The new Terraform Cloud pricing model is so expensive! Or is it?

The new Terraform Cloud pricing model is so expensive! Or is it?

Terraform Cloud recently presented a new pricing model. Previously you paid $20 per user per month, for the?team tier. Now you pay $0.00014 per managed resource per hour. Initially it is difficult to say if this is better or worse for you and your current environment.

Before we dig into the answer, lets first stop and appreciate two important points:

  1. The first 500 resources are free. This means that if you have 500 or less resources, you don’t pay anything. If you go above 500 resources, you pay only for those resources. The first 500 are always free.
  2. With the new pricing model we also get access to features that were previously only available for the priced tiers. This is great news to those who were already using a free tier of Terraform Cloud. They can continue to use a free Terraform Cloud (as long as the number of resources are less than 500), and enjoy premium features such as run tasks and sentinel policy integration.

Let’s move back to the topic of pricing. If the new pricing model is expensive or not depends on how many resources you are currently deploying through Terraform Cloud, and how many users you currently have in your Terraform Cloud environment.

The table below illustrates the situation nicely. On the x-axis (left-right) we have the number of active users, and on the y-axis (up-down) we have the number of resources. The number in the table is the?price per user per month. A green box means we pay less than what we previously paid, a red box means we pay more than we previously paid. A white box simply means we pay an equal amount. Take a look:

No alt text provided for this image

Let us imagine two different possible scenarios:

  1. In scenario one we are two developers managing a large number of resources. We are the two super-heroes of infrastructure-as-code in our company. We have currently deployed 3000 resources throughout all our environments. Unfortunately this new pricing model means we will pay?$151 per user per month. Not good!
  2. In scenario two we are 15 developers split into multiple teams. Each team takes care of their own environment. In total we have 2600 resources. With the new pricing model we pay?$17 per user per month. Nice!

Note again that the first 500 resources are free! So in the first scenario above we are really talking about 3500 resources, and in the second example we are talking about 3100 resources.

Which scenario is most common? In my personal experience scenario two is more common, but more reasonably there would be?more users?and?less resources?than what that scenario has. The first scenario might be common, but if you find yourself in that scenario you should ask yourself why that is. Maybe it is high time to onboard additional users into your Terraform Cloud environment?

Let me know what you think!

#terraform #terraformcloud #hashicorp

Bryan Meier

Director of Data Operations and Warehousing at Mutual of Enumclaw

1 年

Great article Mattias F. your observations are the same as what I have seen. Unfortunately for over half of Hashicorp's customers, this means they are going pay more and in some cases astronomically more. For instance, not everyone is using the predominant providers like AWS, Azure, GCP, etc. In those scenarios it's likely that the quantity of resources is quite manageable and is still cost prohibitive even though it's a little higher. In our case we are using another provider (Snowflake-Labs/Snowflake) that requires a high quantity of resources. We are on the initial Teams plan that was available a couple years ago and paying $260 per month for 8 users. We are running about 20k managed resources. Your chart doesn't go that high but I can tell you that the cost went from $260 per month to about $1900 per month. We are looking to move to another cloud platform for our Terraform needs because of this. What seems to be clear to me is that Hashicorp let someone (maybe marketing) that doesn't have a realistic grasp on how Terraform is being used industry wide come up with a pricing model that has nothing to do with the actual run rate and cost incurred by Hashicorp per run.

回复
Saurabh Negi

Devops Consultant at Coforge | Ex-Amdocs | Ex-Ericsson | Tech Enthusiast | Certified Kubernetes Administrator & AWS Associate Solution Architect, AWS, Kubernetes, Docker Container, IaaC, CI-CD, Shell/Python Scripting

1 年

Thanks for this lovely explaining chart. what i can see from this chart is as long your avg resources / user are 200 you are paying the same ( infact lower since you have 500 free resources additionally). Anything below 200 resources is a WIN and more than 200 resources /User is LOSE. Also i believe in startup companies Scenario-1 is more likely to be prominent while in bigger organizations Scenario-2 is more likely. Appears to be a good pay-per-use model to me as a user.

回复
Christopher (Chris) Johnson

Software architect and entrepreneur

1 年

Mattias Fjellstr?m you haven't accounted for the total cost. At 3000 resources, $151/mo/user x 2 users = $302/mo. At 2600 resources, $17/mo/user x 15 users = $255/mo, not a big savings. And if you use 3000 resources in both cases (not sure why you'd show fewer resources in a bigger team -- I'd think more often the opposite would be true), the second scenario is $20/mo/user x 15 users = $300/mo. That's only $2 less than scenario 1. I don't know what you set out to show here, but the two scenarios are a wash when it comes to total dollars.

回复
Sako M

AI | DevOps | Security | Open Source | #hackpro_sako #everyballwins

1 年

??

回复
Mattias Fjellstr?m

??♂? @ Accelerate at Iver | Author | HashiCorp Ambassador | HashiCorp User Group Leader

1 年

... and as always you can enjoy this article in a nice blue rendition on my blog: https://mattias.engineer/posts/terraform-cloud-pricing/

回复

要查看或添加评论,请登录

Mattias Fjellstr?m的更多文章

  • Azure Flexible Federated Identity Credentials for HCP Terraform

    Azure Flexible Federated Identity Credentials for HCP Terraform

    If you set up workload identity integration between HCP Terraform and Azure today, you need to configure the subject to…

  • Kubernetes provider for Bicep together with user-defined types

    Kubernetes provider for Bicep together with user-defined types

    This article was originally posted on my blog mattias.engineer In the latest release of Azure Bicep user-defined…

  • Azure Kubernetes Service in production - Part 1 - Introduction

    Azure Kubernetes Service in production - Part 1 - Introduction

    Kubernetes can seem notoriously difficult to get right. It is! This is true even if you use a managed service from one…

    1 条评论
  • How to promote releases between GitOps environments

    How to promote releases between GitOps environments

    In this post I will go through an example of how you can promote releases between different GitOps environments. The…

    2 条评论
  • Administering Terraform Cloud using GitHub Actions

    Administering Terraform Cloud using GitHub Actions

    Note that this post is cross-posted here, at my blog mattias.engineer, and at dev.

    1 条评论
  • Kubernetes-101: 1000-Foot Overview

    Kubernetes-101: 1000-Foot Overview

    In this series of articles we have looked at many of the Kubernetes primitives that are needed to run a working…

  • Kubernetes-101: Security concepts

    Kubernetes-101: Security concepts

    Security is in general a large and complex topic, and security in Kubernetes is no different. This is especially true…

  • Kubernetes-101: Ingress

    Kubernetes-101: Ingress

    In this article we will look at the Ingress resource. An Ingress exposes a Service on HTTP or HTTPS outside of the…

    1 条评论
  • Kubernetes-101: Helm

    Kubernetes-101: Helm

    We’re getting closer to the end of this series of Kubernetes-101 articles! In the summary of this article I will list…

  • Kubernetes-101: Pods, part 3

    Kubernetes-101: Pods, part 3

    In this article we will revisit the topic of Pods. Remember Pods? Of course you do, we have been using them all along.

社区洞察

其他会员也浏览了