What are some of the difficulties you may face when developing an enterprise cloud strategy?

Pitfalls of cloud migration for enterprise companies.

Like other large projects, cloud migration in large enterprises brings with it inherent challenges. Which ones do we encounter most often?

Price wars

In large companies, there are still internal tensions between supporters and detractors of cloudification. For example, one side claims that enterprise infrastructure is inflexible, while the other claims that the cloud is expensive. Cost shootouts in particular are common, but they usually have no winners. Why?  

Primarily because few companies are able to allocate proportionate infrastructure costs to individual applications based on actual usage. Moreover, they often don't even know the exact amount of total infrastructure costs. So how many times is it impossible to compare how much more or less expensive it would be to operate the IT infrastructure after cloudification?

It is common that the cost of running an application that has migrated to a trivial option can significantly exceed the original one, but again, it can be argued against such a conclusion that the chosen migration option was not the right one.

We can't, so we don't use

"If an organization lacks sufficient expertise in IT development or DevOps, then using the cloud may not be feasible. This often results in suboptimal solutions being employed."

IaaC, microservices, clever ways of dynamically sizing infrastructure, and monitoring service usage - these are just examples of approaches that unlock cloud efficiency, but someone has to know how to use them. This is especially true for people in the development and operations, or DevOps team and their level of seniority. It doesn't mean that everyone has to be an expert in the field, but they should be shielded by expert mentoring and follow best - practices in both architecture and technical design.

Lack of knowledge also often creates resistance to change on the part of employees. Therefore, we believe it is essential to provide related education and mentoring to internal teams or effective support services in architecture, technical design, and DevOps when cloudification is implemented.

Application portfolio 

A big company means many applications. So where to start? How to evaluate the suitability of an application for migration? What migration options are there to suggest for it? And can it be combined with a business upgrade? How much will it cost? Can it all be done in a reasonable amount of time? Migration brings with it a number of questions that may not be easy to answer.

Application assessment can be approached through questionnaires, personal surveys, or automated application analysis systems. However, it is difficult to assess an entire application portfolio at once, let alone migrate it.

Therefore, we recommend a combination of automated and manual approaches as well as a targeted focus, for example on applications with a certain degree of business or technical connection. This increases the positive effect of the transformation. However, this is not the only selection criterion.

So why do it?

A serious argument for cloudification can be made on the assumption that the steady expansion of cloud offerings will create pressure on the other side in the long run - and gradually reduce the scope of on-premise services. From this perspective, the cloud can be thought of as a possible forced technological change of the future.

Recent global events also offer the opposite - somewhat catastrophic - interpretation. In her view, the global business chain is so vulnerable that we only have real control over what we 'carry'. 

So what to do about it? The golden grail of cloud migration seems to be finding the right reason - for or against it. A company that knows its strategy and finds a strong enough link between it and cloud services is likely to find it worth going in that direction.

It is very difficult to build a universal and robust business case that would magically justify a company's move to the cloud. Too many items in such a study depend on the value of each aspect to a particular company.

For example, resources tied up in proprietary hardware can be a serious negative for the business model of dynamic service testing. Conversely, they are not a disadvantage for a company (or its domain) with stable services.

In the end, it is not just price, flexibility, security, future support, geolocation, fashionability, legislation, or any other argument in isolation that will decide, but an assessment of all aspects of moving to the cloud in the context of a particular company's business model.

If you are considering a migration to the cloud, Trask is ready to help you both make a well thought out decision and implement it effectively.

‍

Ales Kulhanek

Project Manager, Trask

He has worked for large companies in the automotive and finance industries, where he has adopted cloud solutions through development or migration. Alongside Trask's top experts, he has assessed the client's application portfolio, participated in the migration itself, and helped create the related methodology and business case for the migration process.