The OpenStack Foundation is set to officially debut its next major milestone release, dubbed “Icehouse,” on April 17, providing a long list of cloud feature updates and enhancements.
OpenStack was originally started as an open-source project by Rackspace and NASA in July 2010 and has since evolved to become a leading cloud platform supported by many of the world’s top IT vendors. The new Icehouse comes six months after the OpenStack Havana release came out in October 2013.
As part of Icehouse, the OpenStack platform is now gaining a new project with the inclusion of the Trove database-as-a-service (DaaS) technology. Trove was originally known as Project Red Dwarf and got its start in 2011 with the support of Rackspace and Hewlett-Packard. Rackspace today has a cloud database service that is built on top of the same technology that is in Trove.
“Pretty much every application today has a database,” Jonathan Bryce, executive director of the OpenStack Foundation, told eWEEK. “Creating a standard way to provision and manage the lifecycle of those databases is a really useful and important feature to have in OpenStack.”
John Engates, CTO of Rackspace, told eWEEK that Trove helps to enable a managed database platform, instead of users needing to roll their own implementations. With Trove as part of the integrated OpenStack platform, Engates is hopeful that the project will continue to get better, faster. As a community effort, OpenStack offers the opportunity for wider contribution as well as support, which ultimately can improve the quality of software.
“I think Trove becoming an integrated project in OpenStack is important,” Engates said. “The overall benefits of OpenStack that we’ve always enjoyed will now extend in the realm of database as well.”
Overall from Engates’ perspective, the Icehouse release is really about stability and maturity. It’s a theme that OpenStack Foundation executives echo. Mark Collier, chief operating officer of the OpenStack Foundation, told eWEEK that the new Icehouse release was strongly influenced by users of the platform. He noted that user feedback in the release is a clear sign that OpenStack has reached a level of production usage and stability needed for real-world deployments.
Among the new features that have landed in OpenStack Icehouse that were specifically driven by user feedback is an improved federated identity capability.
“Now you can use a single credential and be able to log into multiple public and private clouds,” Collier said. “That’s a key step on the path toward the hybrid cloud requirement that a lot of people have today.”
OpenStack Icehouse Features a Trove of Open-Source Cloud Updates
Storage also gets a boost in the OpenStack Icehouse release with a new replication mechanism for object storage. The OpenStack Swift project is one of the core OpenStack projects and provides object storage capabilities. Swift had previously been using the “rsync” open-source application to enable replication across different storage drives in a cluster, The OpenStack Foundation’s Bryce, told eWEEK.
“There is now a native replication service called sSync, or Swift Sync, that is actually providing better performance for large environments,” Bryce said. “It taps directly into the requests coming in and out of the object storage system to add and remove objects, so it’s able to be more intelligent than rsync.”
Nova Compute
OpenStack Icehouse now includes the ability to perform a live update on an OpenStack Nova Compute cluster. Bryce noted that the Swift object storage system has had the ability to perform live upgrades since the beginning of OpenStack. Getting live update capabilities in Nova has taken some time, he added, since it is so tightly integrated with multiple facets of the entire OpenStack platform.
“With Icehouse, you’re now able to upgrade the control plane and still be able to control OpenStack Havana compute nodes and then upgrade those nodes on a rolling basis,” Bryce explained. “The upgrade on the compute node is just a restart that doesn’t have to affect the running workloads.”
While live Nova Compute updates have progressed in the OpenStack Icehouse release, another key feature known as Cells hasn’t quite moved along as much. The idea behind Cells, a feature that was initially talked about by the OpenStack Foundation in the Grizzly release in April 2013, is to enable multiple Nova Compute modules to be managed by a single Nova API.
“Cells are a fairly light layer that sits on top of an OpenStack environment to provide an aggregation point,” Bryce said. “It has been built out a bit, but I wouldn’t say that there has been a big leap forward for it in Icehouse.”
Overall, as OpenStack continues to evolve with each release, the challenges facing the platform from a development perspective are decreasing, according to Rackspace’s Engates.
“As things mature and settle down a bit with OpenStack, it gets easier, not harder,” he said.
Sean Michael Kerner is a senior editor at eWEEK and InternetNews.com. Follow him on Twitter @TechJournalist.