Cluster costs

There are no set-up charges or commitments to begin using Yugabyte Cloud. At the end of the month, you are automatically charged for that month’s usage. You can view your charges for the current billing period at any time by selecting Invoices on the Billing tab of the Admin page. Refer to Manage your billing profile and payment method.

Only the cloud account Admin user (the user who created the Yugabyte Cloud account) can access the billing and payment details.

Your bill is calculated based on your usage of the following cloud dimensions:

• Instance vCPU capacity
• Disk storage
• Backup storage
• Data transfer

Instance vCPU capacity makes up the majority of your bill, and is the easiest to understand and control. It's purely a function of your total number of vCPUs used and how long they have been running. The cluster's per-hour charge includes free allowances for disk storage, backup storage, and data transfer. If you use more than the free allowance, you incur overages on top of the base vCPU capacity cost.

Dimension	Allowance/vCPU per month
Disk storage	50 GB
Cloud backup storage	100 GB
Data transfer – Same Region	1000 GB
Data transfer – Cross Region (APAC)	10 GB
Data transfer – Cross Region (Other regions)	10 GB
Data transfer – Internet	10 GB

You can see the approximate cost for your vCPUs when creating and scaling clusters, as shown in the following illustration. + Usage in this context refers to any potential overages from exceeding the free allowances for disk storage, backup storage, and data transfer.

Instance vCPU capacity costs

Instance vCPU capacity cost is the cost for the use of the total number of vCPUs in your cloud account.

You can reduce the cost by reducing the number of vCPUs, which may have a negative impact on performance.

Calculating instance minutes

Yugabyte measures vCPU use in “Instance-Minutes,” which are added up at the end of the month to generate your monthly charges. The total instance capacity cost across all your clusters is the sum of instance-minutes across all clusters multiplied by the base per minute rate card ($.00416666666/minute).

Pricing is per instance minute consumed for each instance, from the time an instance is launched until it is terminated.

Assume you start a cluster with 3 nodes x 2 vCPUs (6 vCPUs) for the first 15 days in September, and then scale up to 6 nodes x 2 vCPUs (12 vCPUs) for the final 15 days in September.

At the end of September, you would have the following usage in instance-minutes:

[(6 vCPUs * 15 days * 24 hours * 60 min) + (12 vCPUs * 15 days * 24 hours * 60 min)]
= 388800 instance-minutes

Total vCPU cost/month = Total instance minutes * Per minute base rate
Total vCPU cost/month = 388800 * $.00416666666 ~ $1619.99

Disk storage cost

Disk storage costs are tied to the cost of storing the data on disk in the underlying IaaS storage (for example, EBS on AWS).

The free allowance for disk storage is 50 GB/month for every 1 vCPU per month used in a cluster. Whenever you exceed the 50 GB/month/vCPU threshold, you are billed for the storage used in excess of the free allowance.

You can also specify a custom value greater than free allowance storage capacity. You can customize your cluster storage capacity independently of your cluster vCPU capacity. If you customize an amount of disk storage greater than the free allowance, you are only charged for the amount exceeding the free allowance.

For example, a 3 node x 2 vCPU (6 vCPUs) cluster includes a total free allowance of 300 GB/month (6 vCPUs * 50 GB), which is equally distributed across 3 nodes at 100 GB each. For the same cluster, if you increase per node storage capacity to 150 GB, then your total disk storage will be 450 GB (3 nodes x 150 GB) but you are only charged for the 150 GB above your 300 GB allowance.

Disk storage size is calculated by metering the storage space (GBs) occupied per cluster. The same unit price applies to all regions and clouds.

Calculating disk storage cost

Yugabyte measures disk storage in “GB-hours,” which are added up at the end of the month to generate your monthly charges. The total disk storage capacity cost across all your clusters is the total number of GB-hours multiplied by the base per hour rate card ($0.0001388888889/hr), less the total free allowance based on per month vCPU usage.

Assume you start a cluster for the first 15 days of September with the following configuration:

• Total number of vCPUs: 3 nodes * 2 vCPU = 6 vCPUs
• Disk storage/node: 100 GB
• Total disk storage: 300 GB

Then scale up to the following configuration for the final 15 days in September:

• Total number of vCPUs: 3 nodes * 4 vCPU = 12 vCPUs
• Disk storage/node: 500 GB
• Total disk storage: 1500 GB

At the end of September, you would have the following total usage cost:

Total disk storage
[(300 GB * 15 days * 24 hours) + (1500 GB* 15 days * 24 hours)]
= 648000 GB-hours

Total instance-minutes
[(6 vCPUs * 15 days * 24 hours * 60 min) + (12 vCPUs * 15 days * 24 hours * 60 min)]
= 388800 instance-minutes

Total vCPUs
388800 instance-minutes / ( 30 days * 24 hours * 60 minutes )
= 9 vCPUs

Free allowance (GB/month)
9 vCPUs * 50 GB/month = 450 GB

Free allowance (GB-hours)
450 GB * 30 days * 24 hours = 324000 GB-hours

Disk storage overages
648000 GB-hours - 324000 GB-hours = 324000 GB-hours

Total disk storage cost/month = Total overages (GB-hours) * Per hour base rate
Total disk storage cost/month = 324000 * 0.0001388888889 = $45

Backup storage costs

Backup storage costs are tied to the cost of storing the backup snapshots in the underlying IaaS storage services (that is, S3 on AWS or GCS on Google cloud). It's purely a function of total data backed up from your cluster and the retention period.

