1: \begin{abstract}

2: In the numerical treatment of large-scale Sylvester and Lyapunov equations,

3: projection methods require solving a reduced problem to check convergence.

4: As the approximation space expands, this solution takes an increasing portion of

5: the overall computational effort.

6: When data are symmetric, we show that the Frobenius norm of the residual matrix 

7: can be computed

8: at significantly lower cost than with available methods, without 

9: explicitly solving the reduced problem.

10: For certain classes of problems,

11: the new residual norm expression combined with a memory-reducing device

12: make classical Krylov strategies competitive with respect to more recent

13: projection methods.

14: Numerical experiments illustrate the effectiveness of the new implementation for

15: standard and extended Krylov subspace methods.

16: \end{abstract}