Why consider cloud?

One of the things customer’s need to focus on as they consider moving to the cloud is a strong look at their application portfolio. There are two things that cost money in the IT world. The capacity required to run 100% of the solution (that is the cost of on-premise reality) and the reality of DR. I cannot tell you how many customers have started their cloud migrations assuming that they are truly doing the right things.

First and foremost the only fixed costs you have for cloud are storage and connectivity. No matter what you do, storage will cost you. That means if you put something in the cloud, be it code, data or applications you will pay for the storage. You will also pay for the bandwidth out, as in once connected the information your applications is pushing to the user. In this reality, you have an opportunity. Knowing you won’t change the storage story, you can change the computing story. Compute, being the active computer sessions or VM’s (Virtual Machines) that your organization is running.

Your organization can take a hard look at the reality of your solutions. How long during a standard workday do you need your system operational? If you are truly a 24 hour a day, seven days a week operation you can still save money with the cloud, just not on your initial solution costs. The first thing to do is to evaluate your solution compared to the reality of performance. Many organizations assume that cloud is cheaper and faster. It may never be either if you don’t evaluate what is required. If your solution is 24 by seven what percentage of the solution is required the entire day? Can you throttle down a bit and reduce the overall cost by running fewer servers during the slower hours of the day? Remember the cost of on-premise is always the reality of buying and maintaining 100% of the hardware required for the maximum usage periods. With cloud, you can begin to evaluate the true requirements for the applications. If you max user period is 11 am to 3 pm EDT, when you have 10,000 concurrent users and your lowest period is 11 pm to 3 am EDT where you have 1,000 connected users, you can save money. You only have to pay for 1/10th of your maximum cost for ¼ of the day.

The reality of true consumption is the first value statement of cloud computing!

.doc

wondering about technology

What if we designed systems to succeed?

There are some technologies that I wonder why they aren’t taking off. In part because I find them extremely valuable and something that I use nearly every day. Solar power is one that I find interesting. Solar power to me seems like it is lagging in the US. Part one is the reality of Home Owners Associations and the perception that a solar array reduces the value of a house. I am not sure why free or extremely reduced price power that you don’t have to worry about brownouts and other things like that would make your house worth less. I would think, at least I would personally, that every buyer would demand solar on the house.

The other funky reality that I wonder about lately is that of safety in cars. There are some safety ratings for cars today, 5-star crash ratings, etc. I would like to see the airbag safety listed as well. If the airbag deploys can you get out of the car? If the airbag deploys will you be in better shape than if it did deploy and so on? It is a very similar ask that I have for cloud service providers who run around with cost value equations and performance numbers that are only relevant to the person that put them together.

I wrote about a cloud calculator in my last book published nearly two years ago now. (It may be time to start on a new book). The cloud calculator was designed to help customers understand not only what they wanted to build and deploy but also to understand the impact of a specific cloud provider on the solution they were considering. Not all networks are created the same. Not all solutions operate the same way.

I liken the cloud calculator to building a motorcycle. The most important thing for a motorcycle is that it be balanced. If it isn’t balanced, you are more like to “put the bike” down. That isn’t a good thing. You could, however, take the motor and make the front tires much wider. What this would do is make the big more stable. It would also make it much harder to turn the bike right or left. I always tell customers to make sure you know what motorcycle you want first, then, what do you need that motorcycle to include. That is what the cloud calculator system does for you. Like I said you can find it in my current book, Operating beyond your borders.

Another way you can look at the analogy is the bridge in the cover picture. If you design a boat that is too tall or too wide, you limit the ability to the boat to move. Boats that can’t move have to be docked, but they also cannot move in the case of an emergency. If your goal is movement, that isn’t a good design.

Seamless is only a word if it applies to clothing. It never applies to migrations.

.doc

technology strategist

Why is it so hard to just move to cloud computing?

