When cloud project managers are asked how the transformation to the cloud has been going, they usually give technical answers: they have succeeded in
- closing x data centers
- taking y servers out of operation
- shifting z workloads or
- the time needed for a specific task has been reduced by x percentage points.
Although these metrics are important, they don’t tell the whole story. Cloud transformations are not usually an end in themselves, but mostly pursue very specific objectives.
Employees: a key factor
In fact, those objectives are often very big challenges. Experience from completed cloud initiatives shows that, on completion of migration, it has been possible to shut down around 30% of the data center applications on average. It also shows that 70% of cloud initiatives or shutdowns of equipment fail because the employees have been “forgotten”. Old systems cannot be replaced because it has not been possible to take employees along on the journey to the cloud. The reasons for this are varied and range from active to passive boycotts. Technical reasons, triggers and dependencies turn into insurmountable obstacles at the last minute.
Below are three questions that have a crucial influence on the design and implementation of cloud initiatives and can affect them significantly:
What objectives is your company pursuing?
What business objectives does your company have (e.g. price leadership, quality leadership) and what are the limits? What drivers result from this for the cloud?
- For many companies, costs are the most important or the main determining factor.
- If a company wants to be a competitive, cost-effective provider, it must create a mix that delivers value to the market and also keeps its own costs low.
- If a company has other objectives – short time to market, global expansion, etc. – it may make other decisions in relation to the design of its cloud program.
Naturally, in practice companies have to find their own optimal combination of many parameters – within defined limits, of course. Including those it imposes on itself. In terms of a cloud initiative, for example:
- Security and compliance: Are there security regulations, compliance rules, country-specific or government regulations in relation to data limits and the choice of locations? Consider questions of data sovereignty between countries and even governments. The effects of placing data in a country with heavily regulated data protection could have a significant impact on the success of the program and possibly of the company.
- Culture and organization: How qualified is the workforce and to what extent is the
corporate culture ready to change from one platform to another?
- Processes and governance:
- How disciplined is your company in relation to processes and governance?
- IS IT able to set requirements and take decisions for the long term and compel developers and communities to keep to those decisions, or does the process take place ad hoc?
If the teams are not coordinated, whether for technical or political reasons, the speed of provisioning will be much slower.
Understanding of the business requirements and the limits that are imposed on an organization increases clarity about how companies should structure their interdisciplinary teams and the associated mix of resources.
What does the right use of technology look like?
Decisions about complex, mixed forms of platforms and services are ultimately taken by employees after intensive analysis and long debates. But technical tools can also play a part in this process and bring a certain order into the often chaotic situation.
- Are the applications cloud-compatible?
- Which applications should be migrated first?
- What is the right mix of hybrid cloud platforms for my applications?
- Which applications require a change of code before switching to the cloud?
- What is the fastest way to plan migration?
- How much confidence do I have in the analysis?
- What interdependencies does my application have?
IT teams must then prepare their application migration and programs for any cloud platform that offers the right mix in terms of considerations such as costs, flexibility, security, risk, data protection and a wide range of other factors.
What should be migrated and how, what types of migration should be used?
Rehosting: This migration type is generally referred to as “lift and shift”. This is system-to-system migration of the application and data onto the cloud platform.
Replatforming: With this type of migration, the application servers and software can be run easily on the new platform. Typically it’s just a matter of changing the environment parameters here, but not the code, e.g.:
- Upgrading the operating system and/or database version
- Significant DNS and network changes
- Changes to the INI and configuration file
Refactoring: This type of migration is the most complex and costly, as changes have to be made to the application at code level. The team must have specialist knowledge in the areas of cloud services, security and infrastructure.
Retiring: Shutting down applications – so-called retiring – is very easy. Moreover, applications are frequently run in data centers that are maintained only to meet compliance requirements. These applications should also be examined appropriately and can often be replicated in the cloud with software, which would also make them candidates under this heading.
Replacing: Replacement of applications is more complex and depends very heavily on the business area and those responsible for the applications.
Retaining: Retaining applications essentially involves scenarios at the end of their service life and your plans to shut down the platform. The challenge does not lie in how to retain applications but in how cloud-based applications interact with the applications that are retained. The retained applications are frequently core systems of the company.
The human factor
The cloud makes companies more agile. But switching to the cloud undeniably brings big changes.
Things are done completely differently in the cloud. Employees therefore have to learn new technical skills. The change from a cascading project management model to an agile, continuous scrum-DevOps method requires a completely new approach.
Silos are broken down and employees then have to collaborate in an interdisciplinary team with colleagues they do not know and learn new skills.
Many initiatives for change fail because employees do not know what they should stop doing, what they should start doing immediately and what they should continue to do.
The decisions that you take now will affect the company in the coming years. The composition of the right mix of resources is crucial to getting your company on the path to future success.
“Taking a good look at your employees”, integrating them effectively and selecting the right organization are among the key factors to success. Only if your employees are really “on board” and behind the cloud transformation will it actually turn out to be a long-term success!