In livestock, inbreeding coefficient based on pedigree information is usually used to evaluate the level of inbreeding. Recently, with cost reduction of high-density SNP genotyping, it’s possible to analyze real genomic inbreeding degree using genomic information. In this study, utilizing high-density SNP chip data, we analyzed the frequency and distribution of runs of homozygosity (ROH) in 2107 Chinese Holstein cattle in Beijing area, and calculated 2 genomic inbreeding coefficients, i.e., 1) the proportion of ROH length in the total length of autosomal genome (Froh), and 2) the percentage of homozygous SNPs (Fhom). Then we analyzed the correlation between 2 genomic inbreeding coefficients and the correlation between genomic and pedigree inbreeding coefficients. We totally detected 44 676 ROHs that mainly ranged from 1 to 10 Mb. Various lengths of ROHs existed in the genome. There were more short ROHs than long ROHs. ROHs aren’t evenly distributed in chromosomes. The area with most ROHs is in the middle part of chromosome 10. Strong correlation (r > 0.90) existed between 2 kinds of genomic inbreeding coefficients, but the correlation between pedigree and genomic inbreeding coefficients were much lower (r < 0.50). Our finding suggests that pedigree completeness influences the correlation between genomic and pedigree inbreeding. Genomic inbreeding measures may reflect individuals’ real inbreeding, which could be a useful tool to evaluate population inbreeding.