execution_report.schema.json

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://elixir-europe.org/platforms/compute/execution_report.schema.json",
  "title": "Tool execution report",
  "description": "Gives credit for the specific execution of a data analysis tool.",
  "type": "object",
  "required": [ "report_format_version", "tool_identifier", "tool_name", "infra" ],
  "properties": {
    "report_format_version": {
      "description": "The version of the execution schema used.",
      "enum": [ "0.0.2" ]
    },

    "affiliation": {
      "description": "Identifier of the project associated with this tool execution. Ideally from the Research Organization Registry (ROR) https://ror.org/.",
      "type": "array",
      "items": {
        "type": "string",
        "format": "iri"
      }
    },

    "tool_identifier": {
      "description": "Tool identifier from bio.tools, an RRID, or other source. Multiple identifiers can be supplied for tools known to multiple registries.",
      "type": "array",
      "items": {
        "type": "string",
        "format": "iri"
      },
      "minItems": 1
    },
    "tool_name": {
      "description": "What the tool author calls the tool.",
      "type": "string"
    },
    "tool_version": {
      "description": "Version of the tool that was actually run, ideally captured from the tool itself or a packaging system (apt, conda, pip, etc..).",
      "type": "string"
    },
    "tool_package_version": {
      "description": "Version of the tool as reported from a packaging system (apt, conda, pip, etc..).",
      "type": "string"
    },
    "input_size": {
      "description": "The total size of the inputs, MUST be (rounded or binned, details TBD).",
      "type": "integer"
    },

    "infra": {
      "description": "Physical and virtual infrastructure providers involved in the coordination, planning, and/or execution of this tool. Could be platforms, service providers, etc.",
      "type": "array",
      "minItems": 1,
      "items": { "$ref": "#/$defs/provider" }
    },
    "location": {
      "$ref": "#/$defs/location"
    },
    "cpu_identifier": {
      "description": "For example, the 'model name' field from /proc/cpuinfo on Linux.",
      "type": "string"
    },
    "cpu_mods": {
      "description": "For example, the 'bugs' field from /proc/cpuinfo on Linux.",
      "type": "string"
    },
    "cpu_cores_assigned": {
      "description": "Number of CPU cores that were assigned to the tool.",
      "type": "integer"
    },

    "start_time": {
      "description": "Time & date of the beginning of tool execution.",
      "type": "string",
      "format": "date-time"
    },
    "end_time": {
      "description": "Time & date of the end of tool execution.",
      "type": "string",
      "format": "date-time"
    },
    "duration": {
      "description": "Duration of tool execution. Using start_time & end_time is preferred.",
      "type": "string",
      "format": "duration"
    },

    "final_outputs_size": {
      "description": "Disk usage (bytes), MUST be (rounded or binned, details TBD).",
      "type": "integer"
    },
    "memory_used": {
      "description": "Peak memory usage, measured in bytes.",
      "type": "integer",
      "exclusiveMinimum": 0
    },
    "cpu_cores_used": {
      "description": "Measured number of CPU cores used.",
      "type": "number"
    },
    "gpu_cores_used": {
      "description": "Measured number of GPU cores used.",
      "type": "integer"
    }
  },
  "$defs": {
    "address": {
      "title": "Address",
      "type": "object",
      "$comment": "https://schema.org/PostalAddress",
      "properties": {
        "addressCountry": {
          "description": "Country. 2-letter ISO 3166-1 alpha-2 format.",
          "$comment": "https://schema.org/addressCountry",
          "type": "string"
        },
        "postalCode": {
          "description": "The postal code. For example, 94017.",
          "$comment": "https://schema.org/postalCode",
          "type": "string"
        },
        "addressRegion": {
          "description": "The region in which the locality is, and which is in the country. For example, Baden Wuerttemburg or another appropriate first-level Administrative division. ",
          "$comment": "https://schema.org/addressRegion",
          "type": "string"
        },
        "addressLocality": {
          "description": "The locality in which the street address is, and which is in the region. For example, Freiburg im Breisgau",
          "$comment": "https://schema.org/addressLocality",
          "type": "string"
        }
      }
    },
    "location": {
      "title": "Physical location",
      "description": "Physical location where the computer was. Don't really care more than city/country.",
      "$comment": "https://schema.org/Place",
      "type": "object",
        "properties": {
          "name": {
            "description": "The name of the physical location where the computation took place.",
            "type": "string",
            "$comment": "https://schema.org/Text"
          },
	  "address": { "$ref": "#/$defs/address" }
      }
    },
    "provider": {
      "title": "Infrastructure provider",
      "type": "object",
      "required": [ "infra_name", "infra_identifier" ],
      "properties": {
        "infra_name": {
          "description": "Name of the infrastructure (who should get credit).",
          "type": "string"
        },
        "infra_identifier": {
          "description": "Unique identifier for this infrastructure.",
          "type": "array",
          "items": {
            "type": "string",
            "format": "iri"
          },
          "minItems": 1
        },
        "infra_version": {
          "description": "For software-based infrastructure: the version.",
          "type": "string"
        }
      }
    }
  }
}