Red Hat at Open Infrastructure Summit Shanghai 2019
Going to Open Infrastructure Summit in Shanghai? You’ll have plenty of ways to connect with Red Hat.
VISIT US AT BOOTH #EX6
Stop by to:
- Learn about our products and services.
- Watch Red Hat product demos from experts.
There will also be many Red Hat partners with booths throughout the exhibit hall. Stop by to speak with them about their OpenStack solutions with Red Hat.
Modern infrastructure for your hybrid cloud
This year at Open Infrastructure Summit, Red Hat’s team of subject matter experts will be on-site and ready to talk about our product portfolio and contributions to various projects within various breakout sessions, hands on labs, featured demos, and videos.
Topics in our speaking & hands on lab sessions will range from Machine Learning, Kubernetes, security, and of course - OpenStack including the latest innovations to Red Hat’s OpenStack Platform.
Our featured booth demos will include topics including:
- Red Hat OpenShift on OpenStack Platform
- Infrastructure Migration Solution
- Virtual Central Office
Red Hat network functions virtualization (NFV) solution
Red Hat OpenStack Platform is the foundation upon which telco projects are built. Virtualized network functions, containerized services, and software-defined networking all rely on Red Hat’s NFV Solution, based on Red Hat OpenStack Platform.
Multi-access edge computing
Multi-access edge computing (MEC) is gaining momentum and importance in communication service provider’s digital transformation plans. There are several use cases that can be applied to the virtual Central Office model developed and demonstrated last year at OPNFV, but four have come to the forefront, namely: vCPE, SD-WAN, mobile, and headend virtualization. Join Red Hat’s Chief Technologist, Global Service Provider, Ian Hood, for a discussion of how the modern Telco is adopting open source in it’s digital transformation projects.
Additional resources on Red Hat's work in the telco industry:
Monday, November 4, 2019
Machine Learning benchmarking with OpenStack and Kubernetes
Deep Learning and Cloud Platforms are transforming the field of Machine Learning from theory to practice. However, implementation differences across frameworks and inference engines make the comparison of benchmark results difficult. SPEC and TPCC benchmarks are not accurate due to the complex interactions between implementation choices such as batch size, hyperparameters, or numerical precision. To address this complexity requires systematic benchmarking that is both representative of real-world use cases and valid across different software/hardware platforms. This talk will present the best Machine Learning benchmarking tools to use with OpenStack and Kubernetes. We will show how MLPerf and Thoth help data scientists to improve their system performance and fully benefit from their CPUs, GPUs, or FPGAs. We will share insights and lessons learned over the journey of key Machine Learning training and inference use cases selection.
Automating past human and technical challenges
Andrea Ferraresi and Raul Mahiques with Samuel Graenacher (Unaffiliated)
In a world where collaboration is core and "DevOps" term is often abused, the psychological part of collaboration is often underestimated. As much as we approach a technical issue from a strictly technical standpoint, the role of people on solving it is still the key. At our talk we share our real-world experience about a journey on the slippery ladder of breaking changes: From a silo-based company to a solid and modern Dev-Ops enabled organization. Approaching the topic from "how we learned from our own failures", describing how we should be empowering techies to deliver faster and take advantage of automation. But the most important point of our workshop will be the psychological aspect of collaboration that can safely lead also to better technical solutions. Good teamwork means choosing perfect tools for the job, from a technical AND a synergic point of view.
Quick, Solid, and Automatic - OpenStack Bare-metal Orchestration
Julia Kreger with Rico Lin (EasyStack)
Bare-metal on OpenStack is the trend, and if we can add orchestration around bare-metal will bring you automatic management around the town. But how we orchestrate over bare-metal servers? Is it just the same as orchestrating over VMs? Which services do we need to have a full running bare-metal orchestration service? And how about the performance and what workflow changing when you adding automatic functionality (like self-healing or autoscaling)? In this presentation will introduce bare-metal on OpenStack, and go through detail on how can you use Ironic and Heat to manage your baremetal servers in automatic way. With Baremetal become one common use case OpenStack have. Does OpenStack innovate fast enough for the needs? What's the community' plan/works? And I think what's important question for you, is baremetal on OpenStack a good choice? we will introduce how can we keep it quick, solid and automatic to support baremetal application infrastructure.
It all started with a little playbook
Raul Mahiques with Andres Valero (Unaffiliated) and Mario Vázquez (Unaffiliated)
During this presentation, we will show how to deploy a functional environment following GitOps principles using Ansible. Will give you a quick introduction on how to adapt this playbooks and other contents in the repository to your needs and how could you contribute any changes you want back to the community so that others will benefit as you have benefited. On top of that a simple self-service portal will be created in CloudForms to let non-technical users interact with the Platforms that we will use in a simple manner, as well as deploy an application into OpenShift The presentation will cover different products, mainly: Ansible Tower CloudForms Openshift 4
Deep Dive the essence of Federation in multi-clouds deployment
Lance Bragstad with Xiyuan Wang (Huawei) and Vishakha Agarwal (NEC Technologies)
Wouldn't it be great if you could consume resources across multiple OpenStack deployments without managing separate identities. By using identity federation, you're reducing the attack surface of your deployment by minimizing user accounts, allowing for integration with existing identity systems, while improving experience for your users. While federated identity certainly isn't a new feature, people still struggle understanding some of the underlying concepts and use cases, making it hard for them to know if they're in a situation that would benefit from identity federation. In this session, we're going to set the technicalities aside and focus on establishing a basic understanding of what federated identity is and how it might be beneficial to you and your users.
