Virtual Machine Comparison: AWS, Azure, Google Cloud and OVH

Comparison of Virtual Machines in the Cloud

Public Cloud providers today offer virtual machines that are more or less similar in terms of configuration. Each of these instances has its own specificities: RAM power, CPU power, graphics cards specially designed to drive artificial intelligence models, etc. Whether you are in a multi-cloud development context, or more upstream, in the search for the solution best suited to your needs, it is sometimes difficult to find the right information and the right virtual machine comparison.

For an instance of equivalent level, providers do not offer exactly the same configuration and the same price. We have decided to make a simple and accessible comparison to better understand what each provider offers and thus help you in your decision making.

Sign up for the FinOps Lota.cloud Training

Comparison methodology

To make this comparison, we decided to focus on the three major American suppliers and the European leader: AWS, Microsoft Azure, Google Cloud and OVH Cloud.

Our comparison is based on entry-level Linux virtual machines from these four major vendors. To classify virtual machines, it is usual to rely on 3 characteristics:

  • the number of vCPUs
  • the default number of cores
  • RAM

First, we will make a purely comparative comparison. factual information from these different bodies, with a purely technical approach, based on the configuration.

We will then go further in the analysis by looking at the impact of localization and the ancillary costs that may be added to the prices announced by suppliers.

Technical comparison of virtual machines

Our first analysis is therefore based on the purely technical aspect and on the configurations associated with the different instances studied. We based our analysis on one of the most frequently used instances at AWS: t2.small. We are therefore talking here about an entry-level reference, adapted to test or development environments with limited computing power requirements.

We then searched for reference instances at other suppliers. At Google Cloud, the instance that seems to be the most close to this type of product is g1-small. At Amazon Azure, our choice went to B1s. At OVH Cloud, we've been looking into on s1-2.

Here's our comparison:

Comparison table of basic virtual machines at AWS, Google Cloud, Azure and OVH Cloud. Date: 03/01/2020.

You will have a little more GB of RAM depending on the provider (AWS and OVH Cloud offer the most powerful RAM in the world. configuration) or may eventually depend on a shared core, such as it's the case with Google Cloud.

For equivalent configurations and powers, and as you know we like to talk about the costs of the Cloud, it's essentially the price that's going to attract our attention here. Please note: prices vary by region. Here, we have taken Europe as a reference (we go into more detail in the next paragraph, patience).

OVH Cloud provides a much more affordable offer than the American giants. This significant difference can partly be explained by the fact that OVH's offering is less managed and automated than its competitors.

European Pride

Also note that Google includes a discount directly in its monthly prices, without having to resort to instance booking as with AWS or Microsoft Azure. You will automatically receive a -30% discount on your monthly invoice, which means a more attractive price over a longer period of time. You'll get a little blurry again, won't you?

Comparison by location

To have a more global vision of the different trends, a more detailed analysis is needed, taking into account 3 main areas of concern regions: Europe, Asia and North America. Indeed, we can see sometimes large variations depending on the geographical area in which is your instance.

Location-based comparison of virtual machines in the Cloud
Comparative table of virtual machine instances according to regions. Date: 03/01/2020.

This disparity can be explained by several factors and the choice of region can obviously affect the quality of your service, particularly in terms of latency. The choice of a specific region is only not only to the fare, but also to the direct proximity to your users for example.

This table therefore shows us the variations between the different geographical areas. Confirmation: OVH Cloud offers much lower prices than its GAFAM competitors. In this price opposition, Microsoft Azure nevertheless wins over Amazon and Google, with prices below $10 per month.

Bill Gates facing competition from the GAFA

Please note: we do not intend to highlight one vendor over another, as we know that this comparison only applies to a very specific instance, ideal for test or development environments. This article only aims to establish trends on reference instances for each supplier and to show the influence of regions on invoicing.

This strong difference in price positioning can be explained in particular by the power of the support offered by the American giants compared to the less managed products offered by OVH Cloud. The difference between the supports is also to be taken into account, for example. Not to mention the trade margins, which must certainly have an influence here.

Free FinOps Excel dashboard and 100% manual

Beware of hidden costs

You hear it a lot, but a reminder won't hurt to person: some costs are not directly advertised and could be surprise you when you receive your invoice.

Here, for example, we do not take into account the costs of related to the storage or transfer of data, which are often synonymous with unpleasant surprises when it comes time to settle the bill.

On the other hand, we do not mention the discount solutions proposed. by certain suppliers, such as the Reservation of Instances or Saving Plans at AWS. All these variants are to be taken into account in the development of your strategy and in the implementation of your action plan.

Finally, this short analysis allows us to better understand the importance of making the right choices from the start of your project strategy. We're well aware of how difficult it can be for you sometimes to be met in decision-making, both the data on the costs of the Cloud are blurred and difficult to analyze. This is where a platform of Cloud Cost Management such as ours can help you be more comfortable with your bill. Better anticipation of your expenses, 360° visibility on your billing data and leads optimizations will allow you to make decisions in a more efficient way. serene.

If you are interested in our Cloud Expense Management platform, do not hesitate to test our solution free of charge for 14 days by clicking on the button just below.

Leave a comment

Test Lota.cloud for free for 30 days