What is cloud operation

How much does the cloud really cost? A topic that preoccupies many CFOs and CTOs, as numerous studies show, including the IDG report "The 2020 IDG Cloud Computing Survey" or "Cloud Computing Trends: 2020 State of the Cloud" from Flexera. Despite its importance, the topic of cloud costs and benefit transparency is seldom addressed strategically from the start. Instead, the focus is often either on the technology or the pressure from the business area.

As is so often the case, new tools and startups are created almost every week that offer solutions in the area of ​​cost management and transparency in the cloud. But is another tool really the solution to these challenges? Or are there alternatives?

What are the costs behind cloud services?

When companies start cloud projects, costs are often seen as one of the deciding factors. After all, the cloud not only promises savings, but also full transparency in terms of costs: companies only pay for the exact resources they use.

In practice, however, it looks different, because the maximum transparency leads to a high level of complexity. Anyone who has ever had a cloud billing in their hands knows this all too well: € 0.30 here, € 2.80 there, € 710 for logging or € 48 for support - this level of detail often leads to the opposite of what is desired Simplicity and transparency. The real costs often only become apparent after months.

All providers, such as AWS or Azure, offer price calculators that list prices in a modern design and also make total cost of ownership (TCO) calculations. However, in order to use these tools successfully, you not only have to have in-depth know-how of each cloud service (with over 100 services), but also an exact planning of the individual resources that are to be rolled out. In theory, this may sound feasible at first, but in practice, adjustments are made again and again in the course of a cloud project. This agility is important in order to get the most out of the cloud infrastructure and to optimize costs.

Challenges

As experienced cloud experts, both in the conceptual and operational environment, we have identified some recurring patterns of the challenges companies face in cloud projects.

Cost management

Traditionally, the responsibility for cost management lies with purchasing, license / contract management or controlling. These areas have established processes and mechanisms, from procurement to contract negotiation and contract conclusion to license audits.

A different situation can be found in the area of ​​public cloud and software-as-a-service (SaaS). Every company has a large number of SaaS applications and different cloud providers in use. In addition, it is not easy to establish individual contractual structures with the large cloud providers such as Amazon, Microsoft or Google, as these companies set different standards than are usual in the German enterprise / medium-sized sector.

In addition to these challenges, there is also an unmanageable number of different cloud services, payment models, discount options and contract periods as well as various service level agreements and support services.

The providers also differ in the area of ​​cost management and bring their own tools with them, such as the AWS Cost Explorer, Azure cost management or the Google Cloud platform cost management.

In order to understand all these areas and to use them effectively for your own company, extensive knowledge is necessary.

Resource management

Cloud technologies form the basis for successful digitization and are drivers of innovation. However, the complexity of Infrastructure-as-a-Service (IaaS), Platform-as-a-Service (PaaS) and Software-as-a-Service (SaaS) is often underestimated and the technologies are not used optimally. Because every cloud provider offers a multitude of different services from which the right one for your own company must be selected.

This is crucial, because unfortunately companies still often opt for a pure lift & shift of their infrastructure to the cloud. Such a 1: 1 migration, however, leaves a lot of potential untapped and usually leads to higher costs.

Due to the different handling, use and provision depending on the service and provider, a transparent overview in a dashboard is not simply given per se. Instead, the specific requirements of the cloud provider must be taken into account, such as AWS multi-accounts, Azure subscriptions or Google projects.

Cloud operation

The same applies in the cloud: a well-positioned operating team with a lot of practical experience and technological know-how forms the indispensable foundation for success.

However, classic IT often has little or no points of contact in areas such as cloud-native monitoring, infrastructure-as-a-code, deployment pipelines or well-founded open source expertise in cloud-based applications. Of course, DevOps or Site Reliability Engineering (SRE) are ubiquitous, but a few training courses do not automatically turn a database administrator into a cloud architect and new approaches to infrastructure provision and monitoring also require rethinking and investment

The graphic shows the extensive change and shows that successful IT can no longer work in a silo.

Pure cloud expertise will not turn IT departments into cloud competence centers in the short or medium term. This path is more extensive and requires a variety of adjustments. New learning paths and new roles are required and IT roles and responsibilities will change.

recommendations

  • Cost transparency must be anchored in development, provision and cloud operation right from the start. Not everything is economically cloudifiable.
  • Get your IT department involved as early as possible. Too often these are overrun by agile experts who know nothing but cloud technologies. At the end of the day, however, the decades of experience of your IT and application developers still count.
  • Start with the options that cloud providers bring before you rely on additional tools.
  • Consider cloud migrations carefully and don't look at price and technology alone. Lift & Shift migrations from virtual machines are a start to familiarize yourself with cloud technologies, but are rarely economical. The motto should always be: SaaS before FaaS before PaaS before IaaS.
  • Use the possibilities of the providers such as Azure Budgets or AWS Budgets to avoid cost explosions.
  • A historical course of the use of costs and resources is only possible to a limited extent. Use the native interfaces to store data on a long-term basis. Here, too, there are a few options such as a simple export to a PaaS database, or Elasticsearch, in order to use simple visualization tools based on this (e.g. PowerBi or Kibana).
  • Take advantage of the experience from the Cloud Native Foundation Computing Community and fall back on renowned open source components

Conclusion

In order to fully exploit the potential of the public cloud, resource and cost transparency is more important than ever. Do not rely on one tool alone, but establish a corresponding process in the departments involved at an early stage. It is important to address this topic at the beginning of the cloud adoption journey.

DATAGROUP and Cloudeteer have extensive experience from a large number of successful cloud projects. Benefit from our experience, "lessons learned" and our services such as extensive cloud consulting, our own multi-cloud tools such as our multi-cloud dashboards and automated cloud reports.