Manage a community like conducting an orchestra - with a lot of hand waving
Have you attended an orchestral performance and wondered what it is exactly that conductors do besides waving their arms in the air? Most of the value that the conductor (community manager) brings to the orchestra (community) is carried out before the performance (behind the scenes). In this session, I will share my experiences in orchestras and open source communities big and small across different countries. How do you take care of the different types of contributors/players and bring out their best? What happens when they have differences in opinion on how to collaborate? How do you make the community welcoming and inclusive especially for people with different backgrounds and who may not speak your (programming) language? Join me on a musical journey through some of the common (and perhaps not so common) dissonances of community orchestration and their possible resolutions!
Tuesday, November 5, 2019
Encryption Workshop: Using Encryption to Secure Your Cloud
Ade Lee and Douglas Mendizábal with Dave McCowan (Cisco)
Encryption technology can be used in OpenStack to protect the confidentiality and integrity of data and software. There are services built in to Nova, Cinder, Glance, and Barbican to work together to protect your cloud and its users. This workshop provides a hands-on tour through encryption use in OpenStack. This will include volume encryption, image signing, and protecting encryption keys. Bring your laptop! At the beginning of this workshop, attendees will each be given SSH access to a VM with OpenStack software already installed. After discussing the architecture and technology, we'll hit the ground running with configuration and operation steps to get hands on experience encrypting Nova and Cinder volumes, verifying Glance images, and storing encryption keys in Barbican.
Are Your Secrets Secure?
Ade Lee and Douglas Mendizábal with Dave McCowan (Cisco) and Jimmy McArthur (OpenStack Foundation)
Whether you're following your own infosec policy or trying to meet the requirements of GDPR, ANSI, PCI DSS, HIPAA, or NIST you will want to answer the question: "Are my secrets secure?" in my OpenStack cloud. Barbican is the OpenStack service that allows operators and users to store secrets securely. It consists of an OpenStack API that provides keystone authentication, oslo policy and quotas, and back-ends in which the secret is actually stored. But secrets are only as secure as the storage back-end that is deployed behind Barbican. This talk will focus on the types of secure storage back-ends available, how they work, and the advantages and disadvantages of each back-end. We'll include discussion of HSMs, SGX and TPMs, and Vault. The security of your secrets is important. This session will give you the information you need to confidently make decisions about secret storage for your cloud.
Run Kubernetes on OpenStack and Bare Metal fast
Ramon Acedo Rodriguez
Kubernetes running on top of OpenStack as an on-prem solution, provides high levels of automation and scalability. Kubernetes takes advantage of the OpenStack APIs integrating with OpenStack services such as Cinder, Neutron or Octavia. Kuryr is a CNI plugin using Neutron and Octavia to provide networking for pods and services. It is primarily designed for Kubernetes clusters running on OpenStack virtual machines. We tested Kuryr to check how it increases the networking performance when running Kubernetes on OpenStack, when compared to using OpenShift/OVS SDN. In this talk we will update the latest integrations and architecture to run Kubernetes clusters on OpenStack and bare metal and also aspects of the performance improvements with Kuryr by showing tests results when using Kuryr as the SDN to run Kubernetes on OpenStack.
Easy multi-tenant Kubernetes RWX storage with Cloud Provider Openstack and Manila CSI
Victoria Martinez de la Cruz, Goutham Pacha Ravi and Tom Barron
Kubernetes end users have trouble deploying multiple containers that write to the same persistent volume. Kubernetes administrators lack native solutions for hard separation of shared storage across multiple tenants. The new Manila CSI plugin for the OpenStack Cloud Provider enables solutions for both problems. Kubernetes end-users can follow a simple design pattern to write safely to dynamically provisioned general purpose RWX volumes. Kubernetes administrators can deploy Manila CSI using a straight-forward set of manifests. Manila administrators can set up service for Manila CSI with a minimum of fuss. Cloud Provider OpenStack Keystone can be used for hard tenant separation without sacrificing the economies of shared scale out infrastructure and elastic self-service provisioning. Demos, yaml, and manila configuration included.
