Data Center Intelligence for Transformational Analysis

Just starting out with Virtualization?
Get it Right the First Time!

Many organizations begin the virtualization journey by tackling non-critical, non-production workloads. These environments are often free of the compliance, availability, and performance constraints that govern production servers and make good early targets. Non-production initiatives contain minimal risk for outages or performance issues. 'Do some rough sizing and move the workloads and see what happens'. However, what might work in simpler settings doesn’t apply for more critical environments. Without a more rigorous approach, issues can quickly begin to appear:

Capacity is often over provisioned to ensure safety

Deadlines are missed due to unpredicted issues

Conversions require re-working in order to fix problems

Anticipated consolidation rates may not be achieved

Line of business customers resist over concerns for stability

Critical questions to ask yourself as you start out with virtualization are:

How are you planning to determine what can and can't be virtualized?

Technical missteps are a common reason why consolidation initiatives stall and politics start to drive decision-making. Not every workload is a suitable candidate for virtualization. Some have characteristics that will result in poor performance or failures, while others require specific remediation prior to conversion. Candidate qualification is an essential first step that is often missed or performed by surprise when something goes wrong.

CiRBA provides rigorous candidate qualification analysis, examining both workload patterns and personalities and the configuration details of potential targets against a chosen consolidation strategy. This analysis reveals which workloads may not be suitable for a chosen transformation and those that require remediation and what those preparatory steps are.

How will you determine the actual placements of workloads onto hosts? And which workloads can reside together on the same hosts?

Evaluating workload placements is a critical component of making decisions and plans for consolidation. At the most basic level workload placements determine how efficient and manageable an environment is. When this step is left to ad hoc decisions or performed by “gut”, the result is often over or under provisioning. The placement of workloads is dependent on utilization patterns and personalities, and governed by configurations and business level constraints such as compliance requirements, ownership, and security zones. Advanced organizations enhance efficiency by developing service tiers in their infrastructure to put critical applications into environments with high service levels and others into less expensive infrastructure.

CiRBA enables organizations to evaluate placements as part of the analysis process to determine not just how much capacity is required, but what should go where to meet the goals and sensitivities of the organization.

How will you evaluate different platforms, strategies and approaches?

Sometimes organizations don’t want to consider alternative options and technologies and go into virtualization or consolidation planning with a singular vision of strategy. As the virtualization market in particular matures, multiple viable alternatives are coming to the forefront.. Not considering other options will increasingly result in missed opportunities to lower costs and take advantage of different technology options.

CiRBA provides cross-platform, multi-strategy analysis that enables organizations to quickly model future state using any of the popular platforms, virtualization strategies and alternatives so you can consider different hypervisors, hardware vendors and specifications and even Cloud.

How will you determine the optimal specification and quantity of hardware that will be purchased and how much software you will require? And how will you avoid over or under provisioning?

One of the most common issues facing organizations that have invested in virtualization is inefficient infrastructure. Software and hardware decisions were made based on rough sizing estimates that don’t properly model an environment. Virtualization rates can be overly ambitious and hardware commonly over-sized to accommodate a lack of the detailed analysis required to accurately determine requirements.

CiRBA’s advanced analytics and proven rules enable organizations to accurately model environments and establish true capacity requirements and configurations. In addition, CiRBA’s scope definition and candidate qualification analysis enables organizations to plan waves of virtualization according to it's objectives. Detailed placement and remediation plans enable teams to accurately set and meet virtualization goals, speed the process and ensure that they aren’t stuck holding unconsumed virtualization licenses or large amounts of wasteful whitespace on servers.

How will you obtain buy in from the line of business partners and other key stakeholders?

Internal politics can be one of the biggest issues facing virtualization teams. Common objections range from “my applications are too important to be virtualized” to “there is no way I am sharing infrastructure with another department,” or “our applications aren’t running well because they are combined with that other stuff”. Communication is a critical aspect of success. The ability to show the potential savings and costs of sharing or not, detailing plans and providing visibility into key metrics and considerations all play a role in obtaining buy in.