Last month, Azure announced that it “helps customers achieve more with the best prices”, taking down the cost of the latest version of Azure D-series VMs by 10%-17%. That happened about a week after AWS announced a 5% reduction in prices for its C4, M4 and R3 instance families (for Linux only).
These price cuts are in-line with currently believed notions in the cloud industry that cloud prices continuously decline because of vendors’ inherent economies of scale, competition (especially with AWS), and the fact that innovation lowers infrastructure costs (i.e. Moore’s law). However, this past July, Barb Darrow, a well-known voice on technology and cloud computing, put out an article about Microsoft Azure raising its prices in August 2015, specifically for UK customers (30%), and Australian customers (26%).
In addition, it has become a common assumption that operating in the cloud translates into automatic savings. However, if a box of cereal increases in size along with its price, then the cereal isn’t necessarily more expensive. You then have to ask yourself, is price the only factor for successful cloud operations? Obviously not.
In a previous post that analyzes AWS price reductions over the past five years, we revealed that Amazon hasn’t changed its compute prices for the last two years, but has presented additional types of instances with increased performance at different costs. It’s not enough to say that prices are going up or down without measuring performance levels. For example, even with price increases, overall performance may still be a better value than offerings from other vendors. The cloud is similar to other IT infrastructures in that it needs to serve a business by supporting an increase in performance. Ultimately, you wouldn’t mind paying more if your return increases. Nevertheless, there are still challenges in making the cloud more effective and efficient. Below, we offer guidelines to help address these challenges and cope with the complexities of the cloud.
Complexity, Visibility and Balance
There are challenges in ensuring that you maximize the benefits of your cloud costs. It starts with the diversity of your IT environment, and becomes especially complex with environments that are made up of multi or hybrid cloud architectures. For these cases, you need to find the common ground between your platforms in order to understand where to allocate resources and achieve the best performance/cost for running your applications.
When dealing with private clouds, you need to price your own infrastructure. For instance, with the OpenStack cloud, you have to define instance families and instance types as well as price instances in order to run chargebacks, and have a baseline that you can relate to when deciding on where to run the next workload. Learn more.
Another challenge is the lack of transparency with cloud provider pricing. New instance offerings and constantly changing prices make it difficult for users to stay up to date on pricing. As Darrow revealed with the Azure price increase, it seems like vendors can’t keep up with all of the changes, themselves.
Finally, the last challenge to discuss is finding a balance between performance and price. Returning to the cereal box example, when you purchase instances, the price should reflect a reasonable ROI for your business based on basic economic principles of value. In some cases, in order to get better value from your cloud services, you can adjust the compute power that you purchase to find a balance between what you pay and performance levels.
1. Monitor Rates
You should always be aware of what you pay for your cloud services by monitoring your costs. This means not only the total costs of deployment, but also price rates and how much each service costs. It is best to set cloud monitoring alerts and reports as well as perform thorough financial analyses of your cloud costs.
Check out Cloudyn’s “provider rates” for up-to-date price lists of all Cloudyn-supported providers and learn about how to monitor your cloud costs.
2. Make Commitments
When planning for long-term cloud usage, you should consider making commitments, which are agreements to use a certain volume of compute usage at a fixed rate that is unaffected by market volatility. This is seen with Amazon’s reserved instances (RIs) and Microsoft Azure’s Enterprise Agreements. With Azure’s recent price hike mentioned above, customers with Enterprise Agreements were not affected, which shows the true potential of commitments on cost savings.
Are you an AWS User? Check out the cost effectiveness of AWS Reserved Instances
3. Periodic Assessments
It is imperative to have periodic assessments of your deployment. This includes keeping an eye on different offerings from other vendors. Although it can be difficult to switch over to a new cloud provider, you can use knowledge on what other vendors offer as points to leverage for price negotiations. You can also utilize this knowledge for future projects by using another cloud that better suits your budget or required capabilities.
Learn more: 3 Ways to Avoid Cloud Sprawl from the Get-Go
4. Continuous Optimization
To stay on top of your cloud game, you should be looking to continuously improve your system. Resources that are deployed for an application should be allocated according to your current compute power demands and user activity levels. This allocation should be done on a routine basis, which will allow you to find the point of balance between cost and performance.
Learn more: 10 Practical Tips for Cloud Optimization
New Huge Price Cuts on the Horizon
The slight price reductions, and the fact that we haven’t seen significant price changes since AWS’ huge price reductions in 2014, affirm that vendors don’t see pricing as a competitive edge but look to enhance their pool of resources and platform capabilities, instead. In addition, cloud users should be able to find their own “sweet spot” and optimize their pool of resources considering both cost and performance dimensions.
Start Optimizing Your Multi-Cloud Deployment
Start your free trial today to easily identify greater cloud efficiency opportunities and view your entire AWS deployment in one place. Gain full transparency, accountability, and control.