By Rich Bourdeau, Vice President, DynamicOps
Calculating the return on investment (ROI) for
virtualization is easy: if you consolidate ten physical machines onto ten virtual machines, running on a single physical server, the capital savings can be quite substantial.
However, achieving these savings is not as easy as the simple math might suggest. Many companies adopt virtualization as a cost-cutting mechanism in order to maintain services with flat or shrinking budgets. Most companies that have deployed virtualization to any extent quickly realize that they will not be able to attain the savings they originally envisioned without the efficient and effective management of their virtual infrastructure. In many cases, inefficient management reduces the anticipated savings and can slow the deployment of virtualization solutions.
This article looks at the savings that can be achieved by deploying policy-driven, automated virtualization management software. This type of solution helps improve service delivery, while further lowering the operational costs of managing a virtual infrastructure. In addition, automated virtualization management helps companies ensure that they are more effectively using their shared physical infrastructure. The combination of the increase in operational efficiency and the improvement in resource utilization helps accelerate the deployment in virtualization, thereby generating additional cost savings.
Challenges
All hypervisors come with management software. So why should companies invest additional money for software to help them better manage their virtual infrastructures? The reason is that as virtual infrastructures grow from proof of concept pilots to full production use, the complexities of managing, tracking, and controlling a scaled-out virtual environment very often exceeds the capabilities of these management tools. Some of the challenges that companies face as they try to scale their production virtual infrastructures include:
- Enforce governance and compliance.
Virtualization hypervisor managers have the capability to manage every unique hypervisor feature, but they do little to enforce which resources a user is allowed to consume or how the machine will be built and managed throughout its life. In addition, these tools require new administrative skill sets.
- Managing a virtual infrastructure involves more than just configuring a virtual machine.
Delivering a virtual compute resource to its consumer involves more than just configuring the virtual machine. In most cases, there are requisitioning, planning, pre-configuration, and post configuration processing that must occur. For most companies, these additional tasks are manual processes, which lead to inefficiencies and errors. The problem is only compounded when companies try to integrate their virtual environment with their existing management ecosystem.
- Delivering a virtual machine frequently involves using multiple management tools.
In most cases, tools beyond a hypervisor element manager are required to bring a machine online. An example is the deployment of a virtual desktop. In order to accomplish this task, a system administrator is likely to have to use three separate management tools: the hypervisor management software, the image deployment device manager for software like Altiris, Tivoli, or Citrix Provisioning Server, and the connection broker’s device manager to configure the user’s access rights. There can be additional management tools required, such as patch management applications or CMDBs, to integrate the virtual infrastructure into the existing management ecosystem.
- Multi-vendor environments add additional management complexities.
Today, most companies only have one hypervisor. However, if operating systems are any indication, that status will likely change for most companies. Even with a single hypervisor today, the management tools required to administer a single vendor’s virtual servers and virtual desktops are often not integrated and frequently incompatible.
What is Automated Virtualization Management?
Automated virtualization management software orchestrates the end-to-end service delivery and ongoing management of a virtual machine. Throughout the machine’s life, it monitors, tracks, and enforces policies which control resource consumption, as well as compliance with internal standards regarding how machines should be built and managed.
These software tools manage the hypervisor’s functional capabilities, and help orchestrate the interactions with other complementary virtualization technologies, including integration with a company’s existing management infrastructure.
In terms of capacity planning and management, these automated tools go well beyond capacity and chargeback reporting. They identify stranded, inactive, and abandoned resources and help automate the reclamation and recycling of those resources in order to optimize utilization and further lower operational and capital costs.
The ROI Behind Automated Virtualization Management
Automated virtualization management software is easily cost justified based on its ROI. By enhancing the hypervisor’s native management capabilities with policy-enabled governance and process automation, these software products help companies increase operational efficiency and optimize resource utilization, helping them to further accelerate their virtual infrastructure deployment plans.
- Increase Operational Efficiency
Theprovisioning time for virtual machines has been reduced from months and weeks to days and hours. However, most companies still spend four to six person hours, spread out over a few days of elapsed time, to provision a new machine because there is still manual processing that must take place. The same is true for other management operations like reprovisioning, snapshot, decommissioning, and archival tasks. Errors introduced via these manual processes, and the rework associated with these errors, further contribute to management inefficiencies.
The lack of governance within the hypervisor’s management software also contributes to operational inefficiencies. Many companies start by having a single administrator be the policy enforcer . This works, but in order to scale the environment, more people are required. Additional administrators lead to additional operational costs. Also, it is much harder to standardize manual processes across multiple administrators, which typically leads to the breakdown of governance and compliance controls.
Automated virtualization management software helps companies accelerate the delivery of IT services by automating the end-to-end processes associated with virtual infrastructure management. By removing the manual processes, they eliminate the work necessary to perform these tasks and also eliminate the rework associated with errors. Policy and access controls enforce governance, which limits user access to pre-allocated compute resources and enforces compliance with how machines are built, managed, and eventually decommissioned.
Automated management allows companies to delegate operational control further downstream because the software abstracts the management complexities and its predefined policies ensure that the IT administrators remain in control. Neil Smith, a virtualization engineer at Man Investment Group, said, “Prior to implementing our automated virtualization management solution, it would take half a man day to install and configure a desktop for a developer to use. Now, it is twenty minutes and the developers are self-served.”
