Roughly six months after the launch of its “Folsom” release last fall, OpenStack on Thursday unveiled version 2013.1 “Grizzly,” the seventh and latest release of the open source software for building public, private and hybrid clouds.
More than 500 contributors made 7,620 updates in this new release, which, “more than any before it, was driven by users who have been running OpenStack in production for the past year (or more) and have asked for broader support for the compute, storage and networking technologies they trust and even greater scale and ease of operations,” explained Mark Collier, chief operating officer at the OpenStack Foundation, in a blog post announcing the new software.
Best Buy, Bloomberg, NSA, Cisco WebEx, Comcast, CERN, HP, NeCTAR, PayPal, Rackspace and Samsung are among the companies using OpenStack in production.
Developers who contributed to this release came from more than 45 companies, including Red Hat, Rackspace, IBM, HP, Nebula, Intel, eNovance, Canonical, VMware, Cloudscaling, DreamHost and SINA.
'Support for Security Groups'
Seven integrated projects make up OpenStack, each with source code now publicly available: Compute ("Nova"), Object Storage ("Swift"), Image Service ("Glance"), Networking ("Quantum"), Block Storage ("Cinder"), Identity ("Keystone") and Dashboard ("Horizon").
More than 200 new features are included in this Grizzly release, and some 1,900 bugs were fixed. In anticipation of the launch, Linux.com spoke earlier this week with Thierry Carrez, release manager for the project, to hear about some of the highlights.
“Personally, my favorite key features for this release would be in OpenStack Compute (Nova): introduction of the 'Cells' deployment model for massive scale, and isolation of the compute nodes from the rest of the system for better security ('no-db-compute'),” Carrez began. “I also like how OpenStack Networking introduced support for security groups, as well as a load-balancing-as-a-service feature.”
Meanwhile, “I would also mention how OpenStack Block Storage (Cinder) managed to add a large number of storage drivers from all of the storage industry,” he told Linux.com. Ten new drivers were added, in fact, including Ceph/RBD, Coraid, EMC, Hewlett-Packard, Huawei, IBM, NetApp, Red Hat/Gluster, SolidFire and Zadara.
Finally, “the general drive towards more reuse of code across the various OpenStack projects (through the introduction of the common 'Oslo' libraries) is also worth mentioning,” Carrez said.
Cross-Origin Resource Sharing
Other highlights of the new release include significant improvements in virtualization management on the Compute side, with full support for ESX, KVM, XEN and Hyper-V. Quotas were added to the Object Storage system, meanwhile, as was cross-origin resource sharing (CORS), enabling browsers to “talk directly to back-end storage environments,” Collier noted.
For Networking, Grizzly aims to achieve greater scale and higher availability by distributing L3/L4 and dynamic host configuration protocol (DHCP) services across multiple servers. New plug-ins were also added from Big Switch, Hyper-V, PlumGrid, Brocade and Midonet.
The Grizzly Dashboard, meanwhile, is backwards-compatible with the Folsom release. The video below demonstrates the dashboard in action.
Coming Next: 'Havana'
These, of course, are just some of the highlights in this new release. For a full list of features in OpenStack Grizzly, visit the software's Release Notes.
Looking ahead, October will see the release of OpenStack “Havana,” which will feature two new projects incubated during the Grizzly cycle: “Ceilometer,” for central collection of metering/monitoring data, and “Heat,” a template-based orchestration engine. Planning will begin at the OpenStack Summit in Portland, Ore., later this month.