README.md

fmu-to-resqml

This is a service interacting with fmu-sumo and resqpy to convert data from FMU (Fast Model Update™) to RESQML format.

Want to contribute? Read our contributing guidelines

---

API

Requests:

• POST - Converting and retrieving an ensemble of realizations within a case

• GET - Converting and retrieving one object

• POST - Converting and retrieving several objects

Endpoints:

• url/ensemble/ - Ensemble converting and retrieving a single EPC and HDF5 file zipped together.

• url/ensemble/epc - Ensemble converting and retrieving a single EPC file.

• url/ensemble/hdf - Ensemble converting and retrieving a single HDF5 file.

• url/objects/ - Object converting and retrieving both EPC and HDF5 files zipped.

• url/objects/epc - Object converting and retrieving only EPC files. Will be zipped if using POST.

• url/objects/hdf - Object converting and retrieving only HDF5 files. Will be zipped if using POST.

Parameters:

Ensembles:

Only uses POST - Add case uuid and filter specifications as parameters in request body using JSON object format:

{
    "uuid" : "<case_uuid>",
    "iter" : ["<iteration_1>", "<iteration_2>"],
    "tags" : ["<tag_name_1>", "<tag_name_2>", "<tag_name_3>"],
    "name" : ["<name_1>", "<name_2>", "<name_3>"]
} 

Note: name field can be left blank, and will then return realizations regardless of name, following iteration and tagname filter.

Objects:

Using GET - Add object uuid as parameter in request arguments using url/.../?uuid=<object_uuid>.

Using POST - Add object uuids as parameters in request body using json object format:

{ 
    "uuids" : ["<object_1_uuid>", "<object_2_uuid>", "<object_3_uuid>"] 
}

---

Known Issues !!

There currently exists and issue with the ensemble endpoint.

When requesting smaller ensembles the service works as expected, like; (this one contains 5 objects)

{
  "uuid": "c616019d-d344-4094-b2ee-dd4d6d336217",
  "iter": [
    "iter-0"
  ],
  "tags": [
    "DS_extract_geogrid",
    "apstrend_aps_Crevasse_Average"
  ],
  "name": [
  ]
}

After some wait we get a zip file with a single EPC and HDF5 file back.

However, we end up with a 504 Gateway Time-out error when requesting larger ensembles, like; (this one contains 155 objects)

{
  "uuid": "b90a87fb-6766-457b-9d3b-5ecd442c3e89",
  "iter": [
    "iter-0"
  ],
  "tags": [
    "DS_extract_geogrid"
  ],
  "name": [
    "VOLANTIS GP. Top"
  ]
}

This is most likely due to cases becoming very large, and resqpy taking a long time converting and formatting the whole ensemble into a single epc file.

For reference, these are the error messages;