x86 virtualization and SUSE Linux Enterprise Server



By Carlos E. Montero-Luque

Carlos E. Montero-Luque is the Vice President of Product Management for Novell Open Platform Solutions.

The emergence of broadly deployed x86-based virtualization is causing an evolution of the role of an operating system. In typical physical deployments, operating systems have played the role of running a multiplicity of applications managing local resources and interfacing with the hardware infrastructure. As a way to address conflicts across applications or resource requirements, operating systems have been at times deployed for single applications in single hardware platforms to ensure isolation and resource lockdown.

With virtualization, the operating system role is broken into two distinct pieces: a host component that works in conjunction with a hypervisor to interface with the physical infrastructure (hardware, storage, networks) as well as any centralized management components, and a guest component that becomes a dedicated, perfect container for specific single applications.

These containers (virtual machines) can be deployed in isolation to each other on top of a single hardware platform running on the combination host OS/hypervisor layer. Analogous to the role of client-server computing in the evolution of the monolithic mainframe OS, virtualization separates some of the monolithic aspects of a physically deployed operating system providing additional flexibility, abstraction, and optimization of resource utilization, at a cost in management, security, and orchestration of the new options and choices provided to the user.

Novell is in an excellent position to take advantage of these changes and provide differentiation from other operating system vendors for a number of reasons: as a major operating system vendor specialized in enterprise-class mission critical deployments, with SUSE Linux Enterprise Server we ensure that the operating system both as a host and a guest is optimized for these separate roles while at the same time continuing to be an platform for innovation on traditional physical deployments (which will continue to be a major element of IT infrastructures).

Our role as a key open source vendor in the Linux space also allows us to contribute, guide, and use the innovation from the overall Linux community in virtualization. These assets are extended also in our role in the hypervisor space, whether via the Xen project, the KVM capabilities of Linux, or our strong and growing business, technical, and support partnerships with other hypervisor vendors.

Novell is also in an excellent position to provide world-class solutions to the challenges that broad virtualization deployments bring in terms of additional requirements for monitoring, management, and deployment of virtual and physical elements that make possible their effective use by IT organizations.

Novell’s historical as well as innovative resources for systems management, orchestration, and lifecycle management are being extended to provide support not only to a combination of physical and virtual deployments, but also across heterogeneous environments, on three dimensions: heterogeneous hardware platforms, heterogeneous OS deployments, and heterogeneous virtualization platforms.

In addition, virtualization is opening up for ISVs the opportunity to deploy their applications in self-contained images that are easy to acquire, deploy, install, manage, and run. By separating the “container” element for the application with the interface with the physical elements of IT, applications can be deployed more independently of a single operating system, in pre-defined isolated environments for single use, without demanding as a result dedicated hardware, and able to include simple usage and management interfaces that customers can use independently of OS-specific tools.

This means that SMBs, for example, can use these applications more easily without having internal expensive IT resources and that ISVs can limit their efforts in terms of testing and configuration since they can guarantee for these deployments that the image will be deployed and use in the way they specify only.

Also, if the application needs an update, the ISV can provide a new image to replace the old one altogether, simplifying the update model. In short, by creating virtual appliances that run on host OS/hypervisor combinations, ISVs reduce their own costs, limit their exposure to the customer’s IT infrastructure, and provide customers with a simpler way to use and manage their applications, extending the market reach of their applications to less sophisticated IT environments.

Here Novell is also well positioned as we have an enterprise-ready plaform for virtual appliances with SUSE Linux Enterprise JeOS, a growing ISV ecosystem of applications certified on our platforms, and we also provide management solutions that themselves can make use of the advantages that virtualization provides to vendors and their customers.

Also, since ISVs may deploy full images that include a guest operating system, they can use the large numbers of SUSE Linux Enterprise deployments worldwide to extend their own reach to customers and geographies where they are not leading today. This means that we are creating a powerful symbiotic relationship with ISVs that increases their reach and extends the value of our customers to a broad number of partners.

(Visited 1 times, 1 visits today)

Leave a Reply

Your email address will not be published. Required fields are marked *

No comments yet