• Solution of large linear systems with 
  symmetric positive definite matrices; 
  general symmetric matrices; 
  general unsymmetric matrices;
• Version for complex arithmetic;
• Parallel factorization and solve phases 
  (uniprocessor version also available);
• Iterative refinement and backward error analysis;
• Various matrix input formats 
  assembled format; 
  distributed assembled format; 
  elemental format;
• Partial factorization and Schur complement matrix (centralized or 2D block-cyclic);
• Interfaces to MUMPS: Fortran, C, Matlab and Scilab;
• Several orderings interfaced: AMD, AMF, PORD, METIS, PARMETIS, SCOTCH, PT-SCOTCH.

Recent features

• Reduced/condensed right-hand side;
• Detection of null pivots, null space basis estimate;
• Symmetric indefinite matrices: preprocesssing and 2-by-2 pivots
• Hybrid scheduling;
• Computation of the determinant (with an option to discard factors);
• Sparse multiple right-hand side, distributed solution;