What is ParkMyCloud?

What is ParkMyCloud and how does it work?

ParkMyCloud is a cost-effective, lightweight app that allows cloud computing customers to pay only for the computing power they’re actually using. It does this by scheduling on/off times (also known as “parking”) for idle cloud computing services. As a result, ParkMyCloud reduces cloud computing costs up to 65% in just 15 minutes.

ParkMyCloud is SaaS-based, so there’s nothing to download and no installation required. You’re up and running in just three simple steps:

  1. Set up your ParkMyCloud account.(Free Trial. No credit card required.)
  2. Connect with your cloud service provider to discover your cloud computing resources.
  3. Start “parking” your cloud computing resources with scheduled “on/off” times.

Why do I need ParkMyCloud — what problem does it solve?

The popularity of cloud service providers like Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP) has exploded because they enable easy and rapid access to computing services with no upfront costs.

But hundreds of thousands of cloud customers are now realizing a downside to this ease of use: Cloud computing resources are “always on” unless customers specifically turn them off. Which means customers are paying for computing time they don’t actually use—for example, between 11:00 p.m. and 7:00 a.m. and on weekends.

ParkMyCloud is a simple, lightweight web app designed to solve this problem by scheduling on/off times (also known as “parking”) for your idle cloud computing resources, so you pay only for the resources you’re using.

How does "discovery" of my resources work?

When you create a ParkMyCloud account, you will connect to your cloud service provider (AWS, Microsoft Azure, or GCP). This will allow ParkMyCloud to metadata to extract and load metadata about your resources. ParkMyCloud uses this metadata to control control the state of your resources  (either running or parked), based upon the parking schedules you have applied. As you create new resources via your cloud service provider, they will appear in your ParkMyCloud dashboard.

What does ParkMyCloud do to my resources?

ParkMyCloud starts and stops resources, based upon a schedule you configure and apply. Your resources will not be terminated (except in the case of AWS Auto Scaling Groups). In AWS, ensure that you have selected stop, not terminate, as the default shutdown behavior:


In the AWS and GCP consoles, when the instances are parked, an orderly ACPI shutdown is used and they will be in the “stopped” state.

In Azure, and orderly ACPI shutdown is also used, but your resources will be in the “stop (deallocate)” state when parked.

Which cloud service providers do you support?

ParkMyCloud supports Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP).

Setting up ParkMyCloud

What is ParkMyCloud's pricing model?

You pay for ParkMyCloud monthly or yearly, based on the number of cloud resources discovered and managed in your account. See our pricing information learn more.

Do I have to download any software or plug-ins?

No. There’s no downloading and no plug-ins. ParkMyCloud is completely web-based, and can be accessed from your web browser.

Is there an upfront cost?

No. There are no upfront costs, and no fees, hidden or otherwise. The only cost for ParkMyCloud is the monthly subscription charge based on your pricing tier.

How long does the ParkMyCloud setup process take?

The ParkMyCloud setup process typically takes fifteen minutes or less. It’s simple:

  1. Set up your ParkMyCloud account.(Free Trial. No credit card required.)
  2. Connect with your cloud service provider to discover your cloud resources.
  3. Start “parking” your cloud computing resources with scheduled “on/off” times.

How do I know which cloud resources I should park?

We recommend you park non-production systems, such as development, testing, staging, and QA environments during off-hours. There also may be some production systems which are only used during specific times of the day or week for batch processing.

We suggest you name or tag these resources appropriately within your cloud service provider. You might do this with a standardized naming convention, such as “[server name]-dev” or “[server name]-QA”. Or perhaps you could use a tagging approach, such as “parkmycloud”:”yes”.

Within the ParkMyCloud dashboard, the names of your cloud resources will be displayed, and the tags can be revealed by selecting the tag icon.

ParkMyCloud identifies resources which may be eligible to be parked with a light bulb icon. The recommendations are driven by a default set of keywords. It searches all the instance metadata (including tags) and highlights likely candidates in the Keywords tab on the dashboard.  You may add/delete keywords to match them to your environments and naming structures.

Reducing Costs with Parking

How does ParkMyCloud estimate my savings from parking?

Parking savings is based upon the state of each resource you have ingested (either stopped or running), whether each has a parking schedule assigned to it, the percentage downtime for the chosen schedule and its estimated price per hour, based upon instance size and the region.

ParkMyCloud conservatively assumes the list price for each resource operating system. If the operating system is not available as part of the resource metadata, Linux pricing is assumed, which is the least expensive. If you use a different operating system, your savings may be greater.

Every hour, ParkMyCloud calculates the downtime duty cycle of each instance with a parking schedule attached, keeping a running tally of your parking savings — which can accumulate rather quickly.

What kind of AWS resources can ParkMyCloud park?

ParkMyCloud can park AWS EC2 instances, Auto Scaling Groups, and RDS instances.

What kind of Azure resources can ParkMyCloud park?

Currently, ParkMyCloud can park Microsoft Azure virtual machines (VMs).

What kind of GCP resources can ParkMyCloud park?

