Technical Blog

Technical and Product News and Insights from Rackspace

The art of decision management for multicloud strategies—Part Three

This three-part series introduces you to the art of managing the many necessary decisions for managing your multicloud environment.

If you missed them, check out Part One and Part Two of the series.

Other key dependencies and best practices

Cloud Decision Management Process (CDMP) helps reduce the risks of making inappropriate decisions. A customized decision tree and a platform evaluation framework help match each application with an appropriate cloud platform and filter those workloads that you should not transfer to Cloud.

It should be clear that the effort to construct an effective CDMP is not trivial. The process could be complicated by rounds of facilitated discussion among major organization stakeholders. As you can determine target cloud platforms for the majority of applications either quite intuitively or within Stage 1, only large organizations implementing a multicloud strategy for hosting several complicated application stacks should require the due diligence process at Stage 2. Failure to make the right decisions could entail costly consequences. See Part Two for a refresher on Stages 1 and 2.

Figure 1: CDMP Framework Overall Process Flow

CDMP depends on a few fundamental information sources to function well. Two of the most outstanding ones are the company’s corporate guardrails for multicloud strategy and its application portfolio discovery.

Corporate Guardrails

These guardrails are policies or a wish-list for the company to realize its multicloud strategy. These typically govern the future direction and approach to different application stacks for their cloud journey as the company tries to develop and execute a migration plan over a multi-month timeline. For example, the company might have the following guardrails defined for their cloud journey:

  • We should retire all back-office systems running on legacy or proprietary OS and technologies in three years. We should seek a replacement with SaaS products in the marketplace before considering a new build or a Cloud purchase.

  • We should containerize all digital transformation workloads to accelerate the adoption of DevOps through Agile in a consistent manner.

The preceding guardrails affect how you define questions on the decision tree at Stage 1 and the order in which you execute them. The decision tree should route those retiring back-office systems to SaaS platform options before everything else. Meanwhile, the evaluation epics at Stage 2 could compare functions or ease of use across different SaaS products. By the same token, the second guardrail could make OpenShift a more favorable choice over other platforms for digital stacks. Any evaluation epics concerning digital systems can incorporate credit tiering reflecting the preference for OpenShift.

Application Portfolio Discovery

While corporate guardrails provide the company with the direction its multicloud efforts should spearhead, the application portfolio analysis reveals the constraints and dependencies that the company must consider in formulating their strategy. Before developing a CDMP framework, an assessment of the current application portfolio helps capture important information concerning your target cloud platform selection. These might include the following considerations for each in-scope application:

  • Most notably, inter-dependency with other applications
  • Current and projected TCO
  • Opportunity for system modernization
  • Room for architecture improvement
  • Reasonable migration patterns
  • Systems integration requirements to achieve pre-existing and new functional objectives

Compiling questions for the decision tree at Stage 1 need to incorporate some or all of the preceding considerations to help direct different applications to the appropriate cloud platform choices. Failing to do so might result in unrealistic or undesirable outcomes at the exit of Stage 1. Likewise, at Stage 2, the evaluation epics and credit tiering should also cover these constraints and dependencies to ensure their importance is reflected in the final recommendation.

It is important to note that the corporate guardrails and the application portfolio assessment outcomes might direct the company to move towards different directions. For example, a retiring CRM system should adopt a target SaaS product within 24 months based on the guardrails, while the TCO analysis for its application stack reveals a three-fold expenditure increase over the next few years. Another example is that a mobile application running on SQL databases is subject to be containerized in Azure. In contrast, the portfolio discovery indicates it should adopt open-source databases to integrate with modern analytics technologies and save costs in the long run. These are situations and decision points where a CDMP would be helpful for a reference resolution.


Decision management should keep abreast of the change in business strategy, business and IT requirements, and stakeholders’ priorities over time. Therefore, the lifespan of a static CDMP framework is quite short, and it must be kept up to date to stay relevant. If you break the cloud journey into key phases with multiple weeks or months between successive phases, then you should review and refresh the CDMP process to serve each phase’s specific purpose. With the continuous process improvement mindset, the CDMP is a valuable asset to companies with complicated application stacks and sizeable infrastructure to reckon with for a strategic multicloud journey.

Use the Feedback tab to make any comments or ask questions. You can also click Sales Chat to chat now and start the conversation.

post avatar
Wilfred Wah

Wilfred is a Principal Consultant of Advisory Services, Rackspace Technology North Asia. His area of expertise includes cloud migration assessment, cloud economics, people transformation, cloud adoption strategy, and the transformation roadmap. He has more than 15 years of consultancy experience with major vendors such as Dell/EMC, SAS, and AWS before joining Rackspace.

Share this information: