Glycan biosynthesis is affected by disease states more significantly than protein, which shows greater potential to develop as biomarker. Therefore, qualitative and quantitative analysis is indispensable for glycomics and mass spectrometry (MS) has been the most powerful analytical tool in this hot field. However, the native glycan has lower ionization efficiency and produces more complex fragments than peptide. To solve those problems, permethylation has been developed as the most efficient derivative approach with improved ionization efficiency and simple fragments in MS analysis. In this study, we further optimize the solid-phase permethylation by different parameters and develop a novel strategy for N-glycomics to match the experimental data with theoretical database by R-scripts, which increase significantly the number of identifiable N-glycans with isotope-based data quality control. Furthermore, a novel bundled sequencing algorithm is designed to identify the N-glycoforms at MS2 levels. By this strategy, 57 N-glycan species (133 N-glycoforms) from ovalbumin, 90 N-glycan species (162 N-glycoforms) from etanercept, 245 N-glycan species (395 N-glycoforms) from human acute promyelocytic leukemia cells and 343 N-glycan species (833 N-glycoforms) from corpus callosum of adult mouse are identified. The identified N-glycans are verified by pGlyco software. This strategy is also applicative on O-glycomics. Besides, stable isotopic labeling and label-free quantification are performed for N-glycomics, in which we prove that glycan is more sensitive than its related protein after bioinformatic analysis. Finally, this study provides a novel pathway for N-glycomics to realize deep identification and biomarker discovery.