1. Problem: name of the benchmark problem.
2. f(x): objective function value at solution.
3. eps(C): convergency tolerance.
4. eps(F): feasibility tolerance.
5. # Iter.: number of iterations used by the solver.
6. # x: number of variables in the benchmark problem.
7. # b: number of simple bound constraints in the benchmark problem.
8. # LC: number of linear constraints in the benchmark problem.
9. # NLC: number of non-linear constraints in the benchmark problem.
10. # f(x)-O: number of function calls used by the solver to evaluate objective function.
11. # g(x)-O: number of function calls used by the solver to evaluate the gradient of the objective function.
12. # H(x)-O: number of function calls used by the solver to evaluate the Hessian of the objective function.
13. # f(x)-C: number of function calls used by the solver to evaluate the nonlinear constraint function.
14. # g(x)-C: number of function calls used by the solver to evaluate the gradient of the nonlinear constraint function.
15. Time(ms.): time in milliseconds used by the solver to find the solution.

Optimization benchmark problems are taken from CUTE and COPS problem sets. Click the content table button on the left to browse the benchmark results by problem class.