“Metal as a Service” provisioning tool from Canonical in Ubuntu 12.04 LTS beta
by Canonical on 4 April 2012
- MAAS enables system administrators to provision hyperscale deployments of physical servers, bringing cloud-like semantics for on-demand capacity to the physical server provisioning process.
- MAAS is designed for horizontally scaled environments such as big data workloads and internal clouds, but works just as well for any cloud-like deployment.
- MAAS bridges the worlds of cloud and traditional deployment. MAAS is available for testing in 12.04 Beta 2, with Ubuntu 12.04 LTS scheduled for general availability on 26th April 2012.
The shift from “scale” to “hyperscale” in data centre design warrants a shift in the tooling used to provision that physical infrastructure. Canonical’s provisioning tool “MAAS” (for “Metal as a Service”) is available for testing in Ubuntu 12.04 Server LTS Beta 2. MAAS brings the dynamism of cloud computing to the world of physical provisioning, simplifying the deployment of scalable services for big data and cloud. MAAS supports the deployment of infrastructure such as OpenStack, Hadoop, CloudStack, Load Balanced Web and Cloud Foundry which are often deployed on farms of physical servers without the overhead of virtualisation.
MAAS manages a group of physical servers as a cloud-like resource to be allocated on demand. A large portfolio of reusable cloud services can be deployed on MAAS-managed hardware and scaled up or down on demand. Nodes commissioned through the MAAS can be queued for dynamic allocation to a service, or simply have Ubuntu installed for manual configuration and setup.
Servers are easily assigned to the MAAS through automated or manual enlistment, and the MAAS is then responsible for all aspects of physical server maintenance such as firmware updates, burn-in testing, performance evaluation and the elimination of residual data when machines are decommissioned.
The slick web front-end makes it easy for administrators to add, update, commission and deploy physical servers at record speed. It is straightforward to re-allocate nodes between services dynamically, keep them up to date and, in due course, retire them from use. The web dashboard provides a quick, at-a-glance overview of the status of the MAAS cluster, so it’s easy to see how much computing resource is available for deployment and how much is already in use.
Compatibility of cloud and physical workloads
Because the MAAS provides the necessary dynamic allocation of physical resources just like a cloud, it works with Ubuntu’s cloud deployment tool, Juju. Ubuntu is the number one OS on public, private and hybrid clouds, with a huge range of cloud workloads already defined for instant deployment. That full range of Ubuntu workloads in widespread use on the cloud can now be deployed on physical servers with the MAAS. Most importantly, those tools retain their cloud-like ability to scale services up and down, by allocating more or fewer physical servers to them as needed.
The catalog of cloud services deployable through Juju includes all the popular structural components of typical big data, cloud and web service deployments. That includes relational and nosql databases such as Cassandra, MongoDB, Postgres, MariaDB, MySQL, load balancing web proxies such as HAProxy, Varnish and Squid, multiple web application servers such as Node.js, JBoss, Django and Ruby on Rails, common web applications like WordPress and important PAAS frameworks like Cloud Foundry. These services are deployable, connectable and scalable with a single command on the cloud and, through the MAAS, on physical servers too. The catalogue will continue to be updated throughout the lifetime of 12.04 LTS, and individual service definitions continue to improve at cloud speed, extending Ubuntu’s traditional leadership and strength as the OS with the widest array of pre-packaged open-source and commercial applications.
Mark Shuttleworth, founder of Ubuntu and leader of Canonical product and design, explains: “Metal as a Service – MAAS – is a new way of thinking about physical infrastructure. Compute, storage and network are commodities on the metal just as they are commodities in the cloud. MAAS lets you treat farms of servers as a malleable resource for allocation to specific problems, and for re-allocation on a dynamic basis.”
Hyperscale density requires new thinking in physical provisioning
MAAS is designed for an era where thousands of physical server nodes fit in one rack, and it is inefficient to provision them in the traditional manner. MAAS works perfectly well with typical 1U and 2U servers, but its real strength is for big data, cloud, grid and other horizontally scaled deployments.
The MAAS roadmap calls for an update to MAAS with each point release of 12.04 LTS, extending the capabilities and range of hardware certified and supported by both Ubuntu and the MAAS. In future, the commissioning process will include hardware configuration, for example RAID setup and firmware updates, extended burn-in verification of new or old hardware, hands-off IPMI control, and load testing network, disk, and memory to verify system reliability under load. MAAS will make use of Canonical’s Ubuntu hardware certification database for systems and components, enabling administrators to identify systems that are fully certified, or isolate potentially problematic components.
A daily build of Ubuntu Server 12.04 LTS Beta 2 can be downloaded for testing from http://releases.ubuntu.com/precise/. Ubuntu Server 12.04 LTS will be available for download from ubuntu.com on 26th April, 2012.
Related posts
Data Centre AI evolution: combining MAAS and NVIDIA smart NICs
It has been several years since Canonical committed to implementing support for NVIDIA smart NICs in our products. Among them, Canonical’s metal-as-a-service (MAAS) enables the management and control of smart NICs on top of bare-metal servers. NVIDIA BlueField smart NICs are very high data rate network interface cards providing advanced s […]
Canonical joins the Sylva project
Canonical is proud to announce that we have joined the Sylva project of Linux Foundation Europe as a General Member. We aim to bring our open source infrastructure solutions to Sylva and contribute to the project’s goal of providing a platform to validate cloud-native telco functions. Sylva was created to accelerate the cloudification of […]
A call for community
Introduction Open source projects are a testament to the possibilities of collective action. From small libraries to large-scale systems, these projects rely on the volunteer efforts of communities to evolve, improve, and sustain. The principles behind successful open source projects resonate deeply with the divide-and-conquer strategy, a […]