One of the things I often talk about with customers is the reality of buy vs. build. A conversation that expands beyond simply just infrastructure, applications and does you build your own or buy what is out there? Cloud computing makes this into a what do I want to manage conversation as well.  In the conversation, I usually start off with the email conversation. When I was first in IT, now 30 plus years ago, many companies had multiple mail systems, and some even built their own. With the rise of SMTP mail, most companies that wrote their systems moved to COTs or FOSS mail systems. Why? Because the cost of keeping up with the Jones made building your mail system far too expensive.

Today, cloud computing in the infrastructure space makes it difficult in the short run to decide and determine what you are going to do. The reality of advertising intercedes at this time, first off you are not going to save 20% moving to the cloud. The reality of the  20% savings mark is NOT that it is spread over all the customers of a provider and is most likely a watermark or high level. It is not an accumulated average across all customers. The other things you have to be aware of is the reality of migration. Migrations are painful at times. They are frustrating at other times. Most organizations struggle with migrations, in part because they buy into the expert culture. Experts are focused on a specific point in time solutions, and in a migration, you need to have someone that can step back and say “there are other ways to solve this problem.”

What was once an easy conversation is now much more complex and much harder to have. Effectively you need to have the migration conversation first, then the cloud conversation. Effectively the new world order should be talk to the person with migration scars first, then figure out if you can make the new world order work for your organization.

The tools, concepts and models as well as the service calculator system are detailed in my book.

.doc

technologist

The impacting IoT Devices on Hybrid Cloud…

I won’t spend any column space explaining either of the two terms. They are well explained all over the internet and on Cloudtweaks, much better than I could explain them. As organizations move towards cloud computing more and more of them are implementing a Hybrid Cloud solution. More and more of them are deploying objects that are more Internet of Things than cloud as well.

As far as opening gambits go, this isn’t a radical opening statement. It is simply as an assessment of the current state of business globally. Lots of government and commercial movement to the cloud, mostly to reduce cost. Frankly many governments and commercial deployments of IoT Devices. Pick an industry and notice the change in that industry in just the past five or six years. One that I have been watching the change in just the last year is Real Estate. The concept of 3d models of the home you are considering existed before, but now they are everywhere. Where once a posting with 3, four even five pictures of the house was normal now you have 30, 40 and even music!

IoT is going to modify Hybrid cloud though. How? Simply it is going to force government agencies, and commercial companies are going to continue down the path of building solutions around IoT devices. Over time they are going to also build gateways for the IoT data. That will be the change for the Hybrid Cloud. Today many organizations are considering have multiple clouds in their portfolio. But as you move forward, you most likely will only have one IoT gateway. All four of the major cloud players have released or announced their IoT gateway solution. This will force many companies to move to the one provider model when we are talking IoT, you don’t have a lot of capacity for changing the device.

The reality is IoT will change cloud, forever. Starting with the easy things, like consuming all the bandwidth that exists today. As we improve and make devices smarter, things will improve. But for the short term get ready to see a buffering screen when you play your Friday night Netflix movies!

.doc

and if you don’t watch Netflix on Friday night what happens?

My cloud won’t start…

I can honestly say that the most annoying thing for me is people telling me things that I know aren’t true, and stating them as facts. In part, I realize that this is a much larger issue than myself, it is one that plays out what seems like every day in the media.

In particular, this bothers me from a reality perspective. Many years ago, I developed an application migration framework for a customer. I then took that framework and built a process around it (you can still find that process on Microsoft.com – it was called AAEP or the application analysis Envisioning process, yes I picked the name, yes I am good at names often but not that time).

The process was a mix of Enterprise Architecture (building a capability framework) and creating a technology implementation roadmap.  It was most effective when you mixed not only the art of the possible but peaked over the horizon to see what was possible.

I bring this up because of the reality of cloud. I talked yesterday about the fact that cloud was an answer and it remains so. It is not, however, the answer. The reason it isn’t the answer is what drove the creation many years ago of the AAEP system. I created it for Lotus Domino migrations, and it was extremely effective. But I quickly realized that many pieces of it fit any migration.

