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.

Related OSS Projects [clear filter]
Tuesday, November 4
 

11:15

The Perfect Match: Apache Spark Meets Swift
Apache Spark is a fast and general engine for large-scale data processing that provides efficient support for general data processing, ETL, streaming, and machine learning. It is capable of running analytic jobs up to 100X faster than Hadoop MapReduce. In this talk, we will introduce Spark and the various storage systems Spark can operate on.

In particular, we focus on the integration between Spark and OpenStack Swift, demonstrating the advanced models for executing Spark jobs directly on Swift objects.

Speakers
avatar for Michael Factor

Michael Factor

IBM Fellow, Storage and Systems, IBM, Haifa Research Lab
Dr. Michael Factor is an IBM Fellow with a focus on storage and systems and is the CTO for Cloud Storage Services.   Michael has presented on a range of storage topics at three of the last four OpenStack summits. His interest in cloud storage includes a deep focus on object storage... Read More →
avatar for Gil Vernik

Gil Vernik

cloud research staff member at IBM
Dr. Vernik Gil is as a researcher in IBM, storage clouds, security and analytics group. He received his PhD degree at Mathematics from the University of Haifa and completed his post doctoral position, in the Christian-Albrechts-Universit... Read More →
RX

Reynold Xin

co-founder of Databricks
Reynold is a co-founder and the Chief Architect of Databricks.


Tuesday November 4, 2014 11:15 - 11:55
Room 251

15:40

Converged Cloud Computing That's Secure, Fast, or Cheap: Pick Three
Containers are driving down the overhead that has been necessary for traditional virtualization. But there have been serious tradeoffs made with their adoption. Containerization's resource sharing approach is exposing more of the host system. We're choosing to compromise the isolation of executables from kernel, accepting it as an inevitable price for low overhead. In multi-tenant environments, that's a heck of a gamble.

Tightrope acts like that do not need to happen. 

ZeroVM has taken an entirely distincive approach to lightweight virtualization: strong security, from the beginning. Processes are kept on lockdown, jobs are horizontally scalable, and they execute with such fine-grained accuracy that metering by the second is no problem. End the cost-inefficiencies of metering by minutes and hours.

ZeroVM provides a secure execution environment for untrusted code, by applying the Chromium project's proven Native Client (NaCl) technology for validation and sandboxing. Moreover, when paired with OpenStack Swift, the two become a unique platform for executing arbitrary queries within a datastore, using any language of one's choosing.

We'll walk through the basics of the ZeroVM open source project, compelling use cases, and opportunities to take advantage of ZeroVM + Swift -- for solving problems at scale without compromising on either security, speed, or affordability.

Speakers
avatar for Carina C. Zona

Carina C. Zona

Community Manager, ZeroVM
Carina C. Zona is the community manager for open source software project ZeroVM. She also is an organizer & instructor for many tech women s organizations. Carina is the founder of @CallbackWomen, an initiative to connect first-time speakers with conferences. She has been known to... Read More →


Tuesday November 4, 2014 15:40 - 16:20
Room 241
 
Wednesday, November 5
 

09:00

Monasca Deep Dive: Monitoring at Scale
Monasca is a new cloud-scale, multi-tenant, high performance, fault-tolerant OpenStack Monitoring as a Service (MONaaS) platform described at, https://wiki.openstack.org/wiki/Monasca, that is currently focused on real-time streaming metrics storage/retrieval, alarming and notifications. Development of anomaly detection is in-progress. StackTach.v3, https://github.com/StackTach, is a stream-based processing system for events that is being integrated with Monasca to provide a unified metrics and events processing system.

Join Roland Hochmuth, HP Software Architect, Tong Li, IBM Senior Software Engineer, and Sandy Walsh, Rackspace Senior Software Developer, for an architectural deep-dive of the Monasca components, RESTful API (monasca-api), Agent (monasca-agent), Client (python-monascaclient), Horizon dashboard and the future plans for integrating with the StackTach streaming events-processing pipeline. We will review the technologies used, such as Apache Kafka, InfluxDB and Apache Storm, as well as third-party integrations, such as the excellent metrics dashboard Grafana, http://grafana.org/. Our work in performance analysis will be shown.

Early results of anomaly detection based on the NuPIC Cortical Learning Algorithm (CLA),http://numenta.org/nupic.html, will be presented.

Although Monasca is a relatively new project it is ready for deployment today and comes with a turn-key development environment based on Vagrant and Docker. We will demonstrate several key areas of the solution.


