Serverless Containers pricing
What goes into the cost of using Serverless Containers
In Serverless Containers, you are billed for the number of container invocations, computing resources allocated for the application, downtime of provisioned instances, and outgoing traffic.
Computing resources are billed based on the number of processor cores (vCPUs), the level of core performance, memory allocated for the application, and application running time:
- The number of cores specified when creating a revision is measured in vCPUs and is a positive real number.
- The core performance level specified when creating a revision is measured in %.
- The memory size specified when creating a revision is measured in GB.
- The total container time is measured in hours and is rounded up to a multiple of 100 ms.
Warning
You are only charged for the container invocations that have caused your application to run.
Pricing formula
Monthly cost = ₽3.20 × Memory (GB) × Invocation processing time (Hours) + ₽4.80 × vCPU count × Invocation processing time (Hours) + ₽16.00 × Millions of invocations
The following is offered at no extra charge each month:
- The first 1000000 container invocations.
- The first 10 GBxhour of RAM usage when processing requests.
- The first 5 vCPU×hour of RAM usage when processing requests.
After you've used up your free service amounts, you will be charged at the applicable rate. The unused balance of free services is reset at the end of the month.
Examples of cost calculation
Example of calculating the cost of a container with the following parameters:
- Memory specified when creating a revision: 2 GB
- Number of cores: 20% vCPU
- Number of container invocations: 3,000,000
- Execution time of each call: 150 ms
3.2 × (2 × (150 / 3,600,000) × 3,000,000 – 10) + 4.8 × (0.2 × (150 / 3,600,000) × 3,000,000 – 5) + 16 × ((3,000,000 – 1,000,000) / 1,000,000)
Total: ₽896
Where:
- 3.2 is the cost of 1 GB×hour in excess of 10 GB×hour per month.
- 2 is the amount of RAM.
- 150 / 3,600,000 converts milliseconds to hours, as RAM usage time is calculated in GB×hour.
- 3,000,000 is the number of container invocations.
- 10 is the free of charge RAM usage time.
- 4.8 is the cost per hour of CPU usage in excess of 5 vCPU × hour per month.
- 0.2 is the number of vCPU cores.
- 150 / 3,600,000 converts milliseconds to hours, as CPU usage time is calculated in GB×hour.
- 3,000,000 is the number of container invocations.
- 5 is the free of charge CPU usage time.
- 16 is the cost of 1 million container invocations in excess of 1 million per month.
- 3,000,000 is the number of container invocations.
- 1,000,000 is subtracted because the first million invocations are free of charge.
- 1,000,000 is the divisor used to calculate the number of millions of container invocations.
Example of calculating the cost of a container with the following parameters:
- Memory specified when creating a revision: 2 GB
- Number of cores: 100% vCPU
- Number of container invocations: 3,000,000
- Execution time of each call: 150 ms
3.2 × (2 × (150 / 3,600,000) × 3,000,000 – 10) + 4.8 × (1 × (150 / 3,600,000) × 3,000,000 – 5) + 16 × ((3,000,000 – 1,000,000) / 1,000,000)
Total: ₽1,376
Where:
- 3.2 is the cost of 1 GB×hour in excess of 10 GB×hour per month.
- 2 is the amount of RAM.
- 150 / 3,600,000 converts milliseconds to hours, as RAM usage time is calculated in GB×hour.
- 3,000,000 is the number of container invocations.
- 10 is the free of charge RAM usage time.
- 4.8 is the cost per hour of CPU usage in excess of 5 vCPU × hour per month.
- 1 is the number of vCPU cores.
- 150 / 3,600,000 converts milliseconds to hours, as CPU usage time is calculated in GB×hour.
- 3,000,000 is the number of container invocations.
- 5 is the free of charge CPU usage time.
- 16 is the cost of 1 million container invocations in excess of 1 million per month.
- 3,000,000 is the number of container invocations.
- 1,000,000 is subtracted because the first million invocations are free of charge.
- 1,000,000 is the divisor used to calculate the number of millions of container invocations.
Pricing
Invoking a container
Service | Cost of 1 million calls, including VAT |
---|---|
Container calls, 1 million or less per month | Free |
Container calls, over 1 million per month | ₽16.00 |
RAM usage time when processing requests
Service | Price per GB×hour, including VAT |
---|---|
RAM usage time, 10 GB×hour or less per month | Free |
RAM usage time, over 10 GB×hour per month | ₽3.20 |
CPU usage time when processing requests
Service | Price per vCPU×hour, including VAT |
---|---|
CPU usage time, 5 vCPUs×hour or less per month | Free |
CPU usage time, over 5 vCPUs×hour per month | ₽4.80 |
Provisioned instances
RAM usage time
Service | Cost per GB×hour, including VAT |
---|---|
RAM usage time during downtime of a provisioned instance | ₽0.69 |
RAM usage time when processing queries in a provisioned instance | ₽1.32 |
CPU usage time
Service | Cost per vCPU×hour, including VAT |
---|---|
CPU usage time during downtime of a provisioned instance | ₽1.03 |
CPU usage time when processing queries in a provisioned instance | ₽1.98 |
Egress traffic
When using the service, you pay for traffic from Yandex Cloud to the internet. Traffic between internal IPs of Yandex Cloud services and incoming internet traffic is free.
The first 100 GB of outgoing traffic are provided free of charge every month.
The minimum billing unit is 1 MB.
Resource category | Cost of 1 GB |
---|---|
Outgoing traffic, 100 GB or less per month | Free |
Outgoing traffic, over 100 GB per month | ₽1.53 |