Performance bounds for the effectiveness of pooling in multi-processing systems

The need for quantifying the effect of resource pooling on performance of multi-processing systems arises frequently in the design of a variety of manufacturