MANILA, Philippines - People have all heard of virtualization and how it can serve up a multitude of benefits. These benefits include lower TCO, simplified physical infrastructure, scalability, resilience and reliability, and faster service delivery.
Virtualization also improves readiness for cloud computing, which brings with it its own set of benefits while enhancing those from virtualization.
But while virtualized and cloud-based data centers do deliver significant and demonstrable benefits, they can be rather complex to manage. This is especially so if they were evolved using multi-vendor assets and legacy applications and systems.
Take, for instance, hypervisors, the software used to manage virtual machines. Because most enterprises do not have the luxury of new greenfield-type data centers, more often than not, several hypervisors are used in a single virtualized environment.
Each hypervisor comes with its own management tools, and there is typically very little integration among the different sets of tools. This makes performing tasks across multiple hypervisors — and corresponding applications and hardware infrastructure — a challenge.
In addition, the use of more than one hypervisor makes it difficult to dynamically scale the IT environment to match the variability of workloads. Some IT organizations work around this by putting in a capacity buffer, which comes at a cost in the form of excess hardware and/or software licenses.
The case for a common toolset
To streamline their operational activities and standardize technology management processes CIOs responsible for virtualized and cloud-based data centers need a set of common tools for managing server, storage, networking and other assets.
For those who have yet to go virtual, these tools can help transition existing and new technologies to an open, virtual, cloud-like model that dynamically provisions application workloads and unifies heterogeneous assets into a common pool.
One example of such a transition can be found in Clemson University, one of the top research universities in the United States. Located in South Carolina, the university has 17,500 students and 1,400 faculty members.
Clemson found that its end-users wanted to access their applications in a capacity-on-demand manner. However, its fragmented IT environment could not support these needs.
Moreover, there were a lot of inefficiencies in dealing with and setting up and creating infrastructure needed for deploying new applications.
“We needed to build a more cloud-like infrastructure so we could be agile with our heterogeneous computing environment so we could, on demand, say we need this application, we need more of this application and just turn it on, crank it up,” said Boyd Wilson, executive director for computing systems and operations.
To modernize, consolidate and virtualize its infrastructure to enable cloud services to its end-users, Clemson turned to Dell. “We talked to many companies at a very high level and Dell was the only one that stepped up first in a major way. What we are doing is we are virtualizing the entire infrastructure, not just the servers, but network and storage resources as well, and that led us to Dell Advanced Infrastructure Manager,” said CIO Jim Bottum.
The manageability bugbear
Management complexity in virtualized and cloud-based data centers can also impede the speed of response to user demands, slow down the identification and resolution of technical issues, and make it harder to control virtual machine sprawl.
Not surprisingly, a recent IDC survey found IT decision-makers responsible for managing virtualized environments indicating their most important challenges as those relating to reducing complexity (76.1 percent), integrating virtual and physical server management (73.1 percent), improving virtual server lifecycle operations (69.7 percent), and reducing virtual machine sprawl (65.2 percent).
Besides making data center technology easier to manage, CIOs also need to address the people and processes needed to efficiently operate a virtualized and cloud-based data center.
To be able to do this, they need to first recognize that operational activities (problem management, approval, configuration, provisioning, and so on) in such an environment are different from those in a traditional one.
In a traditional data center, such activities are performed and “managed” in a fragmented and often manual manner to match the heterogeneity of the underlying technology and groups of end-users.
In contrast, in a virtualized environment, the needs of multiple business groups and application environments are supported by a shared resource pool. Accordingly, operational activities have to be approached on a more automated, standardized, and integrated basis, with a view to holistically managing the entire virtualized environment.
Tellingly, 71.1 percent of the participants in the IDC survey mentioned above indicated standardizing management processes as an important challenge.
Dell VIS
Part of the Dell Virtual Integrated System (VIS) suite of virtual and cloud infrastructure management solutions, Dell Advanced Infrastructure Manager (AIM) software simplifies data center management by enabling a single administrator using any leading hypervisor to allocate server, storage and network resources — physical or virtual — against application workloads. The solution can bring together heterogeneous hardware offerings and leading virtualisation hypervisors to create virtual pools of resources that are easy to manage.
With Dell AIM, Clemson completed a rapid data center migration to a new architecture, consolidated infrastructure and created a flexible, agile, resilient cloud-ready environment.
Managed using Dell AIM, this new environment helped Clemson exceed customer service levels and resulted in both operational and capital expenditure reduction. Among the operational benefits: the time needed to provision physical servers has been slashed from eight hours to a mere 15 minutes, disaster recovery is now performed in minutes versus weeks previously, and workloads have become much more mobile between servers.
“The way that AIM handles workload mobility is unique and allows the application to quickly move to new servers in the event of an outage. We were not able to find another cost-effective way to get there. In addition, AIM will allow us to scale the architecture on demand. It frees us up so we’re not constantly upgrading,” said Bottum.
Integrated suite
“Dell recognizes organizations like Clemson have made significant technology purchases. The Dell VIS approach allows them to preserve and fully leverage their existing data centre infrastructure, respond to changes faster by selecting and deploying standard IT resources almost instantly, customize workflows to help automate common tasks associated with workload deployment, deliver right-sized IT capability at the right service levels cost-effectively, and support business growth while keeping existing resources under control,” said Christopher Papa, country manager of Dell Philippines.
Complementing Dell AIM in the Dell VIS suite is Dell VIS Self-Service Creator, which can shorten the time needed to deploy new business applications by standardizing and automating the way applications are deployed.
The solution incorporates a Web-based portal that enables authorized users to select, deploy and manage a customized catalog of IT applications and resources, reducing the time it takes to deploy workloads across multi-vendor hardware and applications.
To help enterprises assess, design and implement VIS into new or existing IT environments, Dell offers a range of service capabilities.