July 24, 2012
The open source cloud space is seeing a lot of activity lately, with OpenStack turning two, and the seemingly never-ending cloud API wars. As with reality television, the mudslinging is surely a big draw, but at least one cloud management platform would like to stay above the fray, and that's OpenNebula. After communicating with OpenNebula Project Director Ignacio M. Llorente, one has a sense of where this equanimity stems from. Mr. Llorente is forthcoming when asked for his take on the open source cloud space, but he also makes a point to convey his respect for the other players. Having spoken with sometime-competitor Eucalyptus Systems, it would seem the positive feelings are mutual.
Currently, OpenNebula is involved in several high-profile use cases, so I asked Mr. Llorente a few questions about the state of adoption and his vision for the future of the project and the open source cloud space.
HPC in the Cloud: Please share a little about the history of OpenNebula and how it has evolved to where it is today?
Ignacio M. Llorente: OpenNebula started as a research project in 2005 and made its first public software release as an open-source project in March 2008. OpenNebula is the result of many years of research and development in efficient and scalable management of virtual machines on large-scale distributed infrastructures, in close collaboration with our user community and the main cloud computing players in the context of flagship cloud computing projects. OpenNebula is now an active project with a very large user base, more than 5,000 downloads per month and thousands of deployments that include leading research centers like FermiLab and ESA; supercomputing centers like SARA and NCHC; telecom operators like RIM, China Mobile and Telefonica O2; and integrators like Logica, Engineering and KPMG. A main differentiation of OpenNebula lies in a roadmap completely driven by users needs with features that meet real demands, not features resulting from agreements between IT vendors. Many of our users, like RIM, Logica, Akamai, FermiLab, are also active contributors to the software. In other words, we are a traditional open-source project.
HPC in the Cloud: We understand that China Mobile has been using OpenNebula since 2008 and is about to launch a public cloud based on the software. Can you tell us a more about that?
Llorente: China Mobile started to evaluate OpenNebula four years ago, as the core component of Big Cloud Elastic Computing, China Mobile's flagship cloud infrastructure. Big Cloud is being used in China Mobile's internal business and is now ready to support China Mobile's operation platform and provide services to its more than 600 million customers – yes, this is a huge number! The Big Cloud stack includes other pioneering components developed by China Mobile Research Institute like a new object store system, named Onest. We are proud to know that OpenNebula is the cloud operating system that meets the demands of China Mobile's cloud for high performance, low cost, high scalability, and high reliability; and very excited with their upcoming contributions to the project.
HPC in the Cloud: OpenNebula is also part of the pan-European Helix Nebula project, in what capacity?
Llorente: OpenNebula is being used as a cloud management platform in some of the proof of concept deployments, being deployed in data centers of CERN, EMBL and ESA to aid their computing needs in a flexible manner. As a highly adopted cloud management platform in research and HPC environments, we know the typical requirements of science cloud environments. This collaboration with Helix Nebula also helps us incorporate new innovations, stemming from the requirements of these humongous research centers, the so called demand-side in Helix Nebula speak.
HPC in the Cloud: Are you at all surprised that it is being used in such a prominent ways?
Llorente: We are very happy to see how OpenNebula has grown to currently become one of the most successful open-source management solutions for virtualized enterprise data centers and private clouds. More considering that we do not invest in marketing, we only invest in technology in order to provide our users with the best fully open-source enterprise-class product to build private clouds. We think this demonstrates our unique features, as well as our mature and proven code-base. We serve our users and they appreciate we are really user oriented, vendor agnostic and free of marketing. Our commercial users also appreciate that we provide all the key functionalities for cloud computing, storage and networking within a single open-source enterprise-class product. This reduces cost and complexity of the cloud, and ensures its long term stability and performance through a single integrated patching and updating process, and one-stop support.
HPC in the Cloud: It seems like there's a lot going on for the OpenNebula.com project. In July, you announced a new release and introduced the virtual appliance marketplace - can you share some of the highlights from these announcements?
Llorente: Some weeks ago we announced our new marketplace as an instrument to support the sharing of virtual appliances within the community. In our view, its main innovation is its complete integration with OpenNebula 3.6, so any user of an OpenNebula cloud can easily find and deploy appliances with a single click, and any software developer can distribute a new appliance, making it available to all OpenNebula deployments worldwide in few minutes. OpenNebula 3.6 came with many other new capabilities that makes it a unique software for the management of private clouds based on KVM, Xen and VMware. Moreover, as an enterprise-class product, OpenNebula offers an update path so all existing users can easily migrate their production and experimental environments to the new version. This third series of OpenNebula brings countless valuable contributions from many industry members of its large user community, including Research in Motion, Logica, Terradue 2.0, CloudWeavers, or Akamai.
HPC in the Cloud: What is your vision for OpenNebula? What do you expect for the project in the next two years?
Llorente: Our aim is to be the leading open-source platform for data center virtualization and private cloud computing. Hence, we will continue implementing the features demanded by our users and provide them with richer functionality and wider integration capabilities. In the last months we have seen a growing number of big companies interested in OpenNebula to build a cloud on VMware as an open alternative to vCloud Director. They are interested in a product that, while providing functionality comparable to vCloud, is infrastructure agnostic to also work on Xen and KVM, flexible to fit into any datacenter, open in order to avoid vendor lock-in, and much more cost effective.
HPC in the Cloud: How do you see the open source cloud space evolving over the same time frame? Is there room for more than one open source platform?
Llorente: Absolutely, our vision is that there will be more than one cloud tool. There is not a single perfect solution for everything. We envision an open-source cloud space with several offerings focused on different environments and/or industries. In this vision we will be the standard open solution for private clouds.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?