Peptides with antimicrobial potential can now be easily and efficiently predicted, designed and validated using in silico tools. Here, twelve peptides with molecular weights between 1.6 and 1.8 kDa were predicted from the sequence of the small subunit of Penaeus vannamei hemocyanin (PvHS) using three online software (AntiBP2, CAMP and APD3). Among the twelve predicted peptides, two peptides, PvHS8 and PvHS9, with the highest prediction score, were selected and their antimicrobial activities ascertained. Both peptide PvHS8 and PvHS9 had strong antimicrobial activity on Gram-negative (i.e., Vibrio parahaemolyticus, Vibrio fluvialis, Vibrio alginolyticus, Escherichia coli, and Aeromonas hydrophila) and Gram-positive (i.e., Staphylococcus aureus, Streptococcus iniae) bacteria, with PvHS9 having the strongest activity. Structural analysis using circular dichroism (CD) and nuclear magnetic resonance (NMR) revealed that peptide PvHS9 (FWVSLKGGKTSIERK) has a β-sheet structure. Mutation of PvHS9 to distort its β-sheet structure and net positive charge attenuated its antimicrobial activity. In addition, scanning electron microscopy (SEM) analysis revealed that peptide PvHS9 destroys bacteria cells by causing membrane rupture and cell death.