The free allowance for backup storage is 100 GB/month for every 1 vCPU per month used in a cluster. Whenever you exceed the 100 GB/month/vCPU threshold, you are billed for the backup storage used in excess of the free allowance. For example, a 3 node x 2 vCPU (6 vCPUs) cluster includes a total free allowance of 600 GB/month (6 vCPUs x 100 GB).

By default, every cluster is configured with 24 hour backups with an 8 day retention period. You can customize your backup schedule and retention period per cluster. Taking frequent backups and retaining for a long period of time can lead to overages. Refer to Back up clusters.

Backup storage size is calculated by metering the storage space (GBs) occupied per cluster. The same unit price applies to all regions and clouds.

Calculating backup storage cost

Yugabyte measures backup storage in “GB-hours,” which are added up at the end of the month to generate your monthly charges. The total backup storage capacity cost across your clusters is the total number of GB-hours multiplied by the base per hour rate card ($ 0.00003472222222/hr), less the total free allowance based on per month vCPU usage.

Assume you start a cluster with 3 nodes x 2 vCPUs (6 vCPUs) for the first 15 days in September, and then scale up to 6 nodes x 2 vCPUs (12 vCPUs) for the final 15 days in September.

At the end of September, you would have the following total backup cost:

Assume an actual backup usage of 720000 GB-hours.

Total instance-minutes
[(6 vCPUs * 15 days * 24 hours * 60 min) + (12 vCPUs x 15 days x 24 hours x 60 min)]
= 388800 instance-minutes

Total vCPUs
388800 instance-minutes / ( 30 days * 24 hours * 60 minutes )
= 9 vCPUs

Free allowance (GB-month)
9 vCPUs x 100 GB/month = 900 GB

Free allowance (GB-hours)
900 GB x 30 days x 24 hours = 648000 GB-hours

Backup storage overages
720000 GB-hours - 648000 GB-hours = 72000 GB-hours

Total backup storage cost/month = Total overages (GB-hours) x Per hour base rate
Total disk storage cost/month = 72000 x 0.00003472222222 = $2.5

Data transfer costs

Data Transfer accounts for the volume of data going into, out of, and between the nodes in a cluster, which is summed up to a cumulative amount within a billing cycle.

Yugabyte meters and bills data transfer using the following three dimensions.

Same region

This accounts for all regional traffic of the cluster. This includes all cross availability zone inter-node traffic, which YugabyteDB automatically manages, and egress cost to a client in the same region as the cluster.

Single-node (fault tolerance of NONE) and three-node (fault tolerance of Node Level) with single availability zone (AZ) topologies will have much lower usage than clusters with three nodes (fault tolerance of Availability Zone) deployed across multiple AZs.

The free allowance for same region transfers is 1000 GB per month for every 1 vCPU per month used in a cluster. You are charged for any data transfer used in excess of the free allowance.

Cross region

This accounts for all of the traffic coming out of the cluster to a different region. This happens if a client is using VPC peering but is in different region than the cluster deployments. Different rate cards apply for clusters deployed in Asia-Pacific (APAC) vs other regions.

The free allowance for cross region transfers is 10 GB per month for every 1 vCPU per month used in a cluster. You are charged for any data transfer used in excess of the free allowance.

Data out (Internet)

This accounts for all of the traffic coming out of the cluster to the internet. This happens when a client is not using VPC peering and connecting to the cluster over the internet.

The free allowance for data out transfers is 10GB per month for every 1 vCPU per month used in a cluster. You are charged for any data transfer used in excess of the free allowance.

Controlling data transfer costs

• Ensure that queries originate from the same cloud region and provider as your database deployment whenever possible.

  • Avoid data out internet costs by using VPC peering and not allowing any client applications to connect over the public internet.

  • Minimize data out cross region costs by making sure your client application and database cluster are deployed in the same cloud and region and connected using VPC Peering.

• Ensure that queries do not:

  • Re-read data that already exists on the client.
  • Re-write existing data to your database deployment.

• If possible, configure your client driver to use wire protocol compression to communicate with the YugabyteDB cluster. Yugabyte Cloud always compresses intra-cluster communication.