cim-bench

CIM/CGMES parser & serializer benchmarks · GitHub repository

Environment: AMD Ryzen AI 9 HX 370 w/ Radeon 890M · 24 cores · Linux 7.1.3-100.fc43.x86_64 · Latest benchmark: 2026-07-16 · Page generated: 2026-07-16 15:40 UTC

Filter by tag:c#c++cligojavaparserpowerflow-toolpythonqueryrustserializersparqltriplestoretyped-modelvalidator

Realgrid (86.5 MB)

ToolTagsVersionLoad timeExport timeMemory (MB)LinesGeneratorsLoadsSubstations
tripletsparser serializer query python0.1.0 (pandas 3.0.3, lxml 6.1.1, polars 1.42.1, pyarrow 24.0.0)159.1 ms946.1 ms370.57,5611,3476,6874,875
OpenCGMESparser serializer query sparql triplestore java1.1.0-SNAPSHOT (jpype1 1.7.1, java 21.0.11, jena 6.1.0)1.10 s4.00 s3055.97,5611,3476,6874,875
cimoxideparser validator typed-model python rust0.0.41.26 s–537.47,5611,3476,6874,875
Apache Jenaparser serializer query sparql triplestore java6.0.0 (jpype1 1.7.1, java 21.0.11)1.58 s4.29 s2586.67,5611,3476,6874,875
PowSyBL CGMESparser serializer query sparql triplestore java7.1.1 (jpype1 1.7.1, java 21.0.11)1.77 s2.17 s2721.57,5611,3476,6874,875
maplibparser serializer query sparql triplestore python rust0.20.19 (polars 1.42.1, pyarrow 23.0.1, fastapi 0.139.0)2.01 s15.26 s454.87,5611,3476,6874,875
PyPowSyBlparser serializer query powerflow-tool python java1.15.0 (prettytable 3.18.0, pandas 3.0.3, networkx 3.6.1)4.24 s1.55 s2089.07,5611,3476,6874,791
GraphDBparser serializer query sparql triplestore java10.8.14 (jpype1 1.7.1, java 21.0.11)5.60 s1.95 s2457.47,5611,3476,6874,875
VeraGridparser serializer query powerflow-tool python6.3.0 (numpy 2.4.6, PySide6 6.11.1, requests 2.34.2, urllib3 2.7.0, websockets 16.0, opencv-python 5.0.0.93, packaging 26.2, VeraGridEngine 6.3.0)10.80 s16.73 s304.77,5611,3476,6874,875
GMSS CIMparser query sparql c#10.1.1.0 (dotnet 10.0.9, GridLab.Abp.Rdf 10.1.1.0, Volo.Abp.Core 10.1.1.0)11.83 s–2646.20000
libcimppparser query typed-model c++2.2.019.53 s–133.87,5611,3476,6874,875
RDFlibparser serializer query sparql triplestore python7.6.023.27 s18.61 s1078.47,5612,69413,3744,875

Query performance

Toolgeneratorslinesloadssubstations
Apache Jena519.2 μs203.5 ms4.9 ms864.4 μs
GraphDB409.6 μs252.1 ms1.2 ms718.6 μs
OpenCGMES521.7 μs144.9 ms2.5 ms744.9 μs
PowSyBL CGMES288.3 μs347.4 ms983.1 μs330.9 μs
PyPowSyBl2.2 ms33.2 ms16.1 ms3.5 ms
RDFlib443.7 μs446.88 s2.1 ms750.0 μs
VeraGrid0.1 μs0.1 μs0.1 μs0.1 μs
cimoxide9.3 ms12.9 ms26.3 ms7.4 ms
libcimpp5.4 ms5.4 ms12.8 ms5.4 ms
maplib459.0 μs44.6 ms1.1 ms485.3 μs
triplets7.4 ms12.6 ms11.2 ms6.4 ms

