openstack-logo-small transparent 2Voting is now open for the OpenStack Summit fall 2013 in Hong Kong. Cloudscalers have submitted a variety of presentations, from Randy Bias’ popular “State of the Stack” overview of the project to technical deep dives on security, block storage, networking, SELinux and more.

Below, we’ve summarized each talk and provided a link to its page on the voting site. (As always, you must be a member of the OpenStack Foundation to vote.) If you like what you see, vote it up.

We’ll see you in Hong Kong!

** **

State of the Stack

Randy Bias

This 2nd major State of the Stack address is a complete refresh of the spring 2013 edition. We’ll examine the progress from Grizzly to Havana and delve into new areas like refstack, tripleO, baremetal/Ironic, the move from “projects” to “programs”, and AWS compatibility. In addition, we’ll have updated statistics on project momentum and a deep dive on OpenStack Orchestrate (Heat), which has the opportunity to change the game for OpenStack in the greater private cloud game. We’ll also discuss the “what is ‘core’” debate and examine the idea that OpenStack is a kernel, not a complete cloud OS.

** **

Panel Discussion With Authors of the OpenStack Security Guide

Panel, incl. Eric Windisch

In June 2013, the OpenStack Security Group assembled a small team of security experts to write the OpenStack Security Guide (http://docs.openstack.org/sec/) over the course of one week.  This panel brings many of the authors together for a discussion of the book and a retrospective view of the book sprint process.

The session will start with some brief remarks from the panel members.  This will include a brief introduction, how each person contributed to the project and their personal views from the book sprint week.  Then we will open discussion to take questions from the audience on any relevant topic such as the book sprint process, OpenStack security, tips on writing another book, how to keep the book alive after the sprint, how people can contribute to the book today, suggested additions for the book, etc. (Abstract by Bryan Payne, Nebula)

** **

Protect Your Bottom-line: Protecting Tenant Data

Eric Windisch

As a cloud operator, you’ve been entrusted with the data of your users. These users might be internal consumers as in the case of private cloud, or external consumers. Either way, compromise of your systems might lead to the disclosure of information belonging to your tenants, their metadata, data at rest, or data in transit. Such exposures put your cloud operation at risk, endangering profits or budget, and may open the door the legal consequences. In this talk, we’ll identify and suggest remediations of these concerns.

** **

Practical SELinux for OpenStack

Eric Windisch

OpenStack doesn’t ship with OS hardening out of the box, but it is available through SELinux policies.  These policies have been written, exist, and are easy to deploy. We’ll show you where to find these policies and how to apply them for a more secure, hardened OpenStack.

** **

What Block Storage Looks like in the Brave New World of Elastic Clouds

Joseph Glanville

Delivering persistent block storage in OpenStack focuses on integrating traditional SAN/NAS solutions, using Distributed File Systems (DFS), or using Direct Attached Storage (DAS). We will look at the implicit and explicit contracts that users and operators get from the different approaches and look at a variety of failure conditions. Finally, we will explain why using Cinder as your storage is better than using Cinder as an API to traditional storage.

** **

Taking OpenStack Networking to the Next Level with Network Virtualization

Joseph Glanville and Ajay Hampapur (Juniper Networks)

Neutron has opened the door to enable Network as a Service within OpenStack.  Come learn and see how Cloudscaling and Juniper partnered to deliver a production ready, performant and scalable network virtualization solution optimized for Openstack.  This technical session will cover the web app, enterprise and service provider use cases that the solution addresses plus how the Nova APIs are extended to support the new capabilities.

** **

High-Performance Networking: Getting The Most From Your Gear

Dan Sneddon

Not sure if you are getting the best performance from your network? Come learn why VLANs are performance-killers, why you should avoid Spanning-Tree, and how to squeeze more performance out of your network gear and simplify your configuration at the same time.

No matter which networking model you are using, there are some universal methodologies that can help you get the best performance from your network. The techniques covered in this presentation apply to all Neutron networking models, as well as network virtualization plugins.

  • Efficiency vs. simplicity

  • Design fundamentals

  • Multi-pathing

  • Scalability

  • Measuring performance

** **

Make the Right Infrastructure Decisions When Building an Elastic Cloud: A Guide

Seth Fox

There’s a wide range of optimization choices when architecting the infrastructure for your OpenStack cloud. We will cover the proper use of the OpenStack scheduler, combined with targeted hardware and node provisioning based on meta data. What’s the ideal power and cooling scheme? What about rack density? Do you embrace Open Compute? What about long-term operational efficiency considerations? What about standardization of CPU, disk/SSD, and RAM components?

In this session, we’ll go deep on rack servers vs. blades vs. sleds. We’ll talk about utility vs. m1.standard vs. hi1 SSD-based nodes for IO instances. Participants will come away with a clear idea of the choices and consequences of infrastructure architecture so they set up their OpenStack cloud right the first time.

** **

Paying For Your Cloud: Using Ceilometer to Extract Billing Data

Eric Windisch

While OpenStack may be free software, your cloud isn’t. Your cloud has operational costs that must be accounted, either for internal budgeting purposes, or for customer billing. Using Ceilometer, it is now possible to extract utilization reports from OpenStack. We’ll show you the basics of using Ceilometer and jump start you toward integrating Ceilometer into new or existing resourcing & billing infrastructure.

** **

Deploying Swift at Scale:  Real World Case Study

Azmir Mohamed

A case study of how to build and deploy a multi- petabyte OpenStack Swift storage cloud.  Learn the practical challenges, technical decisions and results of large scale Swift deployments.  Learn best practices that enable a cloud storage system that can be grown in place plus allows enterprises and service providers to offer a competitive offering for customers.

** **

21st Century Application Development Using OpenStack

Azmir Mohamed

Agile development and OpenStack go together like chocolate and peanut butter. But what are the practical considerations when building an OpenStack deployment to support agile dev? What are the real-world issues with multi-cloud management, PaaS, bursting and repatriation?

In this session, we’ll wash away the marketing gloss and look for answers to tough questions about the myths of cloud bursting, the practical implementations of using OpenStack Heat, and what you need to know when planning to leverage public clouds in a hybrid configuration.

** **

AWS Compatibility - Farce, Fiction, or Fact?

Randy Bias

What is the deal with “architectural compatibility” anyway? Can’t I just use DevOps, PaaS, RightScale, or another solution to move my applications between clouds? Why do I need to care about AWS APIs in OpenStack, anyway? And seriously, who cares? In this session, we’ll examine each of these issues, discussing the implicit contract between APIs and infrastructure, talking about real-world use cases and examining a variety of the many OpenStack reference architectures and how they compare against the AWS/GCE reference architecture. Hecklers, malcontents, and sympathizers are all welcome.

** **

AWS Repatriation – Bring Your Apps Back

Randy Bias

Running your own infrastructure can be as little as half the cost of running on AWS once you are at scale. OpenStack-based cloud systems can provide the same or similar economies of scale if you leverage the lessons of AWS and GCE when building your cloud. This talk will discuss the economic factors in designing a cost-efficient AWS + OpenStack hybrid cloud. We’ll look at the issues involved in repatriating existing applications, and we’ll show a couple of real-world demonstration of tools that can assist in the repatriation process. Repatriation isn’t quite as simple as hitting the “Easy” button, but if you plan your deployment correctly, you can make it work, both technically and economically.