FinOps is the practice of bringing financial accountability to the variable spend model of cloud, enabling distributed teams to make business trade-offs between speed, cost, and quality.

Cert content#

  • Cloud changes IT
  • What is FinOps?
  • FinOps framework and principles
  • FinOps principles and teams
  • FinOps domains and capabilities
  • Motivations and common language
  • Anatomy of a cloud bill
  • Running a FinOps team
  • FinOps inform, optimize, and operate phases
  • Maturing FinOps

Benifits of FinOps#

  • Cost Control: You get to keep a close eye on cloud spending and make sure you’re only paying for what you really need.
  • Better Budgeting: FinOps helps you plan your spending better, so there are fewer surprises in your cloud bills.
  • Efficiency: You use cloud resources more efficiently, which means you get more bang for your buck.
  • Collaboration: Different departments work together, leading to smarter decisions about the cloud.
  • Flexibility: You can adjust your cloud usage quickly as your business needs change, without wasting money.
  • Visibility: Everyone can see what’s being spent and why, which helps in making informed choices.
  • Optimization: Constantly fine-tuning your cloud services and usage means your operations run smoother and cost less.

Principles of FinOps#

FinOps is built on principles designed to optimize cloud spending and ensure it delivers business value. These principles guide organizations in managing their cloud investments effectively:

Collaboration is Key to FinOps Success#

  • Explanation: Different teams within the organization, such as IT, finance, and various business units, must work together to manage and optimize cloud costs.
  • Example: A monthly cross-department meeting to review cloud spending and usage, where IT can explain technical needs, finance can address budget concerns, and business units can discuss their operational requirements.

Decisions on the Cloud are Derived from Business Value#

  • Explanation: Decisions about cloud usage should be based on the business value they provide, not solely on their costs.
  • Example: Investing in higher-cost cloud services that enable faster data processing and improve customer satisfaction, leading to increased sales.

Everyone Takes Ownership#

  • Explanation: Every team member is accountable for the cloud resources they use and should understand the impact of their usage on costs.
  • Example: Developers tagging their cloud resources correctly to ensure accurate tracking and accountability for usage and costs.

Timely Reports, Relevant Teams, and Accountability#

  • Explanation: Regular and prompt reporting on cloud usage and costs is essential, and the information should be shared with the appropriate teams who are accountable for their decisions.
  • Example: Generating weekly cloud cost reports and sharing them with project leaders to quickly adjust any overspending or inefficient resource allocation.

Centralized Team#

  • Explanation: A dedicated FinOps team leads the organization’s FinOps activities, establishing policies, and sharing best practices across teams.
  • Example: A FinOps center of excellence within the company that provides guidelines, tools, and training for cloud cost management.

Take Advantage of the Cloud’s Variable Cost Model#

  • Explanation: Leverage the cloud’s pay-as-you-go pricing model to save costs by scaling services to match business needs dynamically.
  • Example: Automatically scaling down cloud resources during off-peak hours to reduce costs, then scaling up when demand increases.

By adhering to these principles, organizations can ensure that their cloud spending is efficient, transparent, and aligned with business objectives, enabling them to harness the full potential of the cloud.

Basic Terms in FinOps#

  • Cost Allocation: Allocating costs to understand where money is being spent in cloud services. Imagine dividing your monthly internet bill based on each family member’s usage. Example: Your cloud bill is $200, with $150 for storage and $50 for computing. Cost Allocation helps you see exactly where your money goes.

  • Wasted Usage: Paying for cloud services that you don’t actually use. Like a gym membership that goes unused but still costs you money every month. Example: You have a cloud database running 24/7, but it’s only used during office hours, leading to wasted spending.

  • Oversized/Undersized: Not matching the cloud service size to your actual needs. Oversized is like renting a bus for a solo trip; undersized is like using a compact car for a group of 10. Example: Your app is running on a server fit for 1000 users but only has 100 users, indicating it’s oversized and needs rightsizing.

  • Workload Management: Ensuring cloud resources are used only when necessary. Like turning off lights in rooms that are not in use to save electricity. Example: Automatically shutting down development servers on weekends, reducing unnecessary cloud costs.

  • On-demand Rate: The full price for cloud resources without any discounts. Paying the walk-up price for a movie ticket without any discounts or membership benefits. Example: Using cloud compute services without commitments and paying the standard rate per hour.

  • Rate Reduction: Lowering the cost of cloud services through discounts. Getting a discount on your coffee by using a loyalty card. Example: Signing a 1-year deal with a cloud provider to use certain services and getting a 20% discount.

  • Cost Avoidance: Taking actions to prevent unnecessary cloud spending. Deciding not to upgrade to a premium cable package because the basic one meets your needs. Example: Turning off auto-scaling for an application during its low-traffic season to avoid paying for unneeded resources.

  • Cost Savings: Reducing expenses on cloud services. Finding a sale on groceries that saves you money on your weekly shop. Example: Moving non-critical data to cheaper cloud storage options, cutting storage costs by 50%.

