Discussion on the impact of the staff model of cloud computing

Author: Reid Engstrom

The prospect of cloud computing to today’s IT staff poses a real concern to them, not only because of the significant element of change and doing something different outside of their existing skill base, but also because so many current IT roles are involved with the installation, configuration and administration of IT infrastructure.  Think of all the job roles today that end with “Administrator”.  Those roles become greatly diminished in a virtualized and cloud based environment.  Staff may be very reticent to embrace this new environment that now is less dependent on the skills and certifications that they have built up over their tenure.  In addition, the justification on reducing cost and complexity is often built on the reduction of those legacy skills that were seen as critical in a physical world.

Technology depth still plays an important role in the cloud world as staff will still need a foundational understanding of cloud enabled technologies such as server, network, storage and end user device virtualization as the IT architecture changes to a flatter model.  This new environment will now reach outside the walls of the organization with public clouds and Software-as-a-Service offerings that may encompass a greater majority of IT services.

There is a new set of critical skills moving forward in the highly productive and automated cloud environments that will be centered around the ability to design and implement systems and processes that automate the cloud based infrastructure of the future.  Some of these proficiencies include technical and business architecture, project management, service delivery, and contracting.  Soft skills of importance include understanding the business, and communications.  These skills are not totally new but their significance has been elevated in a broader based environment that now include service partners and the need for agility to address the dynamics of business opportunity.

Leadership and risk taking abilities will be critical in order to relate the rapidly changing environment to both staff and the business stakeholders.  Those leaders who are confident enough to move forward aggressively will be able to position their IT organizations into a much more proactive state that not only can lead the business but can be a critical contributor to new revenue opportunities.

ITIL and ITSM are the foundations upon which “Cloud Computing” is built.

Author: Eric Ledyard

Today, most people in the IT community hear the word “cloud” and have a reaction similar to drinking bad milk. Through the last few years of every company abusing the word for the purposes of marketing, the true meaning of the movement has been lost, or at least severely diluted. We in the VMware Vision team are not afraid to use the “c-word” and spend each and every day helping advise strategic companies how they can increase the “cloudiness” of their environment. The end goal of each engagement is to increase the agility, efficiency, and manageability of their IT organization and allow them to provide better IT services to their organizations.

So, what does “cloud computing” mean to us?

Because of the dilution of the term, the best way for us to explain what we feel are the core pillars of cloud are is to explain where the movement came from and what the purpose of it is. As an evolutionary term, cloud computing represents the “end-game” for building an IT infrastructure utilizing all the concepts laid out in the ITIL and by implementing an operational IT organization that is focused on the ITSM model. The primary difference that is difficult for most organizations to tackle is that a “cloud environment” is an environment where the IT organization is no longer a cost center but is truly an internal services provider to the company. In each of our engagements, this is the first step to making an organization more cloud-like. Everyone tends to jump to the thought that cloud computing means you must outsource your environment to a third-party cloud provider. That thought process is completely backwards when held against the ITSM model. You wouldn’t start a project by saying: “Whatever application this is, it doesn’t matter, we are putting it on a 1U pizza box server.” This is the same as saying: “To get to cloud, everything has to be on Amazon or Google.” Not true at all. Cloud computing is a method of computing, not a place.

For those of you that may not be completely up to speed on ITIL or the ITSM model, the short summary is that in order to provide services to the business that are meaningful and impactful, you need to be completely aligned to the line-of-business (LOB) owners from the start. Once you are aligned with the business owners it becomes much easier to build an infrastructure to provide tailored services to them. In the ITSM model, the processes by which services are requested and provisioned need to be tightly aligned to the business. Many IT organizations receive a request today to provide infrastructure and they react by purchasing and deploying additional technology based loosely on some requirements that are provided by the application team. In order to be a true service provider, there needs to be a process that provides alignment of the business owners to the IT platform owners. This is outlined in the following diagram:

As you can see in the diagram above, technology platform is the last piece of the puzzle that needs to be addressed. Prior to any purchase or implementation of a ‘product’ whether it be software, hardware, etc, all of the previous pieces of the ITSM model need to be addressed. The thought process needs to change to more deeply understand how the technology impacts the organization and what level of service needs to be provided to that service in order to best serve the organization overall. Working with organizations for years now on disaster recovery, every client I talk to doesn’t have a good understanding from the business on what the SLA’s for recovery need to be for the applications they support. What IT should do is start by building a service catalog of IT services that support the business.