ParkMyCloud can park Google Compute Engine instances, both preemptible and non-preemptible, as well as Google Managed Instance groups, both with and without autoscaling.

When will I start seeing reductions in my cloud bill?

Once you have parked some of your cloud resources, you will immediately start cutting costs. ParkMyCloud will estimate those savings for you. You will see the actual savings on your next bill from your cloud service provider.

What does a typical parking schedule look like?

ParkMyCloud provides a few pre-created parking schedules for you to apply, and you also have the option to create your own. Typically, our users choose to park resources during the evenings and on weekends, for example, with off-times between 7:00 p.m. and 7:00 a.m. and off completely during weekends. See this example illustrated in the image below:


Some customers choose to use an always OFF schedule for all of their non-production servers, and have their developers snooze the schedule, when they need to get work done. (More on snoozing below.)

I like the parking schedules. What else can ParkMyCloud do?

Features that make it easy for you to manage your instances include:

  • Single-view Dashboard – See all your resources, across AWS/Azure/GCP accounts and locations.
  • SSO – Auto-provision users to a default team using SAML single sign-on (SSO).
  • Users & Teams – Govern user access to instances with permission-limited teams.
  • Policy Engine – Create rules to automatically assign resources to schedules or teams based on keywords.
  • Parking Recommendations – See which resources are recommended to park based on their tags or other metadata.
  • Reporting – View and download savings reports, organized by teams and credentials.
  • Public API – Easily integrate with automation and continuous delivery tools.
  • Audit Logs – Administrators can see a full log of all actions occurring in ParkMyCloud.

Can I park resources for teams in different time zones?

Yes. Each parking schedule allows the selection of a local time zone. For example, if you have a team in London, England you can select the correct GMT / UTC timezone and park instances for your UK team outside their local office hours. We recommend always adding a description to the schedules, for example, “Parked 23:00 – 06:00 London Time Monday to Sunday” to help when reviewing schedules.

What kind of return on investment or payback period can I realistically expect on ParkMyCloud?

A yearly subscription to ParkMyCloud typically pays for itself in less than two months, and often in less than one month. We recommend that you input your own information into the ParkMyCloud savings calculator to find your estimated payback period.

My Account

What access does ParkMyCloud need to my AWS/Azure/GCP account?

ParkMyCloud can manage resources in AWS, Azure, and GCP.  We take a minimal approach to the permissions needed for this access, so you don’t need to give us the “keys to the kingdom”.  ParkMyCloud can use IAM Roles or IAM Users to access AWS, Limited Access Roles to access Azure, or Service Accounts for GCP.  See more details on the exact permissions used for each of these and how to set them up.

What happens if ParkMyCloud service is disrupted? Can I still access my cloud services?

In the unlikely event that ParkMyCloud service were to be disrupted, yes, you would still be able to access your cloud resources directly through your cloud service provider. ParkMyCloud is an out-of-band solution.

What reporting does ParkMyCloud provide?

ParkMyCloud provides several downloadable reports of costs and savings by resource, by team, and by credential, as well as a team roster.

Can a ParkMyCloud account have multiple users? Can an administrator manage user permissions?

Yes. A ParkMyCloud account can have multiple users, organized in teams. User permissions are governed by user roles, managed by the account administrator(s).

How does ParkMyCloud handle public IP addresses in AWS?

ParkMyCloud follows the default behavior of AWS for public IP addresses when instances are stopped and started, whether that is done by parking schedules or by you in the AWS console. If your instances were automatically assigned public IP addresses, those addresses will be released when the instances are stopped. When those instance are restarted, they will get new public IP addresses, which will probably be different from the old ones.

If you wish to maintain the same public IP addresses when starting and stopping your instances, you will need to associate AWS Elastic IPs to those instances or use some other approach (e.g., scripting or automation tools). If you associate elastic IP addresses, then you need to be aware of the fact that there are two different behaviors, depending on whether you are using EC2-Classic or EC2-VPC. If you want the instances to retain the same IP addresses through parking (starting and stopping), then you must use EC2-VPC. That is, your instances must be launched in a VPC and must have an elastic IP addresses associated with them.

EC2-Classic is the old approach, which allows you to launch EC2 instances outside of Virtual Private Clouds (VPCs). In this situation if you associate elastic IPs to instances and then stop them, the elastic IPs will be disassociated.  We have outlined an approach creating a service on Ubuntu instances, which will automatically remap Elastic IP addresses to these Classic instances. We have posted the source code and instructions here on Github.  If others would like to extend this functionality to Amazon Linux, CentOS, RHEL and other flavors of Linux, please clone the repository and open a pull request to share your code contributions. 

What happens if I do a free trial and then do not subscribe to ParkMyCloud?

After the free trial ends, you have the option of continuing to use ParkMyCloud with either a paid or free subscription. If you use the free tier, you can continue to use ParkMyCloud for free for as long as you’d like.

If a valid cloud provider credential is not maintained, or the account is not in good standing for 30 days, your account will be deleted. Contact us with any further questions.


Copyright © ParkMyCloud 2016-2018. All rights reserved|Privacy Policy