Accounting Portal
Overview
The EGI Accounting Portal provides the accounting data for Compute and Cloud services gathered from the data centres of the EGI Federation. Users can employ custom filtering by selecting resource centres, VOs and defining a time period etc. Accounting data gathered by the Open Science Grid (OSG) is also included.
Accounting data
Data for a given month is mapped to the first of the month at midnight, i.e. all usage for July 2026 is stored at ‘2026-07-01T00:00:00Z’.
Usage for the current month will update daily.
Programmatic access
Access to the entirety of the accounting data is provided at the following endpoints, in the following formats.
JSON format explained
The JSON format contains one or more result sets, each consisting of a named series with column definitions and corresponding data rows.
High-level structure
root
└── results[] # Array of result sets
└── series[] # Array of series returned for this result
├── name # Series name ("summary")
├── columns[] # Column definitions
├── values[] # Data rows
└── partial # Optional flag indicating this series is partial
# and more data follows in the next series.
To obtain the full dataset, in a representation of the data where each record is stored as a self-contained object with named fields, concatenate all rows from all series.
all_records = []
for result in data["results"]:
for series in result.get("series", []):
cols = series["columns"]
for row in series["values"]:
all_records.append(dict(zip(cols, row)))
Datapoints explained
Each entry in values represents a summary accounting record containing:
| Column | Description |
|---|---|
| time | Timestamp associated with the record |
| benchmark | Benchmark identifier or value |
| ce | Computing Element |
| country | Country in which the resource is located |
| federation | Federation to which the resource belongs |
| infrastructure | Grid, Grid-Local (local jobs) or Cloud |
| site | Site name |
| tier | Site tier classification within WLCG |
| vo | Virtual Organisation |
| ngi | The national grouping of shared computing resources |
| number_of_jobs | Number of jobs represented by the record |
| raw_cpu_eff | CPU efficiency metric |
| raw_cpu_time | Total CPU time consumed |
| raw_cpu_work | Total CPU work performed |
| raw_wc_time | Total wall-clock time |
| raw_wc_work | Total wall-clock work performed |
Example
{
"results": [
{
"series": [
{
"name": "summary",
"columns": [
"time",
"benchmark",
"ce",
"country",
"federation",
"infrastructure",
"site",
"tier",
"vo",
"number_of_jobs",
"raw_cpu_eff",
"raw_cpu_time",
"raw_cpu_work",
"raw_wc_time",
"raw_wc_work"
],
"values": [
[Datapoint 1],
[Datapoint 2]
# etc
],
"partial": true
}
]
},
{
"series": [
{
"name": "summary",
"columns": [
"time",
"benchmark",
"ce",
"country",
"federation",
"infrastructure",
"site",
"tier",
"vo",
"number_of_jobs",
"raw_cpu_eff",
"raw_cpu_time",
"raw_cpu_work",
"raw_wc_time",
"raw_wc_work"
],
"values": [
[Datapoint 99],
[Datapoint 100]
]
}
]
}
]
}