For each service, we need to understand how it ties back to “the big picture” and how that service is used to help the business meet its objectives and strategic goals. For example, there was a story that I was told about a logistics company that when a new CIO came into the organization that was familiar with ITSM, she asked one of the Sr. Systems Engineers on her team: “Do you know what the impact is to the business if the picker application goes down for even 2 hours?” The engineer looked at her and explained that it would mean the warehouses wouldn’t be able to pick orders for a couple hours… no big deal… and explained that a 4-hour RTO should be fine. Immediately, she booked him travel to visit one of their largest warehouse facilities and had him interview the people that worked on the floor in the warehouse. Once he was there, the foreman explained to him that if the picker system goes down for 2 hours and they are not able to fill the trucks that the trucks start lining up outside waiting… putting them each 2 hours behind. Also, some of the carriers were union carriers, so if the truck didn’t get loaded by the end of their 10-hour shift, they would shut the truck off and go home for the day… which would mean tomorrow morning before they could ship any of that day’s orders they had to first work to fill the trucks that didn’t get done the night before. Hopefully, that didn’t cascade into them not being able to ship out that day’s orders… and on and on. When the SE returned home, he immediately told the CIO that the picker application, an application he had said was non-mission critical only days before, was now one of the most important applications the company had and that they needed to work to provide a higher level of redundancy and SLA for that app.

In a cloud environment, both internal and external, the same pragmatism needs to take place whenever a new product is deployed or an old product is brought into the cloud. When building the service catalog, it is imperative that each application be looked at and understood for how it fits into the organizational goals, then there needs to be understanding from the line of business owners and IT on what business process are used to meet those goals, then on what IT services those business processes rely on, then on what service levels need to be met in order for those services to be maintained, and finally what technologies need to be in place to provide the execution of the service levels. Once you have a fully functional service catalog you are well on your way to having a cloud environment. Then, it will also be possible to determine if you have the ability to leverage third-party cloud providers or if you need to keep your cloud internal in order to meet the service needs of the organization.

“What is your perception of living in the cloud?”

In the engagement I was on last week, one of the most thought-provoking questions we presented to the Sr. IT management team was: “What is your perception of living in the cloud and where do you think your challenges/improvements will be?” The feedback that we got from them made me think that they were not unique in their viewpoints and that most customers that are on the ‘roadmap to the cloud’ are going to face very similar challenges. I wanted to capture their comments and share them because they are a fairly progressive organization and the information we gathered was very insightful into what was ‘front of mind’ for them.

In order to protect their identity, I can only provide a very high level of what the organization looked like, but effectively, they were a centralized IT organization that provides services to multiple subsidiary business groups. One of their primary goals for their future development is that they must be able to provide better services to their customers than Google/Amazon/Microsoft can provide. I am working on a blog post on that entire subject alone because in this space I am working in, this is the primary theme we are getting from every company we engage with. They were fully aware that they were already being measured by their subsidiaries on their ability to deliver services vs what is available in the market. One of their quotes was: “..we believe every one of our customers will say that we provide a very high quality of service, however, they will follow that by saying they wish we would deliver them faster.” Again, that statement alone could drive an entire blog post regarding IT transformation and what it really means to an organization, but it was a strong driver in their response to the question at hand. Each person in the group participated in the feedback and their answers are broken out below.

VP of Open Systems Infrastructure

“I think the biggest challenge that I perceive is changing our customer’s mindset about their infrastructure. Today, we get requests for a specific server with specific processors and memory in them and specific storage they think is the best to support their environment. The challenge is to get them to change their perception of how they receive IT services. They should not care what the infrastructure is behind their services because they should be abstracted into a service catalog. If the SLA's are met, the infrastructure should not matter to them. The other challenge is that some of the subsidiaries will have no issues with Multi-tenancy but others will not accept this without putting up a significant fight. The biggest benefits in my perception are better resource utilization, lower time to market, and being able to get to a service-catalog based environment with pre-defined tiers of service. Also, today our service request process is painful and slow and we need to get to a point where service request and delivery is on par with public cloud providers. Another big component is that none of the subsidiaries pay for their services from us today, so they do not understand the dynamics of what a request entails. We are looking forward to cloud in order to improve transparency through the use of a service catalog. Change and capacity management are other pieces that we need to improve and are hoping to get once we improve our service delivery model.”

