Azure Migrate, Best practices, Hybrid + Multicloud, Migration
Demystifying cloud economics
By Jeremy Winter Partner Director, Azure Management
5 min read
Migrating to the cloud is an evolution, and it’s important to think differently about how you consume resources. As you’re building a business case in your organization, it’s critical to step back and understand the cloud’s key constructs and transform your mindset. It starts by having a conversation about today versus tomorrow and what is possible in the cloud, as with this migration, you will get instant access to innovative technologies and several new options that do not exist on-premises.
Azure is here to help you start your cloud journey strong with key financial and technical guidance as well as best practices from customers who charted a successful cloud journey. With this intention, we recently launched a new initiative to help our customers understand and demystify cloud economics. We will provide a rich set of digital content highlighting key technical and financial tips from Azure experts and share lesser-known tips through this initiative. In this blog, we’ll discuss a set of key considerations that will save you time, budget, and resources as you chart your cloud journey.
How does cloud pricing work?
Cloud billing is tied to compute and storage that includes the underlying software licensing fees. Costs accrue via a pay-for-what-you-consume model versus the up-front server infrastructure and software licensing costs that you would typically pay on-premises in your data center. If you run your workloads on-premises, you have a combination of upfront costs and operating expenditures. When shifting to the cloud, you largely shift to pay-as-you-consume-based models, which results in a largely operating expenditure-based model.
The first thing to note is that the cloud is infinitely flexible and is not “one size fits all.” As you literally pay for what you consume, for the best pricing, you need to consider how you will consume resources for your specific workloads. Then you can establish your fixed and variable cost models to maximize your investment. And these models complement each other. You can layer on top of your variable resources for your seasonal or demand-based activities where elastic computing makes sense and where you can consume on-demand or automate against specific capacity thresholds. That said, we’d like to provide some guidelines to align your cloud spend with underlying workloads.
Align your cloud spend with underlying workloads
Know your workloads
When on-premises, your architecture is likely provisioned for peak capacity. Shifting from on-premises to the cloud with the notion that you can scale up and down and take full advantage of the cloud benefits. Therefore, it is important to know your workloads and understand our key constructs for maximum efficiency.
- Idle capacity: Azure allows you to eliminate idle capacity intended to cover future growth across workloads. Actions like rightsizing or eliminating unnecessary workloads can help you reduce your idle capacity when moving to the cloud.
- Unpredictable workloads: The overall premise and major advantage of the cloud that you’re probably most familiar with is the power it gives you to elastically scale compute resources in response to different peaks in your business. This is great for unpredictable workloads, wherein the Azure service you can add and subtract resources as you need them, resulting in variable costs. Taking advantage of tools and actions like virtual machine scale sets and “snoozing” can help you only pay for the resources needed.
- Predictable workloads: If a portion or all of what you are consuming is more consistent, for example, a batch process that runs every day using the same resources as clockwork on a schedule or what we call a predictable workload, we have options for that, too. You can benefit from fixed costs at reduced pricing by taking advantage of the cost-savings offers such as Azure Reservations.
- Do a clean-up and right-size from the get-go: Just like when you move from one house to the other, you sort the items that you don’t use and decide what to do with them.When planning to move your workloads to Azure, consider which workloads are no longer needed and can be turned off. This can help you build stronger business models and show an immediate impact on your budgets. For workloads still needed, consider what can be done to optimize those resources and operational hours, leveraging tools such as Azure Migrate.
Take advantage of the cost-savings options
Here are the key cost-savings offers for you to consider for keeping your costs in check:
- Azure Hybrid Benefit: A licensing benefit that helps you significantly reduce the costs of running your workloads in the cloud. It works by letting you use your on-premises Software Assurance-enabled Windows Server and SQL Server licenses on Azure. And now, this benefit applies to RedHat and SUSE Linux subscriptions, too.
- Spot virtual machines: Get deep discounts for interruptible workloads that do not need to be completed within a specific timeframe, such as high-performance computing scenarios, batch processing jobs, or visual rendering applications, dev and test environments, including continuous integration and continuous delivery workloads or large-scale stateless applications.
- Reservations: Receive a discount by reserving your resources in advance, which allows you to be more efficient. In return, we pass these savings onto you as discounts of up to 72 percent.1
- Azure dev and test pricing: Get discounted rates for your ongoing development and testing, including no Microsoft software charges on Azure Virtual Machines and special dev and test pricing on other services.
- Extended security updates: We’re providing several options to continue the support for SQL Server 2008 and SQL Server 2008 R2, which have reached the end of their support (EOS) life cycle. You can migrate your on-premises SQL Server instances to Azure Virtual Machines, Azure SQL Database, or stay on-premises and purchase extended security updates. Unlike with staying on-premises, you’ll receive free extended security patches by migrating to an Azure Virtual Machine.
Understand what process and financial stories typically change with cloud migration
In the financial considerations for cloud migration blog, we shared how cloud migration can affect CFO priorities and how the organization’s financial posture and financial KPIs and processes change. Key financial benefits of Azure are driven by a fundamental shift in the IT operating model, which benefits your organization’s core financial statements in the following ways:
- Balance sheet: When you operate your datacenters, you have expensive long-term assets that limit the cash and capital required to grow your business. While in the cloud, you can shift datacenter operations costs into developing cloud applications and other projects that drive business growth, which make your balance sheet more agile.
- Cash flow statement: With the “pay per use” model along with platform capabilities like policy and tagging that Azure enables, you can increase the visibility and predictability of your cash flow statement and delay cash spend.
- Income statement (profit and loss): Over time, you can improve profitability by reducing the cost to deliver equal or larger IT value by taking advantage of Azure’s flexibility, low management costs, and its broad portfolio of services and pricing models.
If you would like to learn more about cloud economics and lesser-known tips, please watch this Mechanics video.
- Financial considerations for cloud migration video
- Technical considerations for cloud migration video
Looking for more technical and financial considerations by Azure experts? Visit the cloud economics webpage today.
1The 72 percent saving is based on one M32ts Azure VM for Windows OS in US Gov Virginia region running for 36 months at a Pay as You Go rate of ~$3,660.81 per month; reduced rate for a three year Reserved Instance of ~$663.45 per month. Azure pricing as of 30 October 2018 (prices subject to change). Actual savings may vary based on location, instance type, or usage.