Archive

5 elements for successfully deploying cloud solutions

Share this post:

Currently, there’s a lot of focus in the marketplace on how to successfully implement, use and manage cloud technology, migrate applications and systems to the cloud and use the cloud to solve all the problems and issues in your organization, company, and your life in general. Supposedly, cloud solves everything.

Back in 2010, I created a list of characteristics of a successful cloud deployment. The focus was on delivering a viable cloud infrastructure and migrating workloads to that model. I wanted to share this model and examine if anything has changed since 2010, especially as the use of cloud has changed; it’s now assumed that part of the approach will include using many different providers and many different offerings in what we call either hybrid information technology or hybrid cloud. Moving to the cloud needs to assume this heterogeneous deployment model.

In order to successfully use the cloud model, you need to ask yourself a couple of questions: What will get fixed, improved, enhanced or created if I use cloud? What problem or challenge am I trying to solve by using cloud?

As an enterprise architect and systems engineer, I’m always ready to solve complex technical challenges and deliver the solution to a client. But before I can do that, I need to know what problem I’m actually trying to solve and whether or not cloud makes sense. The “not” is key here, since I go in assuming cloud is not necessarily the best approach. More specifically, there are often parts of the cloud model that probably apply and others that may not apply. It just depends.

Focusing on the end state is the entry point for moving forward with cloud in the most effective manner. Defining a successful cloud deployment is all about focusing on what really matters. In thinking back through my experience designing and deploying these types of elastic or flexible IT infrastructures, I see the same pattern of factors that must be addressed in order to successfully design, implement, deploy and maintain a cloud environment.

There are a couple of assumptions I always have in mind about what a cloud model involves:

• Pay per use: This is a utility-based and consumption-based model in which you pay only for the technical or business services you use.

• Asset divestiture: Enterprises do not own the physical and logical assets; they just use them and divest those they already have. This holds true for on-premises, off-premises and hybrid models.

Given all this background, let’s review the core success themes that always need to be addressed when deploying a cloud solution.

cloud success graphic

Business and enterprise alignment

Moving to a consumption and utility based model is a transition away from pre-allocation of budget—which has some predictability—to an elastic cost model that is less predictable but can decrease overall costs. Is cloud a good fit for your company, enterprise or group? Is there a justification or reason for moving to a cloud model? What is the right model: on-premises, off-premises or hybrid?

Organizational and skills readiness

This is closely related to business and enterprise alignment, and is focused mostly on the organization’s ability to change in order to use cloud services in the most effective manner. One of the most interesting things to consider is the financial model for cloud usage, both in IT and in the lines of business. Quite often, maximum benefit from cloud is achieved when, among other characteristics, services are used only when they are needed and expense is only incurred when services are used (this is known as a “pay per drink” model).

If it’s true that usage of services will be variable, then how can a CIO continue to justify large annual budgets based on inflexible pricing? This is a key question, and there has been much analysis showing that incremental budgets actually result in lower overall operating costs.

Also, while it’s obviously important to focus on technical skills, it’s perhaps more important to know if the organization that will be using the cloud services has the people and roles in place to effectively use cloud to solve their key business challenges.

Automation and patterns

When starting a cloud design process, automation is usually assumed but not often immediately discussed. Typically, everyone just assumes everything will be fully automated. In reality, there are some processes, workflows or tasks that either do not lend themselves to automation (especially where there are continually evolving regulatory or compliance requirements) or those where the value of automation is unclear.

IBM is focused on software-defined environments to improve two aspects:

• Automation and re-configuration of the infrastructure, containers and connectivity in a hybrid model, which helps to support a multi-environment deployment model.

• Software-defined compliance, to allow policies, practices and security to be incorporated into a cloud deployment

Standards, policies and practices

We often speak about standards and standardization when discussing and implementing technology. We talk about the need for a common operating system, standardized hardware configurations with limited variability, limited connectivity options and the ability to change an existing deployment but only within the options available in a catalog.

Just as important (and maybe even more important) are the policies and, especially, practices in place for utilizing that technology. Often, these practices are not documented in the form of official policies, yet they typically have a huge impact when utilizing a cloud model. It’s true we have to make sure the various software and hardware components are sized properly, but in order for these systems to be compatible with an enterprise level environment, they need to be supported by governance and must be in alignment with policies and practices.

Identify workloads and definitions

The best practice for utilizing a cloud model is to focus on workloads and not on operating systems, hardware specifications or specific software products. It’s important to define the affinity groups where servers, applications, data, policies, practices, API capabilities and connectivity are in alignment as a single deployment pattern.

Combining this with common deployment patterns and a workload definition language are an important part of the approach to defining workloads and describing those workloads in a way that can be deployed. Recent developments in deploying software-defined environments (which are focused on infrastructure automation and optimization) and hybrid cloud platforms utilize workload definitions and patterns as core enablers.

I hope you found this post interesting and useful. If you’d like to discuss things further, please leave a comment below or reach out to me directly via Twitter, LinkedIn or my blog. And keep an eye out for my subsequent posts, in which I’ll review some specific points involved in validating the viability of a workload to run in a cloud model.

More stories

Why we added new map tools to Netcool

I had the opportunity to visit a number of telecommunications clients using IBM Netcool over the last year. We frequently discussed the benefits of have a geographically mapped view of topology. Not just because it was nice “eye candy” in the Network Operations Center (NOC), but because it gives an important geographically-based view of network […]

Continue reading

How to streamline continuous delivery through better auditing

IT managers, does this sound familiar? Just when everything is running smoothly, you encounter the release management process in place for upgrading business applications in the production environment. You get an error notification in one of the workflows running the release management process. It can be especially frustrating when the error is coming from the […]

Continue reading

Want to see the latest from WebSphere Liberty? Join our webcast

We just released the latest release of WebSphere Liberty, 16.0.0.4. It includes many new enhancements to its security, database management and overall performance. Interested in what’s new? Join our webcast on January 11, 2017. Why? Read on. I used to take time to reflect on the year behind me as the calendar year closed out, […]

Continue reading