A typical company managing 1,000 virtual machines can achieve annual savings of $200,000 by deploying an automated virtualization management solution.
- Optimize Resource Utilization
Virtual machine sprawl is one of the biggest concerns facing companies that have deployed virtualization. The ability to quickly create virtual machines without the same disciplines and controls that are in place in the physical world results in machines being provisioned unnecessarily without proper justifications and approvals, machines being over-provisioned (too much CPU, memory or disk), and machines consuming resources when they are no longer required.
Reduce, reuse, and recycle are the three basic rules which drive the overall environmental movement. These same rules directly apply to the efficient management of a virtual compute infrastructure. Automated virtualization management software provides policy governance and compliance enforcement, which optimizes resource utilization in the following ways:
REDUCE - With the appropriate policies in place, administrators control not only how the machine will be built, but what resources will be used and how much of those resources will be consumed by each VM. For additional control, policies can be established which automate the approval workflow, further assuring machines will not be provisioned unnecessarily or over-provisioned without the appropriate business justification.
A typical company with 1,000 VMs, which has five percent of their machines created without proper business justification and another five percent of their machines over-provisioned, could easily save more than $100,000 - $150,000 in capital expenditures through better control of the front-end provisioning process. By delivering the “right sized” machine at the “right service level”, companies can eliminate waste, improve resource utilization, and lower costs.
REUSE - The degree to which resources can be reused has a high impact on the overall efficiency of a virtual infrastructure. There are a number of use cases where machines are only needed for short periods of time, yet they continue to exist well beyond when they are needed. Virtualization management solutions that automatically reclaim and reuse resources when they are no longer needed are key to enabling more efficient resource reuse.
A typical company with 1,000 virtual machines, that has five percent of its machines used for temporary applications, can expect to save about $60,000 - $70,000 by automating the reuse of these machines. Environments like development and testing, which have a larger need for more temporary machines, can expect even higher gains. The same sized company that archives ten percent of their machines on a yearly basis can expect to save between $30,000 and $50,000 annually, by reclaiming and reusing storage resources after the archive period has ended.
RECYCLE - The first problem with reclaiming inactive resources is the identification of inactive machines. For most companies, this is a manual process consisting of data collection scripts and spreadsheets. The second problem is filtering out the truly inactive machines from those that just appear to be inactive. Automated virtualization management solutions not only provide exception reports that help identify stranded, inactive, and abandon machines, but also automate the workflow associated with reclaiming and recycling those resources.
A typical company, with 1,000 virtual machines and ten percent of its resource being consumed by inactive and abandoned VMs, can expect to save between $80,000 and $100,000 on capital expenditures annually. Even if the automated process is run more frequently than their current manual process (e.g. quarterly vs. annually), companies can expect to save another $20,000 in operational expenses associated with identification and reclamation of inactive resources.
- Accelerate Virtualization Deployment
There is very little argument that virtualization can improve overall business agility while reducing costs. If this is the case, why have companies not virtualized a larger percent of their IT infrastructures? While there are some valid concerns and limitations about the core virtualization technologies, one of the biggest barriers to expanded virtualization deployment is the efficient management of the virtual infrastructure.
Virtualization is a big paradigm shift compared to the physical IT world. Many of the system management tools and processes that worked fine in the physical world cannot adequately address the new challenges of the virtual world. Without the appropriate policies and controls in place, most companies have slowed the deployment of virtualization in order to avoid losing control of their virtual infrastructure.
Proper governance and compliance enforcement available in automated virtualization management software enables the efficient management of a virtual infrastructure, allowing additional portions of a company’s infrastructure to be virtualized. The productivity and capital savings related to more efficient management can be used to fund the hardware required to grow the physical infrastructure that is needed to support the virtualization expansion.
A typical company with approximately 2,000 physical servers and 1,000 VMs could save between $400,000 and $500,000 annually by accelerating its virtualization plans by just five percent.
Proven Customer Savings
Virtualization management solutions that automate the workflow and enforce policies related to the provisioning and ongoing management of a virtual infrastructure, can typically justify their additional cost through improved operational efficiency alone. In addition, they provide other savings by optimizing resource utilization and helping accelerate virtualization deployments by supplying the tools necessary to managing a growing virtual infrastructure more efficiently.
The table below highlights the potential savings that can be achieved by deploying more a more efficient virtual infrastructure management solution.
As with any estimate of potential cost savings, the results are somewhat dependent on the specific environment. However, most companies intuitively realize the inefficiencies in both their operational management overhead as well as resource utilization. The ROI model used to calculate the savings for various size configurations shown above helps companies quantify their actual savings.
Brad Novak, managing director of End-User Platforms at Credit Suisse, indicated that "automated virtualization management has made desktop virtualization a win/win for Credit Suisse. We are able to improve the user experience through rapid service delivery, while maximizing the operational ROI through automation and standardization."
Author Bio: Rich Bourdeau has more than 20 years experience developing, managing, and marketing IT infrastructure management solutions. He was vice president of product management at Incipient, and held product marketing and management positions at EMC. He is currently vice president of product management at DynamicOps.
Article Tags