Account for per-job output storage for budget estimations

app/grandchallenge/challenges/forms.py

@@ -405,6 +405,7 @@ class Meta:
             "algorithm_selectable_gpu_type_choices_for_tasks",
             "algorithm_maximum_settable_memory_gb_for_tasks",
             "average_size_test_case_mb_for_tasks",
+            "average_size_job_output_mb_for_tasks",
             "inference_time_average_minutes_for_tasks",
             "task_id_for_phases",
             "number_of_teams_for_phases",
@@ -436,6 +437,7 @@ def __init__(self, *args, **kwargs):
                 "algorithm_selectable_gpu_type_choices_for_tasks",
                 "algorithm_maximum_settable_memory_gb_for_tasks",
                 "average_size_test_case_mb_for_tasks",
+                "average_size_job_output_mb_for_tasks",
                 "inference_time_average_minutes_for_tasks",
                 css_class="border rounded px-2 my-4",
             ),
@@ -477,6 +479,7 @@ def _clean_task_lists_equal_length(self, cleaned_data):
             "algorithm_selectable_gpu_type_choices_for_tasks",
             "algorithm_maximum_settable_memory_gb_for_tasks",
             "average_size_test_case_mb_for_tasks",
+            "average_size_job_output_mb_for_tasks",
             "inference_time_average_minutes_for_tasks",
         ):
             field_value = cleaned_data.get(field_name)

app/grandchallenge/challenges/migrations/0072_challengerequest_average_size_output_mb_for_tasks.py

@@ -0,0 +1,36 @@
+# Generated by Django 5.2.13 on 2026-04-21 11:04
+
+from django.db import migrations, models
+
+import grandchallenge.core.validators
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("challenges", "0071_alter_challengerequest_status_and_more"),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name="challengerequest",
+            name="average_size_job_output_mb_for_tasks",
+            field=models.JSONField(
+                default=list,
+                help_text="Average size of the output per job in MB, for each task.",
+                validators=[
+                    grandchallenge.core.validators.JSONValidator(
+                        schema={
+                            "$schema": "http://json-schema.org/draft-07/schema",
+                            "items": {
+                                "maximum": 10000,
+                                "minimum": 0,
+                                "type": "integer",
+                            },
+                            "type": "array",
+                        }
+                    )
+                ],
+            ),
+        ),
+    ]

app/grandchallenge/challenges/migrations/0073_set_average_size_job_output_for_tasks.py

@@ -0,0 +1,29 @@
+from django.db import migrations
+
+
+def set_average_size_job_output_mb_for_tasks(apps, schema_editor):
+    ChallengeRequest = apps.get_model(  # noqa: N806
+        "challenges", "ChallengeRequest"
+    )
+
+    for challenge_request in ChallengeRequest.objects.all():
+        challenge_request.average_size_job_output_mb_for_tasks = [
+            0 for _ in challenge_request.task_ids
+        ]
+        challenge_request.save()
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        (
+            "challenges",
+            "0072_challengerequest_average_size_output_mb_for_tasks",
+        ),
+    ]
+
+    operations = [
+        migrations.RunPython(
+            set_average_size_job_output_mb_for_tasks, elidable=True
+        ),
+    ]

app/grandchallenge/challenges/models.py

@@ -990,6 +990,7 @@ class ChallengeRequestStatusChoices(models.TextChoices):
     "algorithm_selectable_gpu_type_choices_for_tasks",
     "algorithm_maximum_settable_memory_gb_for_tasks",
     "average_size_test_case_mb_for_tasks",
+    "average_size_job_output_mb_for_tasks",
     "inference_time_average_minutes_for_tasks",
     "task_id_for_phases",
     "number_of_teams_for_phases",
@@ -1183,6 +1184,23 @@ class ChallengeRequest(UUIDModel, ChallengeBase):
             )
         ],
     )
+    average_size_job_output_mb_for_tasks = models.JSONField(
+        help_text="Average size of the output per job in MB, for each task.",
+        default=list,
+        validators=[
+            JSONValidator(
+                schema={
+                    "$schema": "http://json-schema.org/draft-07/schema",
+                    "type": "array",
+                    "items": {
+                        "type": "integer",
+                        "minimum": 0,
+                        "maximum": 10000,
+                    },
+                }
+            )
+        ],
+    )
     number_of_submissions_per_team_for_phases = models.JSONField(
         help_text="Number of submissions per team for each phase",
         default=list,
@@ -1545,6 +1563,13 @@ def average_size_test_case_mb_for_phases(self):
             for task_index in self.task_index_for_phases
         ]
 
