With Havana Release, OpenStack Adds Enterprise Features
October 17th, 2013 By: Jason Verge
The OpenStack Foundation today announced the release of Havana, the eighth version of the open source cloud platform. Havana provides a step forward in app-driven capabilities, an improved operational experience and additional enterprise-grade features. The big two projects that are now fully integrated are Metering & Monitoring (Ceilometer) and Orchestration (Heat). These were both incubated in the previous release, dubbed Grizzly.
Metering and Monitoring (Cielometer) provides a central collection of metering and monitoring data. An example given is collecting usage information for billing systems to determine which workloads are heavy customers.
“The metering and monitoring is critical for anyone who is running a cloud,” said Jonathan Bryce, Executive Director of the OpenStack Foundation.”It gives you visibility into storage, networking, and compute, and is able to aggregate all of that together to a central point. It also allows you to do basic alerting. This is useful from an administration perspective, but one of the big uses of this is with the orchestration project”
Orchestration (Heat) is template-based orchestration engine for OpenStack. For example, developers define application deployment patterns that specify all of the infrastructure resources an app needs.”It allows an application developer to describe all the resources he needs to run,” said Bryce. “You can automate auto-scaling.”
Object Storage Goes Global
One feature that is a little more under the radar but of significant importance is global clusters for object storage.
“The global clustering feature allows you to take your object storage environment — a cost effective system to backup – and run across several data centers,” said Bryce ”Up to now, you’ve been able to run a robust distributed storage system in a data center, or data centers in close proximity. Now you can do it all over the planet – and it acts as a single cluster. This is a pretty impressive feat. Now you can do geographic load balancing, and deliver content from a physical location closest to your users. You could use this to build a CDN, or a private CDN for a worldwide enterprise. If you have a lot of user generated content, you can distributed it out, and it behaves as a single, logical OpenStack environment. Also, If you want a really robust disaster recovery plan, you need that geographical element to it. This takes your disaster recovery to an entirely new scale. It’s a cluster, in-sync, managed across the entire infrastructure.”
The user interface has also been improved, with more functionality exposed on the dash board. There’s continued support for additional plugins as well as metered usage statistics.
Havana also adds some very enterprise-friendly features. All APIs now support SSL encryption, Virtual Private Networks and Firewall as a Service. You can now boot from volume, for live migration, and there’s added support for rolling upgrades.
“Community Has Matured”
OpenStack is heavily pushing community involvement and education.”Because the community has really matured, we’ve been able to get in the major fixes and bug fixes on time, and as promised,” said Bryce. “It’s really cool to see one of the largest collaborations work so well.”
There were 910 contributors to Havana, a more than 60 percent increase from the Grizzly release. There’s a total of 392 new features, representing a 32 percent increase in the total lines of code from April to September. Over 20,000 commits were merged, and on average, an OpenStack cloud deployed for testing occurs 700 times per day.
“The foundation was formed last September, and one of our top priorities at the summit was around education,” said Bryce. “We launched a training marketplace, as well as training courses in close to 40 countries. We’re finding that there’s an incredible demand for OpenStack skills.”