NMR logging can provide the permeability parameter and abundant stratigraphical information such as total porosity,oil,gas and water saturation,oil viscosity,etc. And these physical parameters can be obtained by T2 spectrum inversion. NMR inversion is an important part in logging interpretation. The authors describe a multi-exponential inversion algorithm,solid iteration redress technique( SIRT),and apply the algorithm in real data and compare the results with those based on singular value decomposition( SVD). It shows that SIRT algorithm is easier to be understood and implemented,and the time spent in SIRT is much shorter than that of SVD algorithm. And the non-negative property of T2 spectrum is much easier to be implemented. It can match the results based on SVD very well. SIRT algorithm can be used in T2 spectrum inversion for NMR analysis. NMR logging can provide the permeability parameter and abundant stratigraphical information such as total porosity, oil, gas and water saturation, oil viscosity, etc. And these physical parameters can be obtained by T2 spectrum inversion. NMR transition is an important part in logging interpretation. The authors describe a multi-exponential inversion algorithm, solid iteration redress technique (SIRT), and apply the algorithm in real data and compare the results with those based on singular value decomposition (SVD). It shows that SIRT algorithm is easier to be understood And the implemented, and the time spent in SIRT is much shorter than that of SVD algorithm. And the non-negative property of T2 spectrum is much easier to be implemented. in T2 spectrum inversion for NMR analysis.