The difference between what we expect, what we have, and ultimately what we need is a critical juncture for end users and IT. What any one cloud provider tells you, is something you have to evaluate carefully. It is their information that presents what is happening in their specific instance and therefore has a limited impact on your situation. The value of information has it, evaluating it and using it to create a report that combines what you know, what you have and what you need into a roadmap.

To get to the new DevOps world, you have to know and understand the world you are coming from!

.doc

Dreamer

Cloud Computing is an answer, not the answer…

One of my favorite things in life is thinking about the art of the possible. I don’t always do that from the point of reference of what is now here. I do it, from the combination of what is coming and what is here to arrive at a view of the horizon.

Why? I learned that many years ago from my second IT boss. He was a former GE professional and came in to manage our helpdesk. He taught me to look, not at what I saw, but what could be there if I waited. I have never forgotten that. We are still, to this day friends because of that.

Plus he was an awesome boss. I won’t mention his name because that isn’t fair to him. But he was a great boss!

In looking over the technology window right now, I see some interesting trends. One of them is the reality of cloud computing. In particular the rise of Hybrid (and the drifting off of Cloud Brokers). The economic reality of cloud is saving the organization money. The technical reality of cloud is reducing the churn within the IT organization. If, and it is a big if, we take the long-term view of the cloud solution for our company we see that it pushes us in two directions. We can become a more agile and responsive IT organization at the same time also able to reduce the cost of solutions.

This tipping point though is one that most companies continue to seek and not find. Proactive IT is a logical move for cloud computing, cost savings is a logical outcome. Both, however, are elusive, because effectively we still have the same problems in the cloud that we have on premise. It is the new risk of cloud, the rise of expert cultures that drive towards the right answer.

Cloud is here. Proactive IT has value. Cost savings are good. But reality is, cloud is not the only thing you have to do. Cloud is not the only answer. It is an answer but not the answer.

To paraphrase Star Wars, “These are not the answers you are looking for…”

.doc

technologist.

The future of cloud, as the integration point for IoT devices…

Right now still cleaning my house as far as getting rid of old items I don’t use. I find I have quite a few including three different IoT device kids. I am using the IoT kits, for now, so I won’t get rid of them, I am extremely interested in the direction IoT will take in the next couple of years, so learning is critical.

I shared my initial thoughts on the cloud computing transition to Hybrid from Cloud Broker. One of the things that interest me about the cloud computing short term is the reality of IoT. Everyone is always talking about what IoT is. Billions of devices chattering away. 2, 3 and potentially four times the amount of current bandwidth consumed in the next 50 to 60 months.

In part because we don’t have smart sensors yet. Sensors that can understand when they have to need and shouldn’t be reporting. Those sensors while possible and certainly being considered haven’t hit the market as of yet.

This relates to cloud as the transportation and communication layers within the IoT framework. As more and more of the Cloud Providers begin adding IoT gateways, we will see cloud become the IoT integration point. That will, of course, beget the next big mess.

My system is built to connect to the Azure IoT gateway, but I am on AWS. My system is built for the Bluemix (IBM) IoT gateway, but I am running on Azure. And so on, a mess because the gateways today don’t have standards to adhere to, and well we all know that standards as guidelines aren’t often followed anyway. Or worse, everyone interprets what is is to be something other than what the other is’s are.

The reality is that moving VM’s from one cloud service provider to another can be painful. Sure, you can in most cases easily move from on-premise to the cloud using vendor created tools but cloud to cloud, not so much (at least not without a lot of planning).

IoT is going to bring this issue to the forefront fairly soon. Why? When you are doubling, tripling and so on the amount of data generated by the IoT world, you need to be able to effectively move it to your organization.

Data drives change, and in this case the change will be interesting. IoT may be the next wave what what’s new in cloud, by making the cloud providers work together in offering a unified cloud IoT gateway structure.

.doc

Technologist!