Model and method of an n-tier quality-of-service (qos)

Abstract

A method for increasing the QoS in a data center. The method can be employed in first and second components of the data center. In one embodiment, the method includes a first component of the data center receiving first and second requests. The first component generates first and second priority values corresponding to the first and second requests, respectively. The first component processes the first and second requests to generate third and fourth requests, respectively, wherein the first request is processed before the second request if the first priority value is numerically greater than the second priority value, or the first request is processed after the second request if the first priority value is numerically less than the second priority value. A second component of the data center receives the third and fourth requests. The second component generates third and fourth priority values corresponding to the third and fourth requests, respectively. The second component processes the third and fourth requests wherein the third request is processed before the fourth request if the third priority value is numerically greater than the fourth priority value, or the third request is processed after the fourth request if the third priority value is numerically less than the fourth priority value. In one preferred embodiment, if the first priority value is numerically greater than the second priority value, then the third priority value generated by the second component should not be numerically less than the fourth priority value generated by the second component.