Home > Blogs > VMware Operations Transformation Services > Monthly Archives: August 2016

Monthly Archives: August 2016

Taming the Hydra: IT in a Multi-Cloud World

Kevin Lees

Photo Credit: iStock/ZU_09

Photo Credit: iStock/ZU_09

By: Kevin Lees

I know it’s hard to believe, but VMworld is upon us again. Where did the last year go? I know many of you were busy implementing, updating, extending, and managing your private cloud environment. Perhaps you’re busy implementing and getting comfortable working with NSX; busy realizing the great benefits software defined networking and security brings you as an IT service provider to you line of business customers. Regardless, I’m sure you’re busy making your customers happy with the standardized and customer-driven IT services you’re providing.

So, you’re getting comfortable in your private cloud world – but not too comfortable I hope, as a recent VMware study (VMware Customer Advocacy study”, Q1 2016) shows that 48% of enterprise applications will be cloud based within 3 years. Ok, that doesn’t sound too bad. That is until you see another statistic that says 65% of enterprises will use more than one public and/or private cloud platform and for 67%, and that:

For 67% of enterprises the ideal end state includes relying on multiple cloud platforms.

Relying on multiple clouds, really?

So how are you going to deal with that? How are you going to bring order to potentially multi-cloud chaos? It’s hard enough to get a handle on the workloads your enterprise developers are placing in one public cloud, how bad is it going to get? Well, if you start planning for multi-cloud now, it may not have to be bad at all. In fact, with proper planning, you can drive a good degree of control – without all of those pesky developers even knowing the difference.

If I’ve piqued your interest – or perhaps triggered the onset of a slight panic attack, please stop by my session, SDDC8994: Taming the Hydra: IT in a Multi-Cloud World, next Monday at 3:30 at VMworld 2016, and I’ll fill you VMworldin on some concrete steps you can start taking now to avoid the chaos as well as how the looming multi-cloud world might impact your IT role. I hope to see you there.

Download a full agenda of VMworld breakout sessions that will help IT leaders build a strategy for the digital era.

=======

Kevin Lees is Principal Architect for VMware’s global Operations Transformation practice and is based in Colorado.

3 Steps to Create an Automation Roadmap

Ahmed_croppedAutomation RoadmapBy Ahmed Al-Buheissi

Automation is at the heart of any cloud implementation.  It provides fast provisioning, resource monitoring and self-healing, capacity adjustment, and automated billing.  Also, automation will ensure consistency, prevent errors and free-up valuable staff time to work in more innovative projects.

But in order to embrace automation, the organization needs a roadmap. This roadmap needs to be based on an understanding of the current state of the organization, in terms of technology, people and process. You must also examine where the organization will want to be in terms of automation and define “to be” state. The roadmap creation process will determine what tools, skills and services are required to achieve the automation target, and then schedule these improvements to achieve the requirements.

With a comprehensive roadmap, the organization can be well-prepared for the journey, in terms or time, budget and resources.

Three Steps to Create an Automation Roadmap

There are three recommended steps to take in order to create the Automation Roadmap:

  1. Assess Your Current State: Using industry best-practices, you need to start off by assessing the organization’s current state in terms of:
    1. Technology
      What technology is available and fully adopted, in areas such as virtualization, self-service, automation and orchestration? Even DevOps-related tools should be assessed.
    2. Process
      Are related process and policies documented and implemented? For example, Service Definition, Request Fulfilment and Release Management.
    3. People
      Specific skills and roles are necessary for running an automation-oriented infrastructure. Some of these roles include Service Architect and Infrastructure Developer, which need to be documented, formalized and assigned.
    4. Interactions
      Ensure that proper interaction procedures are in place, such as interactions between groups, to the business, and to service providers.
  2. Get Your Priorities Right: You need to identify potential processes for automation, in areas such as IaaS, PaaS, Proactive Operations and capacity Monitoring. Once these opportunities are identified, they need to be evaluated and prioritized in terms of process, impact and readiness.
  3. Put it all on the Map: Now that we know where we are and what we need, we can put it all on a time-line chart. When creating the roadmap some consideration needs to be given to the length of time for the roadmap, as well as time and order required for implementing tools and processes.

