Open Infrastructure Events

#OpenInfraSummit Track: Private & Hybrid Cloud

We are featuring one of the seven Summit tracks—Private & Hybrid Cloud. Get your Summit tickets for free and add these sessions to your Summit calendar!

October 14, 2020
image

The Open Infrastructure Summit, held virtually for the first time, takes place October 19-23 and includes more than 100 sessions around infrastructure use cases like cloud computing, edge computing, hardware enablement, private & hybrid cloud and security. Thousands of attendees are expected to participate, representing 30+ open source communities and more than 110 countries.

Today, we are featuring one of the seven Summit tracks—Private & Hybrid Cloud. Get your Summit tickets for free and don’t forget to add these sessions to your Summit calendar!

10 years of OpenStack at CERN (From 0 to 300k cores)

  • Presented by Belmiro Moreira from CERN
  • What can you expect to learn?
    • In this presentation, we will describe the CERN Cloud Infrastructure from its early prototypes to today.
    • We will dive into the history, architecture, tools and technical decisions behind the CERN Cloud Infrastructure over the years.
  • Add this session to your Summit calendar!

Achieve a single OpenStack deployment in multiple datacenter

  • Presented by Victor Coutellier from Societe Generale
  • What can you expect to learn?
    • Learn why we choose to have only one Openstack deployment spanned across all the datacenter of a cloud region, which issues we encountered and how we implemented a solution, either by using built-in features of the product or by developing new one for the community.
  • Add this session to your Summit calendar!

Auto-scaling in OpenStack without Telemetry

  • Presented by Dat Vu Tuan & Kien Nguyen-Tuan from Viettel Network
  • What can you expect to learn?
    • Fundamental knowledge of what auto-scaling is and how auto-scaling works
    • Why using OpenStack Telemetry is not a production-ready for auto-scaling at the moment
    • The auto-scaling architecture consisting of Prometheus, Openstack and Faythe, an opensource software we built ourselves, is doing an excellent job, saving us tons of human labor
    • The superior points of the new stack and how we benefited from it

Back to the original mission: an open cloud operating system

  • Presented by Christian Berendt from Betacloud Solutions GmbH and Kurt Garloff
  • What can you expect to learn?
    • Clear separation between an “cloud operating system” and a “cloud software”
    • Overview of the status of existing solutions with focus on OpenStack
    • Deep understanding of the problems (technical and non technical) of existing solutions, e.g.:
      • missing or not matured toolchains
      • missing standardization
      • missing best practices
      • missing good defaults for the production
      • missing free knowledge
      • missing transparency
      • missing and inferior security
      • insufficient testing and validation
      • too few working updates
      • lack of durability and long-term sustainability
      • Overview of possible solutions (technical and non technical) to the problems identified
  • Add this session to your Summit calendar!

Building the European Weather Cloud with Open source software : An end to end journey.

  • Presented by Vasileios Baousis from ECMWF
  • What can you expect to learn?
    • We will present the approach we followed to deploy the European Weather Cloud infrastructure, the decisions made, the challenges and opportunities and lessons learned from this journey, regarding the Openstack deployment (version, operating system, networking, configuration, upgrade approaches/challenges, and Ceph deployment (version, networking, configuration ), OVS vs OVN, GPUs implementation and integration with our existing systems and services, accessing other internal systems like our ~250PB archive of Meteorological data.
  • Add this session to your Summit calendar!

Casting light on (Bare) Metal: Meet the Ironic Prometheus Exporter!

  • Presented by Iury Gregory Melo Ferreira from Red Hat
  • What can you expect to learn?
    • Baseboard Management Controllers (BMCs) can provide useful monitoring data about physical nodes. This information is beneficial to operators, allowing them to detect issues and take actions when something is wrong (e.g., High temperatures can cause unexpected node shutdowns and you can define different alerts). With the Train release a new way to use this data was born, please welcome the Ironic Prometheus Exporter (IPE)!
    • Ironic builds upon the principles of open source and open standards to enable support for a diverse ecosystem of hardware from a variety of vendors. By using the IPE, you will have multiple metrics for each physical node (e.g., temperature, power) available on Prometheus, as long as the vendor supports IPMI or Redfish protocol.
    • During the session, we will show how to configure IPE, share advantages & limitations, show a demo of how to monitor different physical nodes, and what operators can achieve when using it and close with our plans for further improvements.
  • Add this session to your Summit calendar!

Ceilometer, CloudKitty and Gnocchi: a dynamic and agnostic cloud monitoring and billing stack.