The journey to Open Infrastructure: from adoption to production
Julio Villarreal Pelegrinoa and Øystein Bedin
In the last years, the IT landscape has evolved considerably, impacting the way that business runs and deliver services. Today, the evolution continues and Open Infrastructure solutions are a key part of the Digital Transformation movement. In this talk, we will discuss real-life use cases and sample architectures that show how to combine process and software for a successful Open Infrastructure adoption. We will demonstrate how practices typically found in software development can be applied to improve the management of traditional IT infrastructure - with the added benefit of automated tracking of configuration changes and avoiding snowflake environments. In this presentation we will cover the following software stacks and software development practices: Software: OpenStack, Ceph, Ansible, Containers, Kubernetes Practices: DevOps, CI/CD + Continuous Release, Infrastructure-as-Code / Configuration-as-Code
Wednesday, November 6, 2019
Ironically Reeling in Large-scale Bare Metal Deployment Without PXE
Ilya Etingof and Julia Kreger
Provisioning and deploying bare metal hardware in large-scale infrastructure can be challenging. Old technologies, such as PXE, DHCP, and TFTP, cannot always keep up with the demands placed on them by today's emerging environments, such as edge computing. Those environments often require the use of non-dynamic configuration techniques to ensure security, integrity, reliability, traceability, and operational efficiency. This presentation describes and demonstrates the provisioning and deployment of bare metal via static network configuration over a routable, IP network. It relies on the Redfish virtual media boot feature of modern Dell EMC hardware and the latest advancements of the OpenStack Ironic project.
Demo of Oslo Tools That Make Life Easier for Operator
Come see a demonstration of a number of the tools that Oslo provides to make life easier for operators. Topics covered will include: * oslo.config drivers for keeping secrets out of on-disk configuration files * oslo.config validator for finding deprecated or non-existent options present in a configuration file * oslo.policy generator for dumping out a complete listing of a service's policy in code defaults * oslo.policy checker for testing whether a given token is allowed access by a policy rule * oslo.policy redundant rule tool for finding policies that are set to the same as the in-code default
Hotel OpenStack: A New Analogy for Explaining OpenStack
OpenStack contains many moving parts and it can be very complicated to explain how it all works to someone new to the project. Where do you start? How do you explain everything to someone with no knowledge or understanding of how cloud infrastructure works? Introducing "Hotel OpenStack", the best hotel you'll ever visit! Using "Hotel OpenStack" as an analogy for how OpenStack works, you be able to explain OpenStack and general cloud infrastructure to the average person on the street. After this presentation, you'll gain a high-level understanding of what OpenStack does and how to explain it to someone with zero experience in cloud infrastructure.
Avoiding obstacles on the river ride — from upstream to downstream CI
When I joined Red Hat as a CI specialist, there were multiple definitions of what a "CI" system is. My definition is simple: "A set of processes for continuously delivering 'verified' code supported by effective automated tooling" The tools are the means and not an end unto themselves, but a very important means that can make our life easier. At Red Hat we divide CI into Upstream, Midstream and Downstream. A simple division with many teams involved - without which you cannot sustain such a complex, constantly changing environment. "Upstream" is self-explanatory. "Midstream" is the RPM-based RDO project. Finally, the downstream Red Hat OpenStack (RHOS) distribution is the tested and supported variant of RDO. RHOS has a much wider test matrix, with ~2200 CI jobs set, based on RHEL and TripleO running ... on Jenkins. This presentation will show a real world example of adopting Upstream tooling via "Software Factory", for downstream RHOS product.
Connect with Red Hat onsite at Open Infrastructure Summit in Shanghai.
Red Hat associates will present or co-present several breakout sessions, workshops and panel discusions.
Not able to make it to Shanghai or these sessions? Don't worry, the sessions will be recorded and available on-demand (coming soon).
Stay up to date on the latest Red Hat technologies.
Check out these resources before Open Infrastructure Summit, or share them with your team after the event.
OpenStack Summit Berlin 2019 Press Releases
- Red Hat Drives Unified Foundation for Kubernetes and Virtual Machines with Red Hat OpenStack Platform 14
- Introducing Red Hat virtual central office solution: An open pathway to modern telecommunications services
OpenStack Summit 2018 Vancouver press releases
- Red Hat Delivers Cloud Foundation for Digital Transformation with Latest Version of OpenStack Platform
- Red Hat and Juniper Networks Expand Collaboration to Provide a Simplified and More Secure Path to Multicloud
- Red Hat Helps Power Tata Communications’ IZOTM Private Cloud with OpenShift and OpenStack
Red Hat blog posts related to OpenStack Summit 2018 Vancouver
- Red Hat OpenStack Platform celebrates five years of accelerating private and hybrid clouds
- Red Hat OpenStack Customer Survey 2018: containers, technical support top of mind
- OpenStack Stein feature highlights: vGPU support coming in Red Hat OpenStack Platform 15
- OpenStack Stein feature highlights: edge deployments, storage and single node deployments
- The road ahead for Red Hat OpenStack Platform
E-books, trainings, and on-demand webinars
- OpenStack curriculum: Train and test with Red Hat
- Red Hat Learning Subscription
- Red Hat OpenStack Platform evaluation e-book
- Hybrid Cloud Strategy for Dummies e-book
- Defining your cloud strategy step by step webinar series (on-demand)
- Webinar: Power the open hybrid cloud with Red Hat OpenStack Platform
- Turkcell scales cloud storage to improve reliability and stability
- Massachusetts Open Cloud supports big data analysis with open source storage
- Paddy Power Betfair delivers online bets fast and reliably
- MyRepublic adopts unified cloud with Red Hat OpenStack Platform
- LDCC prepares for future with Red Hat OpenStack Platform