theCUBE Live At OpenStack 2014 – Behind The Scenes

In April, when OpenStack released open source OpenStack Icehouse, database as a service project Trove was one of its main features. Fast forward six months: OpenStack Foundation releases OpenStack Juno with several new projects, and lots of new features.

The 1,400 contributors to this version, OpenStack Juno,  included multiple bug fixes (more than 3200 have been squashed) and added more than 340 new features to the cloud framework. The main contributors in terms of backers were Red Hat, HP, IBM, Mirantis, Rackspace, SUSE, the OpenStack Foundation itself, B1 Systems, VMware and NEC.

With so many changes and functionality additions to work with, developers will find a highly configurable playground of modules and APIs to build a robust, enterprise-grade cloud. In this release, OpenStack opened up cutting-edge networking capabilities, enhanced cloud scalability for storage and compute, and bolstered backend and API functionality.

Developers can get hands-on with OpenStack Juno by trying it out in a public cloud or by diving directly into the code by downloading and installing DevStack, the local dev environment.

New modules to ease cloud development

This 10th version of the OpenStack is marked by the arrival of a new module called Sahara and whose goal is to enable the provisioning and administration of Hadoop clusters on top of OpenStack. Sahara in particular benefits from the contributions of Hortonworks and allows the automatic provision of Hadoop clusters.

For those developers looking to create big data analytics solutions, the latest release adds a new data processing service to the existing suite of cloud capabilities–compute, object and block storage, networking etc.

Juno provides greater control over data created with OpenStack clouds through the establishment of storage policies. Users can now decide how they want to store, reproduce and access data across different systems and major geographic regions,” says the foundation. This functionality adds to a long list of updated features including the ability to measure and monitor.

OpenStack focuses on the needs of telecommunications operators. Indeed, a task force was created in May. It is dedicated to NFV (Network Virtualization Functions) functions. The main tools were added to Juno and the work of NFV to continue development as the new releases.

Swift has made ​​a significant contribution to OpenStack since 2010, allowing developers to store multiple copies of the same object on different hardware devices, in order to cope with different hardware failures without losing useful information. With Swift object and block storage developers get a persistent distributed file system for keeping data flowing across the stack.

API storage Cinder (block mode) is too enhanced with new features for managing snapshots, replication of data or management of service quality.

In the new version, OpenStack received an update to Neutron, the portion of code dedicated to Software Defined Network technology. With Juno, it is easier to migrate the network configuration from Nova Neutron and use the advanced features of SDN. In Neutron networking, developers will find multiple networking models including flat networking or VLAN implementations; IP address management for dynamic routing; as well as SDN technology such as OpenFlow to ease network operations.

Juno also improves some functions already present, starting with the Compute brick known as Nova. It includes improvements in rescue mode, option-specific network through improved network Nova-code settings, programming updates for the support services and the scalability and internationalization updates. The network portion (Neutron) offers IPv6 support; the identification module (Keystone) simplifies authentication with LDAP; orchestration (Heat) improves delegation privileges to create resource users; Dashboard (Horizon) enables the deployment of Apache Hadoop groups in a few clicks; and finally, the service database (Trove) has a system of management services in relational databases for OpenStack environment.

Developer community role in Juno

The OpenStack ecosystem continues to experience outstanding growth. There is a lot of input from users running OpenStack in production, and that translates into a lot of work (“How To Contribute”) making OpenStack easier to build, scale, operate and manage. The developer community have made almost 130,000 commits with 61,000 in the last 12 months and an average of 4,000 monthly commits since the beginning of 2014.

The OpenStack developer community continues to be a very impressive animal possessing organs to enhance collaboration such as mailing lists (Announcements, Operators, etc.) in multiple languages. It extends into social networks including Twitter (@OpenStack), Facebook, LinkedIn, and Ohloh. The list of OpenStack user groups is expansive and reaches from North America to the Middle East.

And for the enterprising developer, OpenStack has an IRC channel: #OpenStack on Freenode.

Contributors: Saroj Kar and Kyt Dotson