Three detrimental mistakes companies are making in cloud cost management

The public cloud may seem quite new to many people, but in reality the public cloud as we know it has existed since 2006, when Amazon Web Services (AWS) launched its S3 and EC2 IaaS storage services. The first major organizations began moving their workloads to the public cloud in the early 2000s, before 2010. In the 10 years after migrating and running workloads in the cloud, many lessons will have been learned. However, it is surprising that some of these lessons learned are not yet applied today. The three main lessons learned that everyone who migrates to the public cloud should apply are the following:

• Don't just lift and move your workloads to the public cloud

• Cloud Management Requires New Practices

• The cloud experience shouldn't just be left to application teams

Lesson 1: Don't just lift and move your workloads to the public cloud

Lift and shift applications sound attractive at first: a lift and shift migration is relatively quick and cheap. Since the applications themselves are affected as little as possible during the migration, the risk of defects resulting from the migration is low. But then comes the drawback: the added value of the cloud isn't there. When lifting and moving applications, it also eliminates and displaces errors made in the past and the garbage generated by running the application for years in the data center. In addition, the public cloud is not just another data center. Design decisions that are a perfect fit for the data center often don't work in the cloud. IDC Metri estimates from experience that a lift and change migration gets the best of what you already have in the data center, except for a 30% higher cost.

A cloud migration must be thoroughly prepared, with an application-based approach rather than an infrastructure-based one. For each application, it must be determined if the application will be retired (phased out), retained (kept 'as is' in the data center), rehosted (migrated 'as is'), replatformed (reinstalled on a newly designed cloud platform without adapting it). source code), refactored (adapted and optimized for the cloud) or repurchased (purchased or built from scratch). Tools that offer source-code level scans can greatly improve the speed and quality of such inventory, thus helping to achieve a successful migration with real added value.

Lesson 2: Cloud Management Requires New Practices

A typical way of working when it comes to a data center is that the data center and the infrastructure within it are managed by an infrastructure department (which can also be a managed service provider). The infrastructure department acts like the king of its own castle, with a guardian who controls everything that enters the data center.

The ease of use of the public cloud makes it much easier for application teams to manage the entire application stack rather than just the software part. A tedious task such as installing and configuring infrastructure manually is being replaced by creating infrastructure through infrastructure scripts such as code and ready-made building blocks. In this way, the public cloud has become a driving force behind DevOps and the motto “you build it, you run it”.

But there is also a drawback here. Tearing down the castle wall and the associated bureaucracy between the application teams and the infrastructure department also means that the guardian is no longer there. With this, the natural control mechanism over the deployment of infrastructure and its associated costs also disappears. Organizations are experiencing that their cloud costs are getting out of control due to the lack of a control mechanism.

Controlling the cloud without discarding advantages such as flexibility and scalability requires new practices. Restoring the gatekeeper is a practice that IDC Metri regularly encounters, but this also reestablishes red tape and wastes the benefits of the cloud. The way forward here is to hold application teams responsible for their cloud costs, providing them with knowledge and control mechanisms, building security barriers through policies that restrict them from creating instances of excessively expensive resources, creating alerts that warn them when costs are getting out of control and actively activating helping them analyze and manage their costs.

Controlling the cloud without discarding advantages such as flexibility and scalability requires new practices. Restoring the gatekeeper is a practice that IDC Metri regularly encounters, but this also reestablishes red tape and wastes the benefits of the cloud. The way forward here is to hold application teams responsible for their cloud costs, providing them with knowledge and control mechanisms, building security barriers through policies that restrict them from creating instances of excessively expensive resources, creating alerts that warn them when costs are getting out of control and actively activating helping them analyze and manage their costs.

Lesson 3: The cloud experience should be bundled and not left solely in the hands of application
teams. Lesson two leads to lesson three. While it makes application teams responsible for managing their own cloud costs, they can't be expected to know all the ins and outs of the cloud. In-depth knowledge of the cloud must be grouped into a single entity of the organization: the Cloud Center of Excellence (CCoE). The CCoE acts both as an assistance dog and as a guard dog. The CCoE helps application teams with their design decisions to use the cloud optimally and use the right cloud services for the right task. The CCoE also monitors cloud costs at the organizational level and alerts application teams if they see costs getting out of control. In addition to helping and observing application teams, the CCoE also manages discount schemes, such as reserved instances and savings plans, at the organizational level. In addition, CCoE should drive more complex cost savings, such as license optimization, since license management seems to be a science in and of itself.

The critical success factors for the CCoE are an adequate mandate or support from senior management to prevent it from becoming a paper tiger and, of course, the availability of trained and competent team members. In addition, the CCoE should have an organizational and financial approach, rather than a purely technical one. The application teams that run their applications in the cloud have the technical knowledge; the CCoE enhances it by incorporating knowledge about billing, pricing models and cloud governance.

To conclude, the cloud offers many opportunities, but also many challenges. Ten years of widespread use of the public cloud have taught us valuable lessons about how to use the cloud in a way that actually moves us forward rather than staying where we are at a higher cost. By applying these lessons, the public cloud can be used optimally while maintaining control of spending.

Source: https://blogs.idc.com/2023/02/10/3-harmful-mistakes-companies-are-making-in-the-cloud-and-cloud-cost-management/