Load time - Realgrid (86.5 MB)

triplets
159.1 ms
OpenCGMES
1.10 s
cimoxide
1.26 s
Apache Jena
1.58 s
PowSyBL CGMES
1.77 s
maplib
2.01 s
PyPowSyBl
4.24 s
GraphDB
5.60 s
VeraGrid
10.80 s
GMSS CIM
11.83 s
libcimpp
19.53 s
RDFlib
23.27 s

Export time - Realgrid (86.5 MB)

triplets
946.1 ms
PyPowSyBl
1.55 s
GraphDB
1.95 s
PowSyBL CGMES
2.17 s
OpenCGMES
4.00 s
Apache Jena
4.29 s
maplib
15.26 s
VeraGrid
16.73 s
RDFlib
18.61 s

Memory - Realgrid (86.5 MB)

libcimpp
134 MB
VeraGrid
305 MB
triplets
370 MB
maplib
455 MB
cimoxide
537 MB
RDFlib
1,078 MB
PyPowSyBl
2,089 MB
GraphDB
2,457 MB
Apache Jena
2,587 MB
GMSS CIM
2,646 MB
PowSyBL CGMES
2,722 MB
OpenCGMES
3,056 MB

Average query time - Realgrid (86.5 MB) (log scale)

VeraGrid
0.1 μs
libcimpp
7.2 ms
triplets
9.4 ms
maplib
11.7 ms
PyPowSyBl
13.8 ms
cimoxide
14.0 ms
OpenCGMES
37.2 ms
Apache Jena
52.4 ms
GraphDB
63.6 ms
PowSyBL CGMES
87.2 ms
RDFlib
111.72 s

Svedala (7.3 MB)

ToolTagsVersionLoad timeExport timeMemory (MB)LinesGeneratorsLoadsSubstations
tripletsparser serializer query python0.1.0 (pandas 3.0.3, lxml 6.1.1, polars 1.42.1, pyarrow 24.0.0)10.1 ms147.8 ms82.997397357
OpenCGMESparser serializer query sparql triplestore java1.1.0-SNAPSHOT (jpype1 1.7.1, java 21.0.11, jena 6.1.0)88.4 ms357.3 ms293.597397356
cimoxideparser validator typed-model python rust0.0.4103.8 ms–60.697397357
Apache Jenaparser serializer query sparql triplestore java6.0.0 (jpype1 1.7.1, java 21.0.11)120.1 ms403.1 ms402.097397356
maplibparser serializer query sparql triplestore python rust0.20.19 (polars 1.42.1, pyarrow 23.0.1, fastapi 0.139.0)218.0 ms1.28 s140.697397357
PowSyBL CGMESparser serializer query sparql triplestore java7.1.1 (jpype1 1.7.1, java 21.0.11)252.8 ms168.5 ms467.697397357
PyPowSyBlparser serializer query powerflow-tool python java1.15.0 (prettytable 3.18.0, pandas 3.0.3, networkx 3.6.1)428.4 ms138.0 ms1566.697397357
VeraGridparser serializer query powerflow-tool python6.3.0 (numpy 2.4.6, PySide6 6.11.1, requests 2.34.2, urllib3 2.7.0, websockets 16.0, opencv-python 5.0.0.93, packaging 26.2, VeraGridEngine 6.3.0)701.7 ms1.54 s71.897397356
GraphDBparser serializer query sparql triplestore java10.8.14 (jpype1 1.7.1, java 21.0.11)712.3 ms191.8 ms1344.397397357
GMSS CIMparser query sparql c#10.1.1.0 (dotnet 10.0.9, GridLab.Abp.Rdf 10.1.1.0, Volo.Abp.Core 10.1.1.0)1.01 s–312.9977814657
RDFlibparser serializer query sparql triplestore python7.6.02.00 s1.95 s207.5977814657

