thursday October 29, 2015 5:20pm - 6:00pm JST

Filter sessions
Apply filters to sessions.