service
- Consulting
TCS Consulting Services to Achieve Success in the Business 4.0 Era
Highlights
- Cloud costs are not the same for all use cases; different target solution architectures will mean different costs.
- It is important to design a reasonably cost-effective target state architecture factoring costs of cloud operations.
- A FinOps target operating model can help address the challenges of cloud cost management.
In this article
A double-edged sword
Easy access to unlimited compute resources is great but beware of unexpected costs and inaccurate forecasting.
Cloud bills can be difficult to comprehend. Billing reports generated by cloud service providers do not list charges against complex engineering and their financial implications individually—they are clubbed under one total charge. While that may be, it’s not so difficult to understand that cloud cost is not the same for all deployments. Various application and infrastructure architectures come with different costs. So, it’s vital to design the most cost-effective target state architecture factoring in cost for all phases of adoption from development to operations. It will be very difficult to achieve the stated benefits in the business case if the target state is not properly defined with cost benefit levers such as cloud-native solutions, modernized architecture patterns, no vendor lock-ins, and automated AI-based reporting, monitoring and recommendations.
Cost management during cloud adoption and that at the cloud steady state can also become disjointed if the key success metrics of cloud adoption are not properly monitored during the transition and post adoption. This is where DevFinOps comes in.
DevFinOps for cloud value
DevFinOps encapsulates cost management in every phase of cloud operation.
The cost management spans three significant phases:
Define the cost benefit
This takes place at the outset, in the cloud assessment phase. Business case plays a critical role in value identification and drives value-based architectures. The focus in this stage must be on carrying out a pre-planned cost analysis, considering key cost aspects such as different application and infrastructure architecture patterns, strategies for reporting, and waste management, mapping to the key operational targets. Scope for alteration to factor in unexpected change requirements from business or IT.
Realize the value
This kicks in during the cloud transition phase. Cloud value realization pervades all implementation phases right from strategic planning to operations. It covers the entire spectrum from identifying, designing, quantifying, prioritizing, communicating, measuring, and delivering value. It is key to establish the buy-in, build momentum, and chart the operational value realization plan.
A value realization office, established as part of an enterprise’s cloud transformation strategy, ensures fruition of potential benefits identified in the business case. Businesses need to ensure that the value realization office has the onus, freedom, and capability to define, design, and manage value. It should be able to foresee the business impact, forecast the ROI, and identify and mitigate risks.
A value realization office must maintain artifacts such as a living business case; a detailed benefit realization plan including the success criteria ownership; stakeholder accountability to measure, monitor, and report value; baseline; scorecard; and target KPIs. Its responsibilities should include:
- Measuring, tracking, and reporting on benefits and ROI realized.
- Conducting variance analysis of benefit realization plan.
- Identifying actions to maximize benefits realized.
- Providing updates to executives on a regular basis.
- Conducting a benefit and ROI validation analysis and reconciliation to plan.
Manage and optimize value
This comes into play during what we call the cloud steady state. Inform, optimize, and operate are key phases in the FinOps life cycle. The strategies and tactics of cost optimization vary widely based on the organization and maturity of cloud adoption. Cloud cost management has a series of iterative steps for various considerations by selecting the right trade-offs between architecture and performance metrics.
Fig: FinOps target operating model in the cloud with key domains and levers of cost control in the steady state
A FinOps target operating model addresses the challenging aspects of cloud cost management in the steady state. Most common concerns of cost management are proactive actions to manage cost, managing cost shared across departments, accurate forecasting, and reducing waste. A well-defined cloud governance framework with clear resource ownership, provisioning of controls, billing and metering, waste management, appropriate tagging and labeling, automated reports, threshold-based alerts, and automation strategies play a critical role in cloud cost management.
Fig: FinOps implementation
Chargeback for accountability
Building cost awareness and accountability with a chargeback model
In FinOps in steady state, charge or showback plays a critical role.
Cloud service providers produce a consolidated account-level consumption and billing reports for all the resources used. Businesses must analyze these reports for consumption and budget variance to “chargeback” or bill to the respective departments or services. Business accounts must put in place a chargeback model to assess the alignment and agreement among key stakeholders such as Line of Business, Finance, other functional units, and cloud service providers (CSPs). This will require looking at the account, tagging strategy, and alignment of application data (for instance, the App ID) to the different budget groups while developing the migration plans. It’s a key step to defining an equitable chargeback model.
Depending on cost optimization objectives, define optimal account structures to establish the right level of hierarchical account order and logical account groupings. Start tagging the resources at the earliest and maintain the consistency of tagging strategy. In specific scenarios such as data transfer costs, CSPs’ support fees, or other costs that are shared by all users, it is very difficult to identify and allocate the costs back to individual teams, projects, or initiatives. In the discovery and assessment phase, analyze the current chargeback model and define the cloud cost and usage reports that can help generate the cost and usage information at the desired level of granularity. This will help in internal alignment on how to allocate the shared cost.
So how can businesses get started on a chargeback model? First, build a dedicated team to develop the chargeback model. Then, to ensure a smooth chargeback process, work closely with the Finance department, get their buy-in to implement chargeback, and gain strong support from the senior management.
On an average, enterprises waste 30% of their cloud spend.
Best practices
From identifying idle resources to putting budget thresholds in place, practical steps to creating value
Cloud cost optimization can be achieved by planning target state architecture, cloud resource utilization, smart selection of cloud features, optimizing cloud performance along with governance processes, appropriate tools, and automation. Here are some best practices you can follow:
- Define identification/classifications and operational/automation tags, adding others according to business needs.
- Reduce cloud waste by analyzing consumption and billing reports to identify orphaned resources, overprovisioned resources, idle resources, suboptimal pricing options, and suboptimal use of reserved capacity.
- Put budget thresholds in place and ensure teams are aware of them and cannot exceed them. Sending out alerts and monitoring consumption are other key levers of cost management.
Visibility
|
Accountability
|
|
|
Conclusion
When it comes to cloud, it’s easy to lose track of the resources deployed—and costs. That’s why visibility, accountability, and control are so important. The cost control framework of DevFinOps in the steady state should consider key factors such as assumptions and affinity of migrating workloads, and work in tandem with datacenter objectives such as eliminating or optimizing costs. Adopting effective chargeback models and making simplify, modernize, and automate the mantra can go a long way in helping achieve savings from cloud transformation and realizing benefits from innovation and productivity to revenue uplift.
Contact
Contact
