finished updated 2026-11-07 22:41:02 UTC

Summary

CPU current
1582/1582
1423 solved
CUDA current
1582/1582
1412 solved
setdeviceproblems Ipoptmastercurrent
solvedSGM (s)SSGMsolvedSGM (s)SSGMsolvedSGM (s)SSGM
ACOPFCPU132/13212524.27112619.660.86612619.820.87
COPSCPU65/656423.2916417.610.8296417.330.821
CUTEstCPU1338/1338122311.891118516.581.21119016.001.19
MittelmannCPU47/4743126.14143109.430.87743111.430.892
ACOPFCUDA132/13212319.5711245.650.5291245.740.532
COPSCUDA65/656121.161653.170.423653.120.421
CUTEstCUDA1338/133812057.56111807.781.0111887.440.993
MittelmannCUDA47/473989.6713631.360.4153532.900.43

Solvers

solverdevicerevisionbenchmark
IpoptCPU3.14.19–
masterCPUd9940730 to 315765320.1.0
masterCUDAac03fdde to 315765320.1.0
currentCPU95a2c9ab to 8aaabb200.1.0
currentCUDA95a2c9ab to 8aaabb200.1.0

CPU: shin-compute-005: Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz (40 cores, 80 threads), 123.3 GB of memory

CUDA: shin-compute-005: Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz (40 cores, 80 threads), 123.3 GB of memory; Quadro RTX 6000, 23.5 GB, driver 13.5.0, CUDA runtime 13.4.0

Benchmark 0.1.0: the version fixes the instances, the options, Ipopt and HSL.

Base: `9bbe0480`, on Quadro RTX 6000 (acopf, cops); `31576532`, from #685, on Intel R Xeon R Gold 6148 CPU 2.40GHz (acopf, cops, cutest, mittelmann); `31576532`, from #685, on Quadro RTX 6000 (cutest, mittelmann)

SGM: shifted geometric mean of the solve time, with a shift of 10 s and a failure counting as the time limit of a solve (CPU 7200 s, GPU 600 s, for every solver on those pages), over the problems that all solvers shown have run. SSGM: the SGM scaled so that Ipopt, or the best solver where Ipopt has not run, is 1.