An approach to the study of the conditioning of difference schemes and their stability to data perturbations is developed for the Dirichlet problem for a singularly perturbed convection-diffusion ordinary differential equation with the perturbation parameter ε, ε ∈ (0, 1]. We consider a standard difference scheme, which is a monotone scheme on a uniform grid, and a special scheme of the grid solution decomposition method. Constructing the special scheme, we use a decomposition of the grid solution into a regular and a singular components that are the solutions to grid subproblems considered on uniform grids.The following facts are proved for the standard scheme: (a) the scheme converges only for N −1 = o(ε) at the rate O N −1 (ε + N −1 ) −1 , where N + 1 is the number of grid nodes, (b) the scheme is not ε-uniformly well conditioned and stable to perturbations, (c) the condition number of the scheme has the order O ε −1 δ −2 , where δ is the accuracy of the grid solution, δ = δ st ≈ N −1 (ε + N −1 ) −1 . In the case of convergence of the standard scheme proved theoretically, the actual accuracy of the computed solution decreases with the decrease in the parameter ε under the presence of perturbations and may completely vanish for a sufficiently small ε (namely, under the condition t = O ln ε −1 + ln δ −2 st , where t is the number of digits in the machine word). At the same time, the special scheme of the solution decomposition method converges ε-uniformly in the uniform norm at the rate O N −1 ln N ; in the variables ε and δ , the special scheme is ε-uniformly well conditioned and stable to grid problem data perturbations; the condition number of the scheme of the decomposition method is of the order O δ −2 ln δ −1 , δ = δ dec ≈ N −1 ln N.Methods for the construction of ε-uniformly convergent difference schemes for singularly perturbed partial differential equations have been successfully developed on the base of the method of refining grids and the method of adjustment (see, e.g., [2,6,7,11,12,16,24] and the references therein).An essential drawback of the approach based on standard methods on uniform grids is that the constructed schemes converge formally, i.e., under the condition that all calculations are performed exactly. In reality, if the parameter ε is sufficiently small, the errors appearing in actual calculations Brought to you by | Florida International University Libraries Authenticated Download Date | 5/30/15 10:12 PM Data perturbation stability 383 cause errors in the numerical solution exceeding the accuracy of the solution obtained without computation errors by several orders. The cause is that the standard numerical methods are not, generally speaking, ε-uniformly stable to grid problem data perturbations appearing in calculations even in the case of their convergence (see, e.g., the assertions of Theorems 6.2, 6.3 and Remark 6.1 for a convection-diffusion problem).Thus, in the case of singularly perturbed problems, there is an interest to develop special ε-uniformly convergent numerical method...