VP of Application Development

“Standardization is still the biggest challenge for us currently and I fear it will be as we move forward into cloud. Also, we are challenged with how we will interoperate with other cloud service providers such as MS Azure. Getting all the subsidiaries to the same development platform seems like an impossible task. Also, we have to be fairly flexible in the ability to accommodate and integrate our COTS (Commercial Off The Shelf) applications.”

Chief Security Officer

“Our challenges today exist because there is no set security standard for the entire organizational structure. Many subsidiaries that need higher security do not have it and many that think they need it do not. Post transformation, it would be much better for us because we could enforce the right levels of security for each subsidiary. Another strong benefit of cloud is that we will centralize and consolidate our security audit functions. This is a process today that takes an enormous amount of time and is highly inefficient. We have an education challenge for the subs in that we need to get them over the fear of multi-tenancy. Today, many of them will throw compliance interpretation in our face to resist being multi-tenant. I know that we have the ability to amend most of our compliance SLA’s and I feel we could significantly reduce our exposure to risk if we centralize our security model. One challenge we will face is that we will have to adapt our application certification process and procedures to fit our cloud methodology. Today, we have a very manual process and not much standardization. In the cloud model, we would be able to enforce application development standards that would allow us to certify applications much more efficiently.  

Chief Technology Officer

“We want to establish that we are the premiere service provider for our subsidiaries. We feel we can provide better services than any public cloud provider if we can change our core strategy for IT service delivery. It is important for us to map our IT services to the business. We need to educate our customers on what we offer and how we can help them and then tie a business justification for our services. If one of our customers feels they need to develop their applications to Azure, then they need to provide a strong business justification for why they are going to do that and how they are going to own their security and maintenance of those services. Another part of cloud is that we need to shut down datacenters that are not needed. Compute should be done centrally with a D.R. service offering that we can provide if a customer requests that level of service. A huge benefit of a service catalog and a cloud strategy is that we will be able to modify behavior by changing the way we provision services and charging those services back to the customer. We believe there won’t be many customers who will choose to implement a standalone dedicated physical infrastructure when they see that the time to market is 5-10X longer than our cloud-based offerings and when they see the costs associated with them. Capacity on demand is another benefit of cloud. Our ability to ramp services up and down and only charge business units for what they consume would be a substantial improvement over where we are today. Our goal is to take the quality of service we provide today and accelerate the time to market while reducing our operating costs and increasing our overall IT efficiency through the adoption of a cloud strategy.

My Conclusion

One of the biggest challenges in IT today is being able to quantify what people think “The Cloud” really is. I thought this IT organization had a very concise perspective on the key tenants of the cloud philosophy. Granted, this may just be because their understanding matches very closely what my own understanding of the industry is, but I still think they are on to something. Ultimately, they are very aware of the challenges ahead of them that exist in changing the people and process mechanics of their environment but they are working towards unifying their IT teams to work towards the common goals listed above. Much of their role in the near future is to educate their customers on the importance of pursuing this IT transformation motion and selling them on the value that will be gained through the effort. I definitely feel they have a large amount of strategic opportunities available to them and I think there are still some areas they have not even realized are there. For example, the application development team still hasn’t looked at next-generation application development frameworks and done any assessment of what the impact could be for them if they adopted them. They are, however, much farther down the path than many companies and have a strong sense of purpose as they move their projects forward.

My First ‘Vision’ Engagement.

Last week I got to experience my first ‘Vision’ workshop from start to finish. I have to say, I am completely blown away to see this group in action. I rode along for mentoring and thoroughly enjoyed seeing how many ‘big brains’ we had engaged from VMware’s perspective and how much interaction and participation we were able to get from the company we were helping. The magic of the way we engage seems to be the ability to get CIO’s and CTO’s and IT VP’s all in the room at the same time discussing strategic initiatives that the company needs to achieve in the next 3-5 years. I have personally been on over a thousand customer meetings in the last 5-6 years and I have never been in a meeting that had as many strategic IT management people as these meetings had last week.

Another key aspect of what I was impressed by was the efficiency of the engagement. I asked my mentor: “How long do we typically take to get a full strategy together… 3 months? 6 months?” He laughed and said: “Our goal is to get every engagement we do completely executed in under 30 days.” I did a double-spit-take on that one. How could we possibly understand the organizations’ goals, talk to all groups within IT that are involved, assess where they are today and do a gap analysis, get with the SME’s and discuss possible solutions, put together a roadmap, tie the roadmap to specific work packages, discuss with PSO organization to align resources and project scopes, put together a full product list with pricing, generate a proposal, and get it all presented back to the customer in less than 30 days?!?! Once again, I was greeted with a smile and given the look of: “Watch this, rook… we are about to blow your mind.” Indeed they did.

