Google Compute Engine Pricing

Google Compute Engine charges for usage on a monthly basis, using the following price sheet. A bill is sent out at the beginning of each month for the previous month's usage. Listed prices are in US dollars.

When charging in local currency, Google will convert the prices listed into applicable local currency pursuant to the conversion rates published by leading financial institutions.

Contents

1. Machine type pricing
   1. Sustained use discounts
   2. Instance uptime
2. Premium operating systems
3. Network pricing
   1. General pricing
   2. Load balancing and protocol forwarding
   3. Traffic through the external IP address of a virtual machine
   4. VPN
4. Disk pricing
1. Persistent disk pricing
2. Local SSD pricing
5. Image storage
6. IP address pricing
7. Viewing usage
8. Pricing calculator

Machine type pricing

Google Compute Engine currently offers the following machine types in the US, Europe, and Asia. The billing model for machine types is listed below, but Compute Engine also provides automatic discounts off these prices for sustained use. You can also use our Google Cloud Pricing Calculator to better understand price for different configurations.

Machine type billing model

1. All machine types are charged a minimum of 10 minutes.

   For example, if you run your virtual machine for 2 minutes, you will be billed for 10 minutes of usage.

2. After 10 minutes, instances are charged in 1 minute increments, rounded up to the nearest minute.

   An instance that lives for 11.25 minutes will be charged for 12 minutes of usage.

Sustained use discounts

If you run an instance for a significant portion of the billing month, you can qualify for a sustained use discount. When you use an instance for more than 25% of a month, Compute Engine automatically gives you a discount for every incremental minute you use for that instance. The discount increases with usage and you can get up to a 30% net discount for instances that run the entire month. Sustained use discounts are calculated and applied to your bill at the end of the month.

For computing sustained use, equivalently provisioned machines running non- concurrently are treated as Inferred Instances. This gives you the flexibility to start and stop instances freely and still receive the maximum sustained use discount available across all your instances.

Sustained use discounts are given on incremental use after certain usage thresholds are reached. This means that you still only pay for the hours (or minutes!) that you use an instance for, and Compute Engine automatically gives you the best price. There’s no reason to run an instance for longer than you need it.

The table below describes the discount at each usage level. These discounts apply for all instance types.

Usage Level (% of month)	% at which incremental is charged	Example incremental rate (USD/hour) for an n1-standard-1 instance
0%-25%	100% of base rate	$0.063
25%-50%	80% of base rate	$0.050
50%-75%	60% of base rate	$0.037
75%-100%	40% of base rate	$0.025

Note: Sustained use discounts only apply to machine type prices. All other charges are applied on top of the discounted machine type price. For example, premium operating systems are charged at an additional fee on top of any sustained use discounted machine type prices.

For example, if you run an n1-standard-1 instance for 75% of the month, your charges are calculated as follows:

• The first 25% will be charged at the full on-demand rate.
• The next 25% will be charged at a 20% discount off the on-demand rate.
• The last 25% will be charged at a 40% discount off the on-demand rate.

For this example, sustained use discounts resulted in a net discount of 20% for this instance. The table below demonstrates this example:

Usage level (% of month) for an n1-standard-1 instance	% of base rate usage charge	Example rate (USD/hour)	Calculated charges, assuming 30-day month
First 0-25% of usage	100%	$0.063 (base rate)	30 days x .25 = 7.5 days 7.5 days x 24 hours = 180 hours 180 hours x $0.063 USD/hour = $11.34
Usage between 25%-50% of the month	80%	$0.050	180 hours x $0.050 USD/hour = $9.00
Usage between 50%-75% of the month	60%	$0.0378	180 hours x $0.037 USD/hour = $6.66

The total charge for this instance is $11.34 + $9.00 + $6.66 = $27.

This charge will appear on your bill at the end of the month. Without sustained use discounts, the same n1-standard-1 instance running for the same amount of time would cost 20% more:

• 30 days per month x .75 = 22.5 days
• 22.5 days x 24 hours per day = 540 hours
• 540 hours x $0.063/USD hour = $34.02

The following graph demonstrates how your effective discount increases with use.

Click to enlarge

For example, if you use a virtual machine for 50% of the month, you get an effective discount of 10%, if you use it for 75% of the month, you get an effective discount of 20%, and if you use it for 100% of the month, you get an effective discount of 30%. You can also use the cloud Google Cloud Pricing Calculator to estimate your sustained use discount for any arbitrary workload.