Query performance

Toolgeneratorslinesloadssubstations
Apache Jena165.9 μs4.1 ms295.9 μs64.3 μs
GMSS CIM120.6 μs2.92 s284.7 μs92.1 μs
GraphDB137.9 μs5.9 ms288.6 μs82.4 μs
OpenCGMES160.7 μs4.7 ms251.6 μs81.7 μs
PowSyBL CGMES110.1 μs8.3 ms161.8 μs74.5 μs
PyPowSyBl265.9 μs292.9 μs194.2 μs120.9 μs
RDFlib68.8 μs805.7 ms199.8 μs66.3 μs
VeraGrid0.1 μs0.0 μs0.1 μs0.1 μs
cimoxide292.4 μs259.0 μs878.1 μs286.6 μs
maplib331.3 μs14.6 ms869.9 μs355.0 μs
triplets1.5 ms1.7 ms1.7 ms1.0 ms

Load time - Svedala (7.3 MB)

triplets
10.1 ms
OpenCGMES
88.4 ms
cimoxide
103.8 ms
Apache Jena
120.1 ms
maplib
218.0 ms
PowSyBL CGMES
252.8 ms
PyPowSyBl
428.4 ms
VeraGrid
701.7 ms
GraphDB
712.3 ms
GMSS CIM
1.01 s
RDFlib
2.00 s

Export time - Svedala (7.3 MB)

PyPowSyBl
138.0 ms
triplets
147.8 ms
PowSyBL CGMES
168.5 ms
GraphDB
191.8 ms
OpenCGMES
357.3 ms
Apache Jena
403.1 ms
maplib
1.28 s
VeraGrid
1.54 s
RDFlib
1.95 s

Memory - Svedala (7.3 MB)

cimoxide
61 MB
VeraGrid
72 MB
triplets
83 MB
maplib
141 MB
RDFlib
208 MB
OpenCGMES
294 MB
GMSS CIM
313 MB
Apache Jena
402 MB
PowSyBL CGMES
468 MB
GraphDB
1,344 MB
PyPowSyBl
1,567 MB

Average query time - Svedala (7.3 MB) (log scale)

VeraGrid
0.1 μs
PyPowSyBl
218.5 μs
cimoxide
429.0 μs
Apache Jena
1.2 ms
OpenCGMES
1.3 ms
triplets
1.5 ms
GraphDB
1.6 ms
PowSyBL CGMES
2.2 ms
maplib
4.0 ms
RDFlib
201.5 ms
GMSS CIM
731.2 ms

CLI tools

Benchmarked as subprocesses (full process lifecycle per run) - not comparable with the in-process library numbers above.

ToolTagsVersionDatasetOperationTimePeak RSS (MB)
cimgocli validator serializer go?realgridconvert3.69 s621.7
cimgocli validator serializer go?realgridvalidate4.34 s717.1
pocket-rdfcli serializer query sparql python?realgridquery26.06 s1241.1
pocket-rdfcli serializer query sparql python?realgridserialize28.47 s1250.0
cimgocli validator serializer go?svedalaconvert379.6 ms76.8
cimgocli validator serializer go?svedalavalidate279.9 ms43.9
pocket-rdfcli serializer query sparql python?svedalaquery2.21 s142.4
pocket-rdfcli serializer query sparql python?svedalaserialize2.31 s142.0

CLI time - Realgrid (86.5 MB)

cimgo (convert)
3.69 s
cimgo (validate)
4.34 s
pocket-rdf (query)
26.06 s
pocket-rdf (serialize)
28.47 s

CLI peak RSS - Realgrid (86.5 MB)

cimgo (convert)
622 MB
cimgo (validate)
717 MB
pocket-rdf (query)
1,241 MB
pocket-rdf (serialize)
1,250 MB

CLI time - Svedala (7.3 MB)

cimgo (validate)
279.9 ms
cimgo (convert)
379.6 ms
pocket-rdf (query)
2.21 s
pocket-rdf (serialize)
2.31 s

CLI peak RSS - Svedala (7.3 MB)

cimgo (validate)
44 MB
cimgo (convert)
77 MB
pocket-rdf (serialize)
142 MB
pocket-rdf (query)
142 MB

Generated by tools/generate_site.py from pytest-benchmark JSON results. Sort any table by clicking a column header; tag filters apply to tables and charts alike.