# Glossary This glossary defines the terms that the [GraphQL API](/en/documentation/devtools/graphql/) uses in its queries, in its datasets, and across its documentation. | Term | Definition | | ------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | | adaptive resolver | The mechanism that sets the time interval of each row an aggregated query returns from the length of its time range. A range up to about two days returns minute intervals, a range from 60 hours returns hour intervals, and a range from about 60 days returns day intervals, as [GraphQL API limits](/en/documentation/devtools/graphql/limits/) describes. | | aggregated data | The data that the datasets of the `metrics` endpoint return in intervals of a minute, an hour, or a day, such as `workloadMetrics`. A query summarizes it with the `aggregate` functions and `groupBy`, as [Queries](/en/documentation/devtools/graphql/queries/) describes. | | calculated field | A field of a `metrics` dataset whose value results from a calculation, such as `requestsTotal` or `dataTransferredTotal`. A calculated field takes no `Lte` or `Gte` filter and cannot be the input of an `aggregate` function, and [Real-Time Metrics fields](/en/documentation/devtools/graphql/gql-real-time-metrics-fields/) marks each one. | | dataset | The top-level field of a query that names the set of records it reads, such as `workloadMetrics` or `workloadEvents`. Each endpoint serves its own datasets, and every dataset takes the `filter`, `aggregate`, `groupBy`, `orderBy`, `offset`, and `limit` arguments, as [Datasets and query arguments](/en/documentation/devtools/graphql/features/) lists. | | financial data | The billed and accounted data of an account, which the datasets of the `billing` endpoint, such as `billDetail`, and the `accountingDetail` dataset of the `accounting` endpoint return. These datasets have no `ts` field, so a query filters them by period fields such as `periodFrom` and `periodTo`, as [Queries](/en/documentation/devtools/graphql/queries/) describes. | | GraphiQL Playground | The in-browser GraphiQL editor where you write, validate, and run queries against an endpoint of the GraphQL API. Each endpoint URL serves it to an authenticated request, as [GraphiQL Playground](/en/documentation/devtools/graphql/graphql-playground/) describes. | | `groupBy` | The query argument that lists the fields whose values split the result into rows, each with its own `aggregate` output. It is optional: an `aggregate` with no `groupBy` returns one total row, and a `resample` needs `ts` in `groupBy`, as [Queries](/en/documentation/devtools/graphql/queries/) describes. | | `limit` | The query argument that sets how many rows a query returns, from `0` to `10000`, with `10` as the default. A value outside that range returns a `400` error, as [GraphQL API limits](/en/documentation/devtools/graphql/limits/) describes. | | metadata | The information about the scope of the data the GraphQL API serves: its datasets, their fields, and the name, description, and type of each. An introspection query returns it, as the guide to [query metadata with the GraphQL API](/en/documentation/guides/platform/observability/graphql-metadata/) describes. | | `offset` | The query argument that sets how many rows a query skips before the first row it returns, with `0` as the default. With `limit`, it pages through a result: `offset: 5` with `limit: 5` returns rows 6 to 10, as [Datasets and query arguments](/en/documentation/devtools/graphql/features/) describes. | | operator | A suffix on a field name inside `filter` that sets how the field compares to a value, such as `hostLike` or `statusGt`. A bare field name, such as `host`, works as `Eq`, and [Datasets and query arguments](/en/documentation/devtools/graphql/features/) lists the operators each field takes. | | raw data | The individual event records, without aggregation, that the datasets of the `events` endpoint return, such as `workloadEvents`. Events stay available for about 7 days, as [GraphQL API limits](/en/documentation/devtools/graphql/limits/) describes. | | `resample` | The query argument that groups the rows of a `metrics` dataset into time intervals of the range divided by `points`, and combines each interval with a `function`: `sum`, `max`, `min`, or `mean`. It needs `ts` in `groupBy`, and no `events` dataset takes it, as [Datasets and query arguments](/en/documentation/devtools/graphql/features/) describes. | | top X query | A query that ranks the values of one field by how often they occur. It counts the field with `aggregate`, groups by it with `groupBy`, orders by `count_DESC`, and keeps the first rows with `limit`, as the guide to [select top X queries](/en/documentation/guides/platform/observability/graphql-top-x-query/) shows. | | `tsRange` | The filter that sets the time interval of a query, with `begin` and `end` timestamps in the same timezone. A query on a `metrics` or `events` dataset without a time interval returns a `400` error, and `tsGt` or `tsLt` sets one bound only, as [Datasets and query arguments](/en/documentation/devtools/graphql/features/) describes. | | usage data | The consumption data of an account per workload, product, and metric, which the `workloadConsumptionMetrics` dataset of the `consumption` endpoint returns. Its `accounted` field carries the accounted value, as [Consumption fields](/en/documentation/devtools/graphql/gql-consumption-fields/) describes. |