Viewing sustained use discounts

Sustained use discounts will automatically appear on your bill at the end of each month as separate line items that are credited to your account. You can view this discount in the Google Developers Console billing history. These discounts will appear as credits, with "Sustained Use" in the description. For example, the billing history might look like the following for an n1-standard-1 instance with sustained use discounts:

Date	Description	Debits ($)	Credits ($)
Mar 1 - Mar 31	Google Compute Standard Intel N1 1 VCPU running in NA: 13567 Minutes (Project:1234567895)	$34.02
Mar 1 - Mar 31	Google Compute Standard Intel N1 1 VCPU running in NA Sustained Use Discount: 3381 Minutes (Project:1234567895)		$7.02

Inferred instances

When computing sustained use discounts, Compute Engine will give you the maximum discount available using inferred instances. An inferred instance combines multiple, non-overlapping instances of the same instance type in the same zone into a single instance for billing. With inferred usage, you are more likely to qualify for sustained use discounts.

The example below shows how a customer's usage, which in this case comprised of five distinct instances, is combined to find the smallest number of simultaneous running instances, in this example is three instances, each with the longest possible duration. We call each of these an "inferred instance". Compute Engine then calculates sustained use discounts based on the percentage of the time each of these inferred instances were running.

Instance uptime

Instance uptime is measured as the number of minutes between when you start an instance and when you stop an instance, the latter being when the instance state is TERMINATED. In some cases, your instance can suffer from a failure and be marked as TERMINATED by the system; in these cases, you will not be charged for usage after the instance reaches the TERMINATED state. If an instance is idle, but still has a state of RUNNING, it will be charged for instance uptime. The easiest way to determine the status of an instance is to use gcloud compute with the gcloud compute instances list command or to visit the Google Developers Console.

Instance uptime is rounded up to the nearest minute. For example, if you run two separate n1-standard-1 virtual instances for 14.5 minutes each, you will be billed for 15 minutes per instance, at the n1-standard-1 machine type pricing.

Note that Google Compute Engine bills for a minimum of 10 minutes of usage, so if you run an instance for 2 minutes of uptime, you are billed for 10 minutes. After 10 minutes, your instance is billed on a per minute basis. For more information, see the billing model.

Premium operating systems

Pricing for premium operating systems differ based on the machine type where the premium operating system image is used. For example, an f1-micro instance will be charged $0.02 per hour for a SUSE image, while an n1-standard-8 instance will be charged $0.11 per hour. All prices for premium operating systems are in addition to charges for using a machine type. For example, the total price of using an n1-standard-8 instance would be the sum of the machine type cost and the image cost:

n1-standard-8 cost + image cost = $0.504 + $0.11 = $0.614 per hour

Note: Pricing for premium operating systems are the same worldwide and do not differ based on zones or regions, as machine type prices do.

Red Hat Enterprise Linux (RHEL) images

• $0.06 USD/hour for instance types with less than 8 virtual cores
• $0.13 USD/hour for instance types with 8 virtual cores or more

All RHEL images are charged a 1 hour minimum. After 1 hour, RHEL images are charged by 1 hour increments, rounded up to the nearest hour. For example, if you run a RHEL image for 45 minutes, you will be billed for an hour. If you run an RHEL image for 63 minutes, you will be billed for 2 hours.

SUSE images

• $0.02 USD/hour for f1-micro and g1-small machine types
• $0.11 USD/hour for all other machine types

All SUSE images are charged a 10 minute minimum. After 10 minutes, SUSE images are charged by 1 minute increments, rounded up to the nearest minute.

Windows server images

• $0.02 USD/hour for f1-micro and g1-small machine types
• $0.04 USD per core/hour for all other machine types

Standard machine types, high CPU machine types, and high memory machine types are charged based on the number of cores. For example, n1-standard-4, n1-highcpu-4, and n1-highmem-4 are 4-core machines, and are charged at $0.16 USD/hour (4 x $0.04 USD/hour).

All Windows images are charged a 10 minute minimum. After 10 minutes, Windows images are charged by 1 minute increments, rounded up to the nearest minute.

Network pricing

General network pricing

Traffic type	Price
Ingress	No charge
Egress to the same zone
Egress to a different Google Cloud service within the same region
Egress to Google products (such as YouTube, Maps, Drive)*
Egress between zones in the same region (per GB)	$0.01
Egress between regions within the US (per GB)	$0.01*
Inter-continental Egress	At internet egress rate

Monthly Usage	Network (Egress) Worldwide Destinations (excluding China & Australia) (per GB)	Network (Egress) China Destinations (per GB)	Network (Egress) Australia Destinations (per GB)	Network (Ingress)
0-1 TB	$0.12	$0.23	$0.19	Free
1-10 TB	$0.11	$0.22	$0.18	Free
10+ TB	$0.08	$0.20	$0.15	Free

Load balancing and protocol forwarding

Hourly Service Charge (US$)		Price (per GB) of Data Processed
Up to 5 rules	$0.025	$0.008
Each additional rule	$0.010

Compute Engine charges for forwarding rules that are created for load balancing or protocol forwarding. A load balancing rule is a forwarding rule that is used to load balance a pool of instances and a protocol forwarding rule is a forwarding rule that is used to route packets of selective protocols to a single instance.

