/benchmarking-data-model

OpenEBench Benchmarking Data Model repository

Primary LanguageHTMLCreative Commons Attribution Share Alike 4.0 InternationalCC-BY-SA-4.0

OpenEBench Benchmarking Data Model repository

(born from EXCELERATE WP2)

This repository contains all the developments around the OpenEBench benchmarking data model which was borned from ELIXIR-EXCELERATE WP2.

  • Folder early-drafts contains the initial ideas around what it is needed to be stored in order to model benchmarking events and metrics.

  • Folder json-schemas contains the the JSON Schemas, based on JSON Schema Draft 04 standard, being developed to validate the different objects used by ongoing WP2 benchmarking services and databases. These schemas contain extensions to declare the primary and foreign keys of the documents, when all the documents are validated as a set, or are stored in a database.

    • Primary key is declared on the top of each schema as an array with the key primary_key. A primary key can be composed by more than one property.
    • Foreign keys are declared with keys foreign_keys (which is an object) at any schema level. They map the declared properties to the JSON Schema key referenced by the schema_id property, which is either a relative or absolute URI. When it is relative, it is resolved against the id value of the schema where the foerign key is declared. The members attribute is an array, and it must contain the list of attributes (or the path to them, separated by dots) which map to the ones in primary_key declaration in the referenced schema. Attribute order does matter. When the foreign_keys declaration points to a primary key with a single attribute, it can be included at atomic value level, and use the special attribute name ..

  • Folder prototype-data contains several JSON data sets, which validate against the benchmarking JSON schemas.

  • Both fairtracks_validator and extended-json-schema-validators contain several reference programs, which validate the JSON data sets against the benchmarking JSON schemas. See Tools for validation in order to learn their lineage.