Plant breeding is a dynamic process that incorporates new germplasm to introduce genetic variation. Knowledge gained from genetic diversity studies and identification of potentially useful germplasm is critical for efficiently utilizing these new materials for breeding program. Linkage disequilibrium (LD), diversity, and association mapping analyses in barley (Hordeum vulgare L.) were done using a mapping panel that included Ethiopian landraces, and cultivars and breeding lines from the Ethiopian, ICARDA, and NDSU breeding programs. LD decayed within 10 to 20 cM in the mapping panel and large proportions of unlinked loci were found to have large LD estimates, indicating that factors other than linkage contributed to LD. Diversity analyses using phenotypic data and molecular markers indicated that the mapping panel was highly structured according to spike row-type, geographic origin, and breeding history; thus, accounting for population structure and familial relatedness was crucial for association analyses. Comparison of the four models (Naïve, P, K, and P+K) indicated that the P+K model is the best model for the current mapping panel. The genome-wide association study (GWAS) identified 94 QTL for 14 agronomic and disease resistance traits; and 145 QTL for 11 malt and grain quality traits. Association mapping of agronomic and disease resistance traits identified six photoperiod related loci (Ppd_H1, HvFT4, HvGI, HvFT2, HvCO2, and HvCO1) and one vernalization-related locus (VRN-H1) for days to heading, one semi-dwarf locus (sdw3) for plant height, and four resistance loci (Rrs1, Rrs15, rpt.k and rpt.r). The largest number of QTL for malt and grain quality traits was detected in chromosome 5H, followed by chromosome 7H. QTL for malt and grain quality traits were mapped near the Hor1, Hor2, Upg2, Dor4, Ltp1, Amy1, and Amy2 loci. Several unique QTL were identified in the ICARDA and NDSU accessions, with the NDSU materials having the favorable marker genotypes. These regions could be useful to the Ethiopian breeding program for improving malt quality. The current study indicated that association mapping provided useful tool to identify QTL for several traits simultaneously. Because the QTL had small effect and distributed across the genome, genome selection may be warranted for improving these traits.