  • Savings Realized: The actual amount saved on cloud services after taking specific actions. Checking your receipt and seeing you saved $30 by buying items on sale. Example: After optimizing your cloud usage, your monthly bill decreases from $1000 to $800, realizing $200 in savings.

  • Savings Potential: The expected savings from planned actions on cloud services. Estimating you’ll save $100 a month by cutting out takeout coffee. Example: Predicting a $500 monthly saving by planning to migrate several workloads to a more cost-effective cloud service.

  • Commitment-based Discounts: Discounts received for committing to use a certain amount of cloud services. Like getting a lower rate on a hotel room by booking it for a whole week instead of just one night. Example: Committing to a certain level of cloud compute usage for a year and receiving a 25% discount on the rate.

  • Commitments Unused/Unutilized: Paying for more cloud services than you actually use. Like buying a 10-visit pass to a play center but only going twice. Example: Pre-purchasing 1000 hours of compute time but only using 800, wasting 200 hours.

  • Covered Usage: Cloud usage that benefits from pre-purchased commitments or reservations. Using all the visits on your play center pass, getting full value from the purchase. Example: All compute hours you actually used that were discounted thanks to your upfront commitment.

  • Coverable Usage: Cloud services that could be discounted if they match your usage commitments. Potential visits to the play center that could be cheaper if you had a pass. Example: Identifying additional cloud services that could be covered under a new or existing commitment for additional savings.

  • Unblended Rates (AWS specific): Different prices for cloud services based on usage amounts. Paying for each ride at an amusement park separately, with prices varying by ride. Example: Paying varying prices for AWS services based on the individual usage and service type.

  • Blended Rates (AWS specific): An averaged cost for cloud services across all accounts or services. Buying an amusement park day pass that lets you ride everything at an average price. Example: AWS averaging out costs across multiple services and accounts to give you a consistent rate.

  • Amortized Costs: Spreading the upfront costs of cloud services over their useful life. Like paying off a car in monthly installments rather than all upfront. Example: Spreading the cost of a 3-year reserved cloud instance across each month of its term to reflect its monthly cost impact.

  • Fully Loaded Costs: The total cost of cloud services, including all associated fees and discounts. The total cost of owning a car, including insurance, gas, and maintenance, not just the purchase price. Example: The comprehensive cost of using cloud services, factoring in rates, discounts, and additional charges like network fees.

Finance Terms for Cloud Users#

  • Matching Principle: When you match costs with the revenue they help generate, not when you actually pay for them. For example, if you pay for a year’s worth of cloud services upfront, you recognize the expense each month as you use the service, not all at once.

  • Cost of Capital: The cost of borrowing money or the return expected by investors. It’s like the interest or profit you need to pay for using someone else’s money. Imagine you take a loan to start a cloud project; the interest you pay on this loan is your cost of capital.

  • Cost of Goods Sold (COGS): The direct costs tied to producing what you sell. This includes things like raw materials and labor. If you sell an online service, COGS could include the cost of the server time and software licenses needed to deliver that service.

  • EBITDA: A way to measure a company’s financial health without getting into taxes or interest. It stands for Earnings Before Interest, Taxes, Depreciation, and Amortization. Consider a cloud company that makes $1 million before paying any interest, taxes, or depreciation; this is its EBITDA.

  • Abstraction: Simplifying complex numbers into more understandable terms. It’s like explaining cost savings in terms of everyday items. For instance, saying “saving $1000 on cloud services is like getting 250 free coffees” makes the savings more relatable.

  • Cloud Language vs. Business Language: Using simpler, more general business terms instead of cloud-specific tech jargon. This helps everyone understand financial reports and decisions. Instead of saying “We reduced our API Gateway requests to lower costs,” you might say, “We optimized our customer interaction points to save money.”

  • Universal Translator: A method that helps finance and tech teams understand each other by translating technical terms into business-friendly language. Imagine explaining the need for increased cloud storage by comparing it to needing more shelves in a warehouse for a growing inventory.

  • Benchmarking and Gamification: Comparing your cloud spending and saving efforts to others and making a game out of cost-saving measures. For example, departments compete to see who can lower their cloud costs the most, with rewards for the winners, like a special mention in the company newsletter.

Types of Cloud Bills#

  • Invoices:

    • Summarized and lack granularity.
    • Serve as accounts payable tools, not suitable for detailed FinOps analysis.
    • Example: A monthly cloud service bill without detailed usage statistics.

  • Cloud-native Cost Tools:

    • Provided by cloud service providers (e.g., AWS Cost Explorer).
    • Useful for basic reporting in single cloud environments.
    • Have limitations in multicloud setups and for complex reporting needs.
    • Example: AWS Cost Explorer offering insights into AWS-specific spending.

  • Detailed Cost and Usage Billing Data:

    • Offers a wealth of information but is complex.
    • Requires specialized knowledge and tools for full comprehension.
    • Contains granular charges updated multiple times a day.
    • Example: A report detailing every instance hour, storage usage, and data transfer costs.

Key Personas in FinOps#

Understanding the different roles and their perspectives on cloud spending and management is crucial for effective FinOps practices. Here are the key personas:

The CEO Persona#

  • Objectives: Accelerate company growth, strategic advantage, faster market time, market-leading solutions.
  • Frustrations: Unpredictable cloud spend, uncertain ROI on cloud investment.
  • Key Metrics: Revenue growth, cost of goods sold (COGS), unit cost economics.
  • FinOps Benefits: Risk management, connecting engineering decisions to business outcomes, cloud spend prediction, guiding cloud investments.

The CTO and CIO Persona#

  • Objectives: Similar to the CEO, with an emphasis on technological leadership and innovation.
  • Frustrations: Pressure to justify cloud expenses, lack of spend guardrails and reliability.
  • Key Metrics: Revenue growth, COGS, unit cost economics, R&D tracking.
  • FinOps Benefits: Strategic cloud investments, spend prediction, and cloud usage freedom.

The CFO Persona#

  • Objectives: Engineering accountability, cost-effective applications, collaborative cost reduction and allocation.
  • Frustrations: Unsatisfied engineering teams, unpredictable budget impacts, difficult app ownership identification.
  • Key Metrics: Revenue vs. infrastructure cost, cost per service, showback/chargeback.
  • FinOps Benefits: Increased visibility into cloud costs, enhanced accountability, incentives for robust architecture.

The Engineering Persona#

  • Objectives: Company growth, competitive edge, quick market entry, innovation delivery.
  • Frustrations: High cloud costs, team dissatisfaction, ensuring business continuity and reliability.
  • Key Metrics: Revenue growth, COGS, time to market, engineering productivity, R&D spend.
  • FinOps Benefits: Encourages wise cloud investments, spend predictability, freedom in cloud technology use.

The FinOps Lifecycle#

The FinOps lifecycle is designed to help organizations better manage their cloud spending through a series of structured phases. Each phase focuses on specific activities and goals, promoting continuous improvement and collaboration across teams. The lifecycle consists of the following phases:

image

1. Inform#

  • Objective: Provide visibility into cloud spending and usage.
  • Activities: Collecting and analyzing data on cloud usage and costs, establishing a baseline for spending.
  • Outcome: Stakeholders have a clear understanding of current cloud spending patterns, enabling informed decision-making.

2. Optimize#

  • Objective: Ensure cloud resources are used efficiently to balance cost, performance, and capacity.
  • Activities: Identifying waste, rightsizing resources, choosing the right pricing models (e.g., reserved instances, savings plans), and optimizing across different cloud services.
  • Outcome: Reduced waste and improved efficiency in cloud spending without compromising on performance or capacity.

3. Operate#

  • Objective: Implement governance and operational practices to manage cloud spending proactively.
  • Activities: Establishing policies for cloud usage, setting up budgeting and forecasting, implementing cost allocation tags, and conducting regular reviews.
  • Outcome: Effective governance over cloud resources, ensuring alignment with budget and operational goals.

4. Report#

  • Objective: Communicate financial management and performance metrics to stakeholders.
  • Activities: Creating detailed reports and dashboards that provide insights into cloud spending, trends, and optimization opportunities.
  • Outcome: Stakeholders are well-informed about the financial performance of cloud investments, supporting strategic decision-making.

5. Iterate#

  • Objective: Continuously improve financial management practices of cloud resources.
  • Activities: Revisiting the inform, optimize, and operate phases with new data, insights, and operational feedback to refine strategies.
  • Outcome: A cycle of continuous improvement that adapts to changing needs, market conditions, and cloud pricing models.

FinOps for Containers#

Understanding Container Impact on FinOps#

  • Containers and Orchestration: They change how we manage costs because they share resources and move around a lot.
  • Cost Visibility and Allocation: It’s harder to see and manage costs because everything is more dynamic.

Transition to Container Orchestration#

  • Why Use Containers: They make apps scalable and reliable.
  • Kubernetes and Resource Utilization: Like playing Tetris, Kubernetes helps fit containers efficiently on servers, but it can hide the cost of each container.

The Container FinOps Lifecycle#

  • Inform Phase:
    • Reports: Make reports to show what each container costs.
    • Cost Allocation: Figure out how much each container uses and costs, including extra expenses like licenses.
    • Container Proportions: Understand how containers use server resources and use tags and labels for cost allocation.
  • Optimize Phase:
    • Rightsizing: Adjust the size and number of containers based on need.
    • Workload Matching: Make sure the resources match what the container needs.
    • Server Instance Rate Optimization: Use reserved and spot instances to save money.
  • Operate Phase:
    • Scheduling: Plan when containers should run to save resources.
    • Removing Idle Containers: Get rid of containers that aren’t being used.
    • Maintaining Tags and Labels: Keep container information up to date for better cost tracking.

By understanding these steps and applying FinOps practices, businesses can handle the financial side of using containers effectively, making sure they get the technical benefits without overspending.

Finops tools#

assessmen tool

Backward Operation and Support Aws Forward