Day 1: We all arrived at a hotel near the customer the day before the first meeting. I watched as people starting rolling into the conference room and introduced myself. I was shocked as more and more people came into the room. At the end of the flow, we had anyone and everyone from VMware that would be involved with the account. The TAM, the account manager, the SE, the SSM, SME’s from each discipline (VEUC, Cloud, Etc), and then the vision team members. (The BSS and BSA team.) I have to say, I was extremely impressed. I have never really been in a position where getting all these people in the same room at the same time would be possible. Then, the magic happened. Reid Engstrom was my BSS on this account. (Yes, I am the BSA on this account, with my mentor John Steiner helping whip me into fighting form.) Reid is to me as Gandalf was to Frodo. He is absolutely a candidate for ‘the most interesting man alive’ and has been in amazing roles in this industry. His previous experience was as CTO of one of my favorite companies… Harley-Davidson. He immediately took the reins and kicked things off. With a wisdom born of experience he laid-out the agenda for the following two days, explained the process we would go through, cleared-up any questions and then instructed each member of the team what their role was and made sure they were enabled to deliver their piece of the puzzle. We went through together, as a team, each persons’ presentation and polished them up. Each person gave feedback and was engaged. It was the most collaboration I have witnessed. Amazingly… it was done in about 4 hours.

Day 2: We arrive onsite at the customer at 8:00AM and start to sit down. All the VMware team came in and then it was clear that there were just as many people from the companies’ IT department. As we started introductions, that is when I was bewildered again. CIO, CTO, VP of App Dev, VP of Infrastructure, VP of Mainframe, CSO. They were all there! At the same time! Then, we listened as the CIO level-set us on their top goals, challenges, etc and we went through our prepared presenters… gathering absolutely amazing amounts of information from the IT executive team. At one point, I actually threw a question out expecting that everyone would vehemently disagree. I asked: “Based on what you have told me, do you think you are at risk of having ‘Shadow IT’ in your environment?” After I clarified what Shadow IT is, they all answered: “Oh yeah… I am SURE we do! But, today, we have no way of knowing… they don’t need approval for anything under $100K..” I sat there stunned. They were being completely transparent. Completely divulging information to us. And then it dawned on me….

“WE were doing THEM a favor. And they were grateful for it.”

No one had ever engaged them in this manner before. No one had come to them to bring the forces to bear to affect real transformative change in their environment. They were getting extreme pressure from the industry competitors (Amazon, Google, Teremark[Verizon]) and needed help. So much so, they spent 4 hours with us to give us as much data as we needed. 4 HOURS! ;-p

Day 2 – Part 2: After our meeting with the team, Reid gathered us all up back at the hotel conference room and we began our debriefing. We went through what we heard. Did a rough swipe of a SWOT analysis. Identified risks to them and us. We gathered everyones’ notes. (Reid actually has pre-built notes pages designed in order to ensure people capture relevant data in the workshop… awesome.) After gathering this, we then shifted gears to identify next steps and give people ownership of follow-ups.

All in all, after only two days of meetings it was amazing to see the amount of work that we completed in a fully collaborative fashion. It truly felt like 1+1 = 3 when we left. We are still in the middle of this, but I am extremely certain this engagement will be completed within 30 days. Amazing.

What is the ‘Vision Program’?

It is my 4^th full week of getting indoctrinated into the Vision Program within VMware and I wanted to do a write-up to explain what it is we do, how it is we do it, and why we do it. Effectively, we are a new venture for VMware that is focused heavily on changing the way the VMware field sales team works with strategic accounts through the use of ‘value engineering’. Our mission statement is as follows:

To collaboratively develop an actionable Cloud strategy, business case, and execution plan through clear alignment to the customers’ goals and objectives.

The key differentiator for our group is that we are strategic in nature and not transactional. Our execution is focused on the top strategic VMware customers and the team we have is full of senior-level people who have been in strategic leadership roles before within corporate IT. Within the group, we have two distinct roles: Business Solutions Strategists (BSS) and Business Solutions Architects (BSA). The BSS group are folks that used to be CIO’s and CTO’s of companies themselves and understand people, process, and politics of corporate IT organizations and how to engage with senior management teams to define the strategic direction of their own companies. The BSA’s are the senior architects that define the strategic technology roadmap for the organization and ties together the products and services that are needed to transform the IT organization. The BSA aligns products and services to a roadmap that breaks out strategic IT initiatives and then outlines the work stream and the work packages for each.