If you want to learn more about establishing your Automation Roadmap, please join my Quick Talk at VMworld in Las Vegas:

VMworldAugust 28th 2016 –  1pm
“Service Automation Roadmap: Approach and Samples”
Add session SDDC7876 via the VMworld Schedule Builder

Download a full agenda of VMworld breakout sessions that will help IT leaders build a strategy for the digital era.

=======

Ahmed Al-Buheissi is a Senior Solutions Architect with the VMware Operations Transformation global practice and is based in Melbourne, Australia.

How NOT to Fail at Process Automation and Continual Improvement

Chris KunselmanProcess Automation and Continual Improvementby Chris Kunselman

A big part of my job at VMware is to help customers transform to an SDDC environment. For VMware Operations Transformation Services, that means understanding the business reasons behind their strategy for an SDDC and Private Cloud, and then defining the roadmap to achieving this strategy.  The transformation is the journey along the way.  Typically, this means significant changes to an organization’s culture, paradigms, skills, technology and operational processes.

Usually our customers already have some level of IT Service Management (ITSM) maturity around operations.  They also usually have highly qualified talent who understand the management of technology.  However, what they sometimes lack is the mindset, skills, and experience necessary to effectively scale a privately operated cloud.  To do this, they need to focus on end-to-end process automation through a rigorous approach to continual improvement.

Without an approach to automation, companies will not be able to scale their operations, retire their legacy environments, and realize the full business value from their investment in VMware technology.  Hear more about this at my VMworld 2016 breakout session, “Tips for Realizing the Full Value of Your SDDC Transformation” (add SDDC7692 to your agenda).

Continual improvement and process automation go hand-in-hand if an organization truly intends to operate their cloud as a business, and gain the benefits from the use of vRealize Automation (vRA), vRealize Orchestration (vRO), and vRealize Operations (vROps).  VMware solutions bring a new potential for an organization to finally achieve the highest level of process maturity.  Automation and continual improvement should be a component of every private cloud operations team.

The Importance of Continual Process Improvement

Imagine that you currently have six globally located private cloud data centers, and four thousand application workloads already migrated to your private cloud.  Not only that, but you are continuing to migrate an average of 400 legacy workloads per month into your SDDC.  That means in one year, the number of your workloads will have more than doubled!

In my experience working with customers, most of them don’t plan to manage their IT services in the same way and with the same number of resources as they manage their legacy environment.  They don’t plan to double their team size as their SDDC grows, but this is a very likely scenario.

Virtualizing your infrastructure creates more demand from the business, for even faster delivery of the same legacy services.  Merely moving your legacy environment to your new virtual private cloud does not give you the full payback you would expect, unless you also have a strategy for continual improvement and automation.  A vast amount of operational cost can be saved if you automate.  However, to achieve this, your continual improvement team should have a good mix of business process re-engineering, systems integration, and development skills

VMware’s Approach to Automating Processes and Establishing Continual Improvement

Say, for example, the request for increasing capacity occurs very frequently in your environment.  Let’s say it takes your team members eight hours to provision 20 of these requests in vRA.  The reasons it takes so long may vary, but it usually boils down to regulatory controls, change control approvals, documentation, and many tedious steps in various systems, performed by different teams.

You start with automating this process so that it takes just a button push to approve it and the rest happens automatically through vRO orchestration, and vRA to provision the request.  After taking out most of the manual tasks in this process, your same resource could provision 20 of these requests in an hour, becoming eight times as productive.  This is the type of scenario you want to tackle on a continual basis.

VMware operations transformation consultants use an incremental approach with customers that works well within agile IT environments.  We work with your business process team to identify these opportunities, prioritize them, and drive them to completion through a series of development sprints.  This requires a strong collaboration between IT business process experts, your ITSM systems and your VMware automation/development team.