+    @property
+    def average_size_job_output_mb_for_phases(self):
+        return [
+            self.average_size_job_output_mb_for_tasks[task_index]
+            for task_index in self.task_index_for_phases
+        ]
+
     @cached_property
     def number_of_submissions_for_phases(self):
         return [
@@ -1601,15 +1626,35 @@ def total_compute_time(self):
 
     @cached_property
     def data_storage_size_gb_for_phases(self):
-        return [
-            round(n_images * image_mb * settings.MEGABYTE / settings.GIGABYTE)
-            for n_images, image_mb in zip(
+        input_size_mb_for_phases = [
+            n_images * input_mb
+            for n_images, input_mb in zip(
                 self.number_of_test_cases_for_phases,
                 self.average_size_test_case_mb_for_phases,
                 strict=True,
             )
         ]
 
+        output_size_mb_for_phases = [
+            n_jobs * output_mb
+            for n_jobs, output_mb in zip(
+                self.number_of_algorithm_jobs_for_phases,
+                self.average_size_job_output_mb_for_phases,
+                strict=True,
+            )
+        ]
+        return [
+            round(
+                (input_size_mb + output_size_mb)
+                * (settings.MEGABYTE / settings.GIGABYTE)
+            )
+            for input_size_mb, output_size_mb in zip(
+                input_size_mb_for_phases,
+                output_size_mb_for_phases,
+                strict=True,
+            )
+        ]
+
     @property
     def number_of_docker_images_per_team_for_tasks(self):
         # A docker image for a later phase should also be submitted to an earlier one.
@@ -1770,7 +1815,7 @@ def compute_costs_euros_for_tasks(self):
             for task_id in self.task_ids
         ]
 
-    @cached_property
+    @property
     def data_storage_costs_euros_for_tasks(self):
         return [
             sum(
@@ -1892,6 +1937,9 @@ def costs_for_tasks(self):
                 "average_size_test_case_mb": self.average_size_test_case_mb_for_tasks[
                     task_index
                 ],
+                "average_size_job_output_mb": self.average_size_job_output_mb_for_tasks[
+                    task_index
+                ],
                 "compute_costs_euros": self.compute_costs_euros_for_tasks[
                     task_index
                 ],
@@ -1928,6 +1976,9 @@ def get_costs_for_phases_in_task(self, task_id):
                 "data_storage_size_gb": self.data_storage_size_gb_for_phases[
                     phase_index
                 ],
+                "number_of_jobs": self.number_of_algorithm_jobs_for_phases[
+                    phase_index
+                ],
                 "compute_costs_euros": self.compute_costs_euros_for_phases[
                     phase_index
                 ],

app/grandchallenge/challenges/templates/challenges/partials/budget_table.html

@@ -133,7 +133,9 @@ <h5>Task {{ task.id }}</h5>
                         <tr>
                             <td>Data storage phase {{ forloop.counter }}</td>
                             <td>{{ task.average_size_test_case_mb }} MB average size per case,
-                                {{ phase.number_of_test_cases }} case{{ phase.number_of_test_cases|pluralize }},
+                                 {{ phase.number_of_test_cases }} case{{ phase.number_of_test_cases|pluralize }},
+                                {{ task.average_size_job_output_mb }} MB average output per job,
+                                 {{ phase.number_of_jobs }} job{{ phase.number_of_jobs|pluralize }},
                                 for {{ num_support_years }} year{{ num_support_years|pluralize }}</td>
                             <td class="text-right">{{ object.storage_costs_euros_per_gb|euro }}</td>
                             <td>{{ phase.data_storage_size_gb }}</td>

app/grandchallenge/challenges/views.py

@@ -355,6 +355,7 @@ def get_initial(self):
             ],
             "algorithm_maximum_settable_memory_gb_for_tasks": [32, 32],
             "average_size_test_case_mb_for_tasks": [10, 40],
+            "average_size_job_output_mb_for_tasks": [1, 1],
             "inference_time_average_minutes_for_tasks": [5, 10],
             "task_ids": [1, 2],
             "task_id_for_phases": [1, 1, 2, 2],

