Pricing

Goldsky allows you to pay for exactly what you use. There is also a generous free tier for developers and hobbyists.

We track usage based on two metrics - the amount of active subgraphs, and the amount of data stored across all subgraphs in your project.

The amount of active subgraph workers, tracked hourly. If you pause or delete a subgraph, it no longer gets billed.

Examples:

1. If you have 10 subgraphs, you will be using 10 subgraph worker hours every hour. At 730 hours in the average month, with 10 subgraphs you would incur 7,300 subgraph worker hours.
2. If you start with 10 subgraphs in a billing period and delete all of them in the middle of the billing period, you are charged for 5 subgraphs that billing period.

The number of entities stored across all the subgraphs in your project. This is tracked hourly as well. If you delete a subgraph, it no longer gets tracked. All subgraph entities in a project count towards the projects usage cumulatively.

Examples:

1. If you have 3 subgraphs that have 30,000 entities collectively, you will be using 30,000 subgraph entity storage hours every hour. At 730 hours in the average month, you will incur 30,000 * 730 = 21,900,000 subgraph entity storage hours in that month.
2. If you start with 3 subgraphs, each with 10,000 entities, and you delete two of them after 10 days, you will be using 30,000 entity subgraph hours for the first 10 days, then 10,000 entity subgraph hours after that.

Up to 3 concurrent subgraphs a month, or more precisely, 2250 worker hours.

Up to 100,000 entities stored for a whole month, or more precisely, 75,000,000 entity storage hours.

You will be incurring usage for each hour that each subgraph in your project is deployed and active. If you have 2 subgraphs deployed and active for 2 hours each, you will accumulate 4 hours of usage.

When you exceed free tier limits in the Starter plan, subgraph indexing will be paused but they will still be queryable.

The number of active workers, billed hourly. A small pipeline (the default) will be 1 worker. Pipelines can have multiple parallel workers, and each worker incurs usage separately. This is billed hourly.

If you have a small pipeline and a large pipeline both deployed for 2 hours, you will accumulate 2*1 + 2*16 = 2 + 32 = 34 hours of usage.

Note that pipelines with a single subgraph as a source, and webhooks or graphql APIs as sink(s) are not metered as pipelines. You will still however accumulate hourly subgraph usage.

Examples:

1. If you have 1 S-sized pipeline, you will be using 1 pipeline worker-hour every hour. At 730 hours in the average month, with that pipeline you would incur 730 pipeline worker-hours over the course of the month.
2. If you start with 10 such pipelines in a billing period and delete all of them in the middle of the billing period, you are charged the equivalent of 5 pipeline workers for the full billing period.
3. If you have 2 L-sized pipelines, you will be using 20 pipeline worker-hours every hour, equating to 14,600 pipeline worker-hours if you run them the entire month.

The number of records written by pipelines in your project. For example, for a PostgresQL sink, every row created, updated, or deleted, counts as a ‘write’. For a Kafka sink, every message counts as write.

Example:

1. If you have a pipeline that writes 20,000 a day for 10 days, then 20 a day after that for 10 days after that, you will be using 200,200 pipeline event writes.
2. If you have two pipeline that have written 1 million events each during a month, then you will be charged $0 for the first million events, then $1 for the next million, as the free tier would be exhausted.

You are able to run 1 s-sized pipeline for free (equating to 750 pipeline hours) each billing cycle.

You are able to write up to 1 million events to a sink, cumulatively across all pipelines, per each billing cycle.

When you exceed free tier limits in the Starter plan, pipelines will be paused but they will still be queryable.

You will be incurring usage for each hour that each pipeline in your project is deployed and active. One thing to note is that pipelines have a notion of resource size (this maps to the underlying VM size) which is a multiplier on the hourly usage.

We want to give our customers a usable free tier. We started with the idea of giving our users 3 free concurrent subgraphs per month with up to 100,000 free subgraph entities per month. Similarly, we want to give our users 1 free pipeline per month with up to 1,000,000 free pipeline entities per month.

One tricky thing about our pricing model are the numbers for how we measure subgraph entities. This arises from the fact that we conceptualize our billing on a monthly basis but we bill on an hourly basis. Unfortunately, there aren’t a consistent number of hours in a month, some months have 672 hours (February) and others have 744 (August). We decided to assume that each month has 750 hours. This assumption is what underlines our Subgraph Entity Count billing table.

In otherwords, the estimated monthly pricing you see is more expensive than you would actually pay given the same monthly usage.

From this there are two important concepts for us to call out: entity-months and entity-hours. An entity-month is an entity that has been active for an entire month. An entity-hour is an entity that has been active for an hour.

To give our everyone 100,000 free subgraph entity-months, with hourly snapshotting, we multiplied 100,000 * 750 = 75,000,000 to get our free tier. This means that you can run a subgraph with exactly 100k entities for an entire month without exceeding the free tier.

Beyond the free tier, we want to charge $4.00 per 100k entity-months up to 10 million entity-months. This comes out to 4 / 750 / 100000 = 0.000000053333 per entity-hour up to 10,000,000 * 750 = 7,500,000,000 entity-hours.

Similarly, when exceeding 10 million entity-months (or 7,500,000,000 entity-hours), we wanted to charge $1.00 per 100k entity-months. The math on this works out to 1 / 750 / 100000 = 0.000000013333 per entity-hour.