In this paper, we propose a new scheduling algorithm with economic theory, called black Scholes market (BSM) algorithm for a class of dynamic jobs (DJ). BSM is based on the classic option pricing theory in investment - black Scholes pricing model. The algorithm could meet the needs of dynamic flow jobs and select server to provide specific service through simulating an irrational market. Compared with dynamic weighted round robin (DWRR) and dynamic statistical random (DSR) scheduling algorithms, BSM algorithm achieves a better performance in long time scheduling and the best average delay rate in different maximum job arrival rates. And from view of the stability, BSM is also much better than the other two algorithms