 

Speakers
avatar for Rob Basham

Rob Basham

Architect, Cloud Systems Software, IBM
Rob works at IBM designing systems management software with a focus on monitoring and automation. He is relatively new to OpenStack but has been working on various aspects of systems management for quite some time. Rob is an innovator and author of dozens of patents across a broad... Read More →
avatar for Roland Hochmuth

Roland Hochmuth

Tech Lead and Software Architect on Monasca, HP
I am the Tech Lead and Software Architect on Monasca, the open-source Monitoring-as-a-Service (at-scale) OpenStack project (https://wiki.openstack.org/wiki/Monasca). We focus on developing a highly performant, scalable and reliable turn-key monitoring solution that leverages the industries... Read More →
TL

Tong Li

Senior Software Engineer, IBM
Tong Li is a senior engineer at IBM. He has been a software architect, developer, deployer, operator, advocator and a consultant through his professional career. His main focuses in recent days have been cloud computing, social software and Blockchain technologies. He is very passionate... Read More →
avatar for Sandy Walsh

Sandy Walsh

Software Developer - Rackspace
Based in Halifax, Alex 'Sandy' Walsh is the owner of Dark Secret Software. He has been a senior professional developer for nearly 20 years and a Pythonista for 10 years. He is a developer on the OpenStack project with Rackspace and created the StackTach monitoring application. You... Read More →


Wednesday November 5, 2014 09:00 - 09:40
Room 241

09:50

Docker All The OpenStack Services
Managing the lifecycle of OpenStack's services is no easy task. In this session users will learn about the lifecycle management challenges faced by enterprises deploying OpenStack including:

   Using lifecycle management tools with a fast moving development model
   Ensuring service isolation
   Maximizing utilization of resources by consolidating services

Currently, lifecycle management tools are attempting to solve these problems, but the approach requires far too much logic within the management tool and is inefficient. Virtual machines are also being used by many, but involve too much overhead and lack rich metadata without complex configuration management tools.

Enter Docker, Kubernetes, and Atomic which hold the promise of a more efficient and reliable way of managing the lifecycle of OpenStack services. After exploring the challenges and shortcomings of current methods of lifecycle management of OpenStack services attendees will learn how Docker, Kubernetes, and Atomic can be used to solve these challenges and reduce deployment complexity. Users will learn:

   How performing automated Docker builds of OpenStack services on commits can increase developer    productivity
   The benefits of deploying OpenStack services in containers from Docker images using Kubernetes    templates on hosts
   Considerations when deploying OpenStack services as containers and the future of container based  deployment orchestration and management
   The Kolla StackForge Project

Join us for this talk and learn to Docker All The OpenStack Services!

 

Speakers
avatar for Brent Holden

Brent Holden

Chief Field Architect, Red Hat
Brent Holden is Red Hat's chief field architect with a speciality in cloud infrastructure, specifically OpenStack. Brent is responsible for bridging the office of the CTO to senior technology leaders throughout Red Hat's customer base. He has experience with customers of all sizes... Read More →
avatar for James Labocki

James Labocki

Technology Strategy, Red Hat
James Labocki is a technology strategist for Red Hat's Infrastructure Group. In his role James works to shape Red Hat's roadmap for technology integration of Red Hat's Virtualization, Private Cloud, and Cloud Management offerings. Previously, James assisted Public Sector organizations... Read More →


Wednesday November 5, 2014 09:50 - 10:30
Room 241

11:50

OpenDaylight and OpenStack Developer Panel

OpenStack and OpenDaylight are two open source projects with a lot in common, including the same developers who work on both simultaneously. What makes these projects work so well together and how can you get involved? Ask the developers themselves in this panel discussing policy and control -- the point of intersection for the two projects -- and the pros and cons of uniting OpenStack s cloud with OpenDaylight s controller. They ll also cover the challenges encountered while building code and the ultimate long term benefits for the networking industry. 



 




Moderators
PR

Phil Robb

Director of Networking, OpenDaylight

Speakers
avatar for David Lenrow

David Lenrow

Distinguished Architect, HP
David Lenrow was trained as a computer scientist and has spent more than 20 years driving innovation in digital technology with an emphasis on networks, storage and media.  His career spans multiple roles from individual contributor to executive across all major functional areas... Read More →
avatar for Chris Wright

Chris Wright

CTO, Red Hat
Chris Wright is vice president and chief technology officer (CTO) at Red Hat. Wright leads the Office of the CTO, which is responsible for incubating emerging technologies and developing forward-looking perspectives on innovations such as artificial intelligence, cloud computing... Read More →


Wednesday November 5, 2014 11:50 - 12:30
Room 241

13:50

Meet Mimic: An API-compatible Mock Service for OpenStack
This talk introduces Mimic, a lightweight framework that imitates a subset of OpenStack services, to speed development and improve code quality. Mimic can be faster, easier to deploy, and more configurable than DevStack with null drivers. Unlike many replay services, Mimic provides dynamic, stateful responses based on templates of expected behavior for the supported services.

Mimic was created to facilitate complex system integration tests for an application sensitive to network and upstream failures due to significant dependencies on external OpenStack services.  By speeding up feedback and enabling configurable error injection to enhance test coverage, Mimic is a valuable tool for the development of applications consuming OpenStack services. Our goal is to bring this tool to the wider community, solicit contributions, and find more cases for its use and enhancement of the OpenStack ecosystem.