app/tests/challenges_tests/test_forms.py

@@ -105,6 +105,7 @@ def test_budget_update_form():
         "algorithm_maximum_settable_memory_gb_for_tasks": "[32, 32]",
         "algorithm_selectable_gpu_type_choices_for_tasks": '[["", "T4"],["", "A10G", "T4"]]',
         "average_size_test_case_mb_for_tasks": "[10, 100]",
+        "average_size_job_output_mb_for_tasks": "[1, 1]",
         "task_id_for_phases": "[1, 1, 2, 2]",
         "number_of_teams_for_phases": "[500, 500, 500, 500]",
         "number_of_submissions_per_team_for_phases": "[10, 1, 10, 1]",
@@ -121,6 +122,7 @@ def test_budget_update_form():
         "algorithm_maximum_settable_memory_gb_for_tasks": "[32, 32]",
         "algorithm_selectable_gpu_type_choices_for_tasks": '[["", "T4"],["", "A10G", "T4"]]',
         "average_size_test_case_mb_for_tasks": "[10, 100]",
+        "average_size_job_output_mb_for_tasks": "[1, 1]",
         "inference_time_average_minutes_for_tasks": "[5, 10]",
         "task_id_for_phases": "[1, 1, 2, 2]",
         "number_of_teams_for_phases": "[500, 500, 500, 500]",
@@ -150,6 +152,10 @@ def test_budget_update_form():
             {"average_size_test_case_mb_for_tasks": "[10]"},
             "not all tasks defined",
         ),
+        (
+            {"average_size_job_output_mb_for_tasks": "[1]"},
+            "not all tasks defined",
+        ),
         (
             {"inference_time_average_minutes_for_tasks": "[10]"},
             "not all tasks defined",
@@ -186,6 +192,7 @@ def test_budget_update_form_invalid(invalid_data, reason_invalid):
         "algorithm_maximum_settable_memory_gb_for_tasks": "[32, 32]",
         "algorithm_selectable_gpu_type_choices_for_tasks": '[["", "T4"],["", "A10G", "T4"]]',
         "average_size_test_case_mb_for_tasks": "[10, 100]",
+        "average_size_job_output_mb_for_tasks": "[1, 1]",
         "inference_time_average_minutes_for_tasks": "[5, 10]",
         "task_id_for_phases": "[1, 1, 2, 2]",
         "number_of_teams_for_phases": "[10, 10, 10, 10]",

app/tests/challenges_tests/test_migrations.py

@@ -7,10 +7,14 @@
 
 from tests.factories import ChallengeRequestFactory
 
-_migration = importlib.import_module(
+_migration_0070 = importlib.import_module(
     "grandchallenge.challenges.migrations"
     ".0070_assign_review_perm_to_reviewers_group"
 )
+_migration_0073 = importlib.import_module(
+    "grandchallenge.challenges.migrations"
+    ".0073_set_average_size_job_output_for_tasks"
+)
 
 
 @pytest.mark.django_db
@@ -30,7 +34,9 @@ def test_assign_review_perm_to_reviewers_group(settings):
         reviewers_group, challenge_request
     )
 
-    _migration.assign_review_perm_to_reviewers_group(apps, schema_editor=None)
+    _migration_0070.assign_review_perm_to_reviewers_group(
+        apps, schema_editor=None
+    )
 
     assert "review_challengerequest" in get_group_perms(
         reviewers_group, challenge_request
@@ -53,8 +59,34 @@ def test_assign_change_perm_to_creators():
         challenge_request.creator, challenge_request
     )
 
-    _migration.assign_change_perm_to_creators(apps, schema_editor=None)
+    _migration_0070.assign_change_perm_to_creators(apps, schema_editor=None)
 
     assert "change_challengerequest" in get_user_perms(
         challenge_request.creator, challenge_request
     )
+
+
+@pytest.mark.django_db
+@pytest.mark.parametrize(
+    "task_ids, expected",
+    [
+        ([], []),
+        ([1], [0]),
+        ([1, 2, 3], [0, 0, 0]),
+    ],
+)
+def test_set_average_size_job_output_mb_for_tasks(task_ids, expected):
+    challenge_request = ChallengeRequestFactory(
+        task_ids=task_ids,
+        average_size_job_output_mb_for_tasks=[],
+    )
+
+    # The field is empty before the migration.
+    assert challenge_request.average_size_job_output_mb_for_tasks == []
+
+    _migration_0073.set_average_size_job_output_mb_for_tasks(
+        apps, schema_editor=None
+    )
+
+    challenge_request.refresh_from_db()
+    assert challenge_request.average_size_job_output_mb_for_tasks == expected