Title :
An approach to reduce the Erlang B probability of the M/M/2/2 system
Author_Institution :
Dept. of Syst. Eng., Arkansas Univ., Little Rock, AR, USA
fDate :
28 Nov.-2 Dec. 2005
Abstract :
It is well known that the Erlang B probability of an M/M/1/1 system can be reduced by evenly splitting the capacity of its single server into two servers, which leads to an M/M/2/2 system with homogeneous servers. In this paper, we analytically prove that the Erlang B probability can be further reduced by unevenly splitting the capacity of the single server into two servers, which leads to an M/M/2/2 system with heterogeneous servers. As a matter of fact, there exists an optimal load ratio of the two heterogeneous servers at which the Erlang B probability reaches its minimum.
Keywords :
network servers; probability; Erlang B probability; M/M/1/1 system; M/M/2/2 system; server; Capacity planning; Constraint optimization; Constraint theory; Control systems; Operating systems; Probability distribution; Queueing analysis; Steady-state; Systems engineering and theory; Telephony;
Conference_Titel :
Global Telecommunications Conference, 2005. GLOBECOM '05. IEEE
Print_ISBN :
0-7803-9414-3
DOI :
10.1109/GLOCOM.2005.1577358