Customers often miss crucial aspects of continual improvement, leading us at VMware to create a practical and comprehensive step-by-step approach to follow continual improvement practices by automating end-to-end IT processes.  By following this approach, customers can derive measurable value from their private cloud transformation by increasing the quality of service delivery, removing unnecessary labor, and human intervention

Deep Dive Analysis

We begin with a deep dive analysis of the particular processes we aim to automate.  We look at all types of changes that exist to identify changes that are executed frequently, performed with consistency and require many steps and approvals, then determine which of these can be automated.  We establish a backlog of these automation opportunities.

It is important to note that due to regulatory concerns, it is not possible to automate the provisioning steps of every type of change, especially major changes.  Typically, the list of potential automation opportunities is made up of standard change types that do not require approval.

Because we want to automate the end-to-end IT process, it is important to look at every segment of the manual process.  Some process segments will require forms and data capture, or integration that populates form data and automatically passes this data into vRO, and/or vRA.  Other processes may require approval steps.  This type of end-to-end automation involves integration between multiple systems, such as request ticketing, change ticketing, and VMware vRA, vRO systems for provisioning and orchestration.  We look at the business processes, business and compliancy requirements, as we engineer the automation to ensure we automate the right activities.

Calculate Expected Business Value

Next, we calculate the expected business value to be gained from the automation of each change type that we have identified as an automation candidate. This is a cost analysis of the duration, time, and number of people it currently takes perform each step in the process.  From this information, we can determine the “time-cost” of the entire process.  Once we have an idea of the time cost of every process in the backlog, we can determine which automation opportunities will drive the most benefit.  We rank and prioritize these opportunities based on how much value they will bring to the overall virtualization business case.

Develop a “To-Be” Design for Each Process

Starting at the top of the priority list, we write flow diagrams for how the process will work in an automated fashion.  The workflow diagrams include the user to user, user to system, and system to system interactions and use cases.  From these diagrams, we define functional requirements, inputs, outputs, triggers, roles and responsibilities.  Most importantly, during this step, we establish reasonable metrics for measuring the outcome of the effort.

Validation & Implementation

We then hand-off the functional automation requirements to the automation development team.  This team analyzes the functional requirements, and then creates a technical design.  During the design stage, the technical design team comes back to the business process automation engineer who validates whether or not the technical design meets the functional requirements.  This ensures that the goals for time-cost reduction will be met.

Very often, the development team cannot fit all the required functionality into the first release of this process automation.  So we also must prioritize the technical requirements associated with each automation opportunity, placing them into various releases.

By the time the final release is implemented, the full end-to-end automation exists and we are able to measure the business value metrics that we set out to achieve.  The customer begins capturing performance information on the automated process to determine if the automated effort is consistently saving the time cost it is intended to save.

Keys to Success

Following this methodology ensures that our customers get the benefits they were looking for, maximizing cost savings and optimizing operational efficiency.  By effectively prioritizing your process automations, we ensure the ongoing success of your continual improvement program.  We analyze the functional requirements of each end-to-end process first, ensure that we evaluate improvements to the process before automation, and make sure improvements align to existing policies and regulatory requirements.  It doesn’t make sense to automate a bad process.

Tight integration between your business process and your technical development teams is also key to success.  In some situations, a technical team will attempt automation on its own, often with little understanding of the full end-to-end process.  When that happens, their effort may not focus on achieving a clear business objective. By not evaluating the full end-to-end process, their attempt of automating only parts of a process, will not reduce the cost of labor effectively

By formalizing automation efforts as part of a continual improvement program, customers can achieve these improvements incrementally, over time, and within their normal operational budget.  But most importantly, they gain increasing value from the investment in VMware technology.  With our continual improvement approach, one successful automation project creates a “quick win scenario” that sets you up to tackle all the others that follow, with a focus on achieving and measuring specific business outcomes.

If you are interested in creating a plan for process automation and continuous improvement, reach out to your local VMware representative and my team would be happy to help.

=======

Chris Kunselman is a Senior Transformation Consultant with VMware Operations Transformation Services and is based in Pennsylvania.  ckunselman@vmware.com