There are multiple key strategies that we execute on today. Some examples of these are: Optimizing your virtual datacenter, Enterprise management, Application modernization, Virtual End User Computing, and Infrastructure as a Service (IaaS) to name a few. Each of these has a distinct work stream associated with it and within each work stream we define a work package to complete each step of the stream.

The work streams break out all the logical steps for how an organization needs to address each strategic initiative and assigns work that needs to be done for each. We take these and shift each tab out into a timeline that meets the project goals of the teams we interview during our engagement. The work package is a visual way for us (the BSA’s) to assign products and services (VMware or other) to each work stream component. This acts as a high-level roadmap for how the organization will transition their environment from where they are to where they want to be.

Using these tools, the Vision Program provides strategic guidance to large customers to get them to realize the goal of VMware to become an IT transformation company in the future. One of the best quotes that I have heard was: “You are the leading edge of the ‘Cloud Company’.” Our goal is to tie together the field sales teams, the SME’s, the PSO team, and the customer’s entire IT management team and provide an actionable approach to moving to a cloud IT organization. We truly believe this is the future of VMware as a company and how we will grow to being seen as what we truly are, rather than just a virtualization company.

“vFabric: It has, and always will be, about the apps.”

In my past life I was focused primarily on hardware and software sales for building internal private clouds. The thing that always seemed to be a limiting factor for us was: “Well, what about the apps? Are they designed to leverage the power of cloud computing or are we just trying to architect hardware to overcome software limitations?” One of my favorite quotes that I have heard is:

“You can’t hardware yourself out of a problem that you softwared yourself into.”

In this same vein, I always felt that we needed to start these discussions more strategically at the app layer and discuss this, at the very least, in parallel to infrastructure. Much of the power of cloud adoption comes from the ability of an organization to leverage PaaS and SaaS as part of their ongoing IT operations transformations.  They also need to have more strategic discussions around how they will manage their application environment once they move into cloud services; both internally and externally.

In my new role my focus is now to have these strategic discussions with large enterprise customers and help provide a roadmap for how they can start transforming their IT operations to get them best prepared for the future of cloud computing. The most enjoyment I have had thus far was getting up to speed on the vFabric family of products and services that VMware is currently offering and will be offering in the future. What is amazing to me is how much of the story is actively available today and I definitely do not feel that the industry even has a good grasp (as a whole) of what VMware’s current capabilities and future capabilities are/will be. What VMware has put together is an entire framework for developing applications and for deploying them into both internal and external cloud infrastructures and then provided a way to monitor and manage them holistically. You can learn more about the framework here: http://www.vmware.com/products/vfabric/overview.html

I could do entire posts on each component of the vFabric platform, but the theme of this post is not so much about the technologies themselves, but the fact that most VMware customers are focused on only 2 of the 3 layers of VMware’s solution sets. Many of you are familiar with these, but to restate, VMware has a cloud vision in 3 major layers: 1) End User Computing (DaaS), 2) Cloud Infrastructure and Management (IaaS), and 3) Cloud Application Platform (PaaS). In my humble opinion, from my industry experience most companies are already well along the adoption curve of IaaS and overall infrastructure virtualization, they are starting to look pretty heavily at end user computing, but they have not even started to look at how to change the application layer of their environment. Most organizations have not strategically assessed the impact they could make on their environment by looking there first, rather than as an afterthought. In my mind, it seems logical that we should start there by understanding application interdependencies, change elements, and business impact of each before we would try to architect infrastructure or end user connectivity. I know there are significant political implications of this at most organizations but as with everything cloud, you have to tear down the isolation and work together to effectively drive efficiencies within an organization or you cannot be successful.

And remember… to the business…it is now, and always has been… all about the apps.

So today marks my first day at my new career at VMware! I am very excited. I will be a Business Solutions Architect in the Vision Program - Value Engineering team. I am relocating the family to Jacksonville, FL and will be primarily covering the South East and Mid-Atlantic regions of the US.


I got into blogging a bit at my last organization and decided to brush the dust of this blog and perhaps use this as my portal of information that I come across in my travels. Check back soon for more info. Follow me on Twitter @ericledyard.