Parallel solving symmetric eigenproblems

In this paper parallel solving symmetric eigenproblems, which include standard and generalized eigenvalue problems, is discussed. For standard eigenvalue problem and tridiagonal eigenvalue problem is not the key point. For symmetric-definite generalized eigenvalue problem, which arises in solving many actual application problems, we give a new parallel computational method for reducing the generalized eigenproblem to standard one. The parallel algorithm is fully considered in reducing the communications. The numerical measurement is given in SGI/Cray T3E and Hitachi SR2201. Some computational results are compared with ScaLAPACK.