Speakers
avatar for Glyph

Glyph

Rackspace
avatar for Lekha Jeevan

Lekha Jeevan

Software Developer in Test, Rackspace


Wednesday November 5, 2014 13:50 - 14:30
Room 241

14:40

Use OpenStack as a Vagrant Provider
Vagrant is a very popular tool within developer communities. It makes easy to manage virtual machines and is very useful to set up development or staging environments. Vagrant can be plugged on various VM providers using its plug-in mechanism, VirtualBox, Docker, AWS... In this talk we will present the open source project Vagrant Openstack Provider, implementing a plug-in to use Openstack as a cloud provider for Vagrant. We will present how to use it, the key features, some interesting use cases and the roadmap for future releases.

Speakers
GG

Guillaume Giamarchi

Developer / Devops
Developer and devops, mainly experienced on backend architecture and development, Guillaume currently develops enterprise products built on top of Openstack API and works on Openstack integration within an enterprise information system.
JV

Julien Vey

OpenStack Developer, Numergy
Active Technical Contributor to the OpenStack project, Julien's work is focused on the area of application lifecycle management in the Cloud. Core contributor to the Solum project, he designs and develops the next generation CI/CD for OpenStack. Active Technical Contributor to the... Read More →


Wednesday November 5, 2014 14:40 - 15:20
Room 241

15:30

Beyond Horizon: Open Source Management On the Go for OpenStack and the Rest of Your Computing Infrastructure
Horizon is a powerful UI for OpenStack management. But when working on multi-cloud/hybrid environments you end up having to log in and use several different UIs. Mist.io (https://github.com/mistio/mist.io) is an open-source, minimalistic and mobile-friendly multi-cloud server management interface. It supports launching, destroying, tagging and rebooting OpenStack VMs, and provides a touch friendly web console based on term.js to access your servers safely from any device. It can connect to a monitoring service (https://mist.io) that providers monitoring, alerting and automation. It uses Apache Libcloud to talk to the backend, the client is built on ember.js and jQuery Mobile and uses Socket.IO for real time updates.

Speakers
avatar for Chris Loukas

Chris Loukas

SRE & QA, mist.io
Linux enthusiast. Loves python, C, cannot tolerate javascript! Became QA Manager out of need for automation. In a love/hate relationship with anything automation-related.
avatar for Dimitris Moraitis

Dimitris Moraitis

CTO & co-founder, mist.io
Dimitris is a full stack hackerpreneur, passionate about solving hairy technical problems without sacrificing the UX. He loves Python, tolerates Javascript and constantly looks for ways to simplify the management of computing resources. To that end, he co-founded Mist.io where he... Read More →


Wednesday November 5, 2014 15:30 - 16:10
Room 241

17:20

Percona Server Features for OpenStack and Trove Ops


Over the last few years, Percona, in partnership with other Openstack Community members, has developed and refined several features in Percona Server to facilitate its use in cloud scenarios and, specifically, within OpenStack Trove. Since Percona Server is a drop-in replacement for MySQL, many operators (as well as end users) have benefited from these features designed to improve cloud deployments.

 

In this talk I will cover some of these new Percona Server features that OpenStack operators can use to better manage their Trove guest instances of MySQL. These include:

 

 



  • How to use the utility-user functionality to limit access to hosted data and help limit the liability of the operator while still allowing a certain level of access to manage the MySQL server.



  • How to use the enhanced options modifiers to prevent end users from changing critical MySQL system variables.



  • How to use the various options to limit the space utilized by binlog and slow query logs (max-binlog-size, max-binlog-files, max-slowlog-size, and max-slowlog-files).



  • Share research on preventing InnoDB/XtraDB from getting into an unrecoverable state due to ENOSPC and some techniques that can be used to avoid this situation.



  • Discuss the research being done to help limit runaway UNDO space.




 

 

After this presentation, OpenStack operators will have a better understanding of the capabilities available to them in their Trove offerings when using Percona Server as their MySQL database choice.

Speakers
avatar for George Lorch

George Lorch

Software Engineer, Percona, Percona
  George joined the Percona development team in April 2012. He has over 20 years of experience in software support, development, architecture, and project management. Prior to joining Percona, George focused on Windows-based enterprise application server development, network protocol... Read More →
avatar for Vipul Sabhaya

Vipul Sabhaya

Lead Software Engineer, HP
Vipul Sabhaya works in the HP Cloud Platform Services team out of Seattle as a Lead Software Developer. Vipul led the HP Cloud effort to get a Database as a Service product to Public Beta with Openstack Trove. Vipul is excited about building awesome community-based Platform Services... Read More →


Wednesday November 5, 2014 17:20 - 18:00
Room 241