  • Presented by Rafael Weingartner
  • What can you expect to learn?
    • Attendees should expect to get insights on how to address the monitoring and billing needs of their organization, by leveraging OpenSource components such as Gnochi, Ceilometer, and CloudKitty. The talk will present the idea behind a dynamic monitoring system, and how it fits nicely in Cloud companies needs. Moreover, we will showcase real-life use cases of the proposed stack, and how it was leveraged to integrate different system/services into the OpenStack billing solution that was already in place.
  • Add this session to your Summit calendar!

China Mobile software and hardware integrated portable cloud platform test system

Discover OpenStack’s nerve with oslo.metrics: Have a robust private cloud on a large scale

  • Presented by Motomu Utsumi & Reedip Banerjee from LINE Corporation
  • What can you expect to learn?
    • WHow we are collecting metrics from oslo libraries
    • What kind of metrics we are getting from oslo libraries
    • How oslo.metrics can support monitoring and parameter tuning
    • What characteristics of OpenStack’s RPC we observed
  • Add this session to your Summit calendar!

Effective migration from Newton to Queens with minimal downtime

  • Presented by Attila Szlovencsák from GE Digital, Viktor Schlaffer from Nokia Networks
  • What can you expect to learn?
    • Cloud to cloud migration is complex (even if done between two OpenStack platforms)
    • The biggest challenges during migration (and a proposed solution)
    • Scale integration: moving instances between Scalr environments/farms
    • Methods to copy data effectively between regions.
  • Add this session to your Summit calendar!

Elastic Secure Infrastructure (ESI): I Learned to Share my Hardware; You Can Too!

  • Presented by Tzu-Mainn Chen from Red Hat
  • What can you expect to learn?
    • Attendees will learn about the ESI project, and the planning decisions that led to our initial phase of work. Detailed explanations of that work will allow attendees to understand how features will be used for the MOC, and how it can be adapted by other projects as well. Finally, a demo will showcase how all our efforts are coming together to form a hardware leasing system.
  • Add this session to your Summit calendar!

Hassle-free migration from OVS to OVN

  • Presented by Frode Nordahl from Canonical
  • What can you expect to learn?
    • Quick introduction of OVN
    • Planning a migration and steps required prior to a migration
    • Steps required during a migration
    • Demonstration of how the process is automated for OpenStack deployments deployed with OpenStack Charms
  • Add this session to your Summit calendar!

Hybrid Cloud Ecology of Tencent, Reconstructing the future digital infrastructure

  • Presented by Ruan He from Tencent.
  • What can you expect to learn?
    • Tencent has been deploying a hybrid cloud architecture within the company since 2012 and is a pioneer in China’s hybrid cloud construction. Chief Architect Dr. Ruan He will share Tencent’s best  practice experience in hybrid cloud, sharing one-stop hybrid cloud  solutions in terms of cloud native architecture, platform capabilities, and service capabilities, helping traditional industries build more  flexible enterprise IT architectures
  • Add this session to your Summit calendar!

I Don’t Think This Means What You Think It Means: Red Herrings in OpenStack

  • Presented by Florian Haas from City Network International AB
  • What can you expect to learn?
    • OpenStack’s complexity comes with operational challenges. And in situations where OpenStack misbehaves, it is frequently non-trivial to find the actual cause of an issue. This talk includes several examples of red herrings in OpenStack, and suggestions for spotting and avoiding them.
    • This is useful for both OpenStack cloud operators and users of OpenStack clouds.
  • Add this session to your Summit calendar!

Introduction of FaaS (Function as a Service) in StarlingX

  • Presented by Sharath Kumar K & Poornima Nagaraju from Intel.
  • What can you expect to learn?
    • Starlingx architecture and implementation, how FaaS as service will help StarlingX end users to run the code in most effective and cost-efficient manner.

Lessons Learned from a Large Scale OpenStack Deployment with Tripleo

  • Presented by Pradipta Sahoo & Sai Sindhur Malleni from Red Hat.
  • What can you expect to learn?
    • In the session, we will talk about our journey to over 700+ overcloud nodes using Tripleo with OpenStack Ussuri, lessons learned, issues identified and fixed. We will also talk about control-plane performance with real-world use cases

Live migration from VMware to OpenStack

  • Presented by Jack Lee, Brin Zhang from Inspur.
  • What can you expect to learn?
    • How does VMware virtualization use agentless technology and VMware data changed block tracking technology (CBT).
    • Data compatibility issues between OpenStack and VMware.
  • Add this session to your Summit calendar!

OpenStack Cluster Installer: the Debian way to deploy OpenStack

  • Presented by Thomas Goirand from Infomaniak
  • What can you expect to learn?
    • Atttendees will learn about this new deployment solution, how to deploy and use it.
  • Add this session to your Summit calendar!

Performance optimization of Neutron server and Agent for large-scale scenario

  • Presented by Haizhong Qin from Inspur, Yongfeng Du from Intel.
  • What can you expect to learn?
    • Analysis and optimization of the process for neutron control plane including dbcp agent.
    • Analysis and optimization of the process that the ovs agent deal with the change of members in security groups.
  • Add this session to your Summit calendar!

Practice and Thinking of Telecom-cloud Management Platform based on Cloud-native

Private cloud price-performance analysis

  • Presented by Tytus Kurek from Canonical.
  • What can you expect to learn?
    • Attendees will learn about the economics of private clouds vs public clouds and why organisations decide to run their workloads in private clouds. They will study what criteria to take into account when designing a private cloud infrastructure to ensure TCO reduction. They will see how the TCO per VM varies between leading public cloud and private cloud platforms.
  • Add this session to your Summit calendar!

Scaling Bare Metal Provisioning with Nova and Ironic at CERN

  • Presented by Arne Wiebalck & Belmiro Moreira from CERN
  • What can you expect to learn?
    • In this talk you will learn about the issues we encountered and the solutions we deployed to scale our bare metal deployment with Nova and Ironic to several thousand nodes.
  • Add this session to your Summit calendar!

SLO’s for Openstack – the API for Engineering Teams

  • Presented by Kit Merker from Nobl9, Joseph Sandoval from Adobe Systems
  • What can you expect to learn?
    • How to deliver reliable features faster without degrading the customer experience with private cloud. Service Level Objectives (SLOs) are customer-centric goals that define expectations between the stakeholders of your service. SLOs are an essential tool for any Site Reliability Engineering (SRE) team to achieve sustainable customer happiness when running Openstack. After the session you will know how to implement and build a culture around it with your organization .
  • Add this session to your Summit calendar!

Tackling operational and capacity utilization challenges at a large scale enterprise deployment

  • Presented by Bogdan Katyński, Silvano Buback, Imtiaz Chowdhury from Workday
  • What can you expect to learn?
    • Attendees of this presentation can learn:
      • Overview of capacity management, planning and operational challenges an enterprise cloud operator faces with large OpenStack deployments.
      • Concepts like memory fragmentation and blast radius. How to measure them and how they impact server capacity utilization.
      • How Workday manages capacity with multi-cluster deployment across data centers
      • Difference between scheduler filters and weighers.
      • How to simulate/visualize different weighers configurations. We are going to show the tools developed by Workday.
      • Scheduler limitations with very high concurrency deployments and how to overcome.
      • How to write custom weighers and how to deploy them without changing OpenStack original packages.
      • How Workday developed custom weighers to maximize capacity usage while minimizing blast radius and service downtime.
  • Add this session to your Summit calendar!

The amphorae of application delivery & security with Octavia

  • Presented by Saurabh Sureka, Richu Channakeshava, Hunter Thompson from A10 Networks
  • What can you expect to learn?
    • Introduction to Octvaia LBaaS
    • Challenges, best practices and path to migrate to Octavia LBaaS (from Neutron)
    • Enable consistent framework for application delivery and security for hybrid cloud
  • Add this session to your Summit calendar!

The Magic of the Libvirt Driver – What to Do When You Invariably Hit “No Valid Host”

  • Presented by Stephen Finucane from Red Hat
  • What can you expect to learn?
    • Upon conclusion, attendees should have a better understanding of how some of the more obscure race conditions and errors in nova can be triggered and how you can work around them if you hit them. This will be illustrated through a relatively deep dive into nova’s architecture.
  • Add this session to your Summit calendar!

Don’t miss these Summit sessions and get your Summit ticket for free!

Still on the fence?

View the full Summit schedule and check out some most anticipated Summit sessions that you might love!

Participate:

Follow the #OpenInfraSummit hashtag on Twitter, Facebook, LinkedIn and make sure to subscribe to the OpenStack Foundation (OSF) YouTube channel to get exclusive behind-the-scenes content on how the Summit is being organized!

Participate in the conversation on

Twitter: twitter.com/OpenStack

Facebook: facebook.com/OpenStack

YouTube: youtube.com/user/OpenStackFoundation

WeChat ID: OpenStack

Tags: ,