Up to 5 forwarding rules you create are charged at $0.025/hour. For example, if you create one forwarding rule, you will be charged $0.025/hour. If you have 3 forwarding rules, you will still be charged $0.025/hour. However, if you have 10 rules, you will be charged:

• 5 forwarding rules = $0.025/hour
• Each additional forwarding rule = $0.01/hour

$0.025/hour for 5 rules + (5 additional rules * $0.01/hour) = $0.075/hour

Traffic through the external IP address of a virtual machine

When you send traffic between virtual machines through the external IP addresses, it will be charged as follows:

• Traffic sent between the external IP address of two virtual machines in the same region will be charged as Egress between zones in the same region, even if the virtual machines are in the same zone.
• Traffic sent between the external IP address of two virtual machines in different regions within the US will be charged as Egress between regions within the US.
• Traffic sent between the external IP address of two virtual machines in different continents will be charged at Internet Egress pricing.

It is not possible to charge traffic through the external IP address of two virtual machines in the same zone as Egress to the same zone because Compute Engine cannot determine the zone of a virtual machine through the external IP address.

VPN

Component Billed	Price (US$)
Per tunnel	$0.05/hour*
Traffic through tunnels	Charged the same as if it wasn't in a VPN tunnel. See General network pricing.
Public IP for VPN Gateway	Charged according to IP address pricing.

^* Not charged during Beta.

Compute Engine does not charge for forwarding rules that send traffic to the VPN gateway.

Disk pricing

Persistent disk pricing

Persistent disks are charged for the amount of provisioned space per disk. Persistent disk I/O operations are included in the charges for provisioned space and persistent disk performance grows linearly to the size of the persistent disk volume, so you might want to create a larger or smaller persistent disk to account for your I/O needs. For more information, see the persistent disk and the v1 transition guide documentation.

Persistent disks are prorated based on a granularity of seconds. For example, a 200GB standard persistent disk volume would cost $8.00 for the whole month. If you only provisioned a 200GB volume for half a month, it would cost $4.00. Likewise, a 200GB SSD persistent disk volume would cost $65.00 for the whole month. If you only provisioned a 200GB volume for half a month, it would cost $32.50.

After you successfully delete a persistent disk, you will no longer be charged for that disk. If your persistent disk isn't accessible due to maintenance events, we won't charge for the time during which your persistent disk was inaccessible.

Type	Price (per GB/month)
Standard provisioned space	$0.04
SSD provisioned space	$0.17
Snapshot storage	$0.026
IO operations	No additional charge

Standard network pricing applies to snapshot restore. For example, a snapshot restore across regions or continents incurs the same fees as Google Compute Engine egress. Addition of egress fees for restore will be delayed as a promotional offer until April 1, 2015.

Persistent disk snapshots

Persistent disk snapshots are only charged for the total size of the snapshot. For example, if you only used 2TB of disk space on a 5TB persistent disk, your snapshot size will be around 2TB, rather than the full 5TB of provisioned disk space. Google Compute Engine also provides differential snapshots, which means that after the initial snapshot, subsequent snapshots only contain data that has changed since the previous snapshot, providing for a generally lower cost for snapshot storage.

Note that snapshot storage is also calculated based on a prorated rate.

Local SSD pricing

Local SSD devices are charged for the amount of provisioned space per device for the lifetime of the instance it is attached to, at a rate of $0.218 per GB per month. Since local SSD can only be purchased in 375GB increments, this comes to approximately $81.75 per device per month. Actual data storage and usage are included in that price and there is no additional charge for local traffic between the virtual machine and the local SSD device.

Type	Price (per GB/month)	Estimated monthly* cost per device**	Estimated hourly cost per device**/month*
Local SSD provisioned space	$0.218	$81.75	$0.113

Image storage

Type	Price
Image Storage (per GB/month)	$0.085

IP address pricing

Type	Price/Hour
Static IP address (assigned but unused)	$0.01
Static IP address (assigned and in use)	No charge
Ephemeral IP address (attached to instance or forwarding rule)	No charge

Note: If you want to remove a static IP address from your project, you can do so using the Addresses collection.

Viewing usage

The Google Developers Console provides a transaction history for each of your projects. This history describes your current balance and estimated resource usage for that particular project. To view a project's transaction history:

1. Log in to the Google Developers Console.
2. Select the project for which you would like to see the current usage and balance.
3. Click Billing & Settings in the navigation menu.

Pricing calculator

To help you understand how your bill is calculated, use the Google Cloud Pricing Calculator.

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 3.0 License, and code samples are licensed under the Apache 2.0 License. For details, see our Site Policies.