A key issue when designing and implementing large-scale content2based publish/ subscribe systems is how to efficiently match high volumes of events against large numbers of subscription. A fast and efficient algorithm is presented ,according to the characteristic of logical expression ,this algorithm makes pretreatment with all the subscription by the priority ,in this way ,the comparison of the same attributes is one time or less than one time ,there by reducing the cost of matching. Experiment proves that it is more efficient than other commonly used algorithms ,and has good expansibility ,suits large-scale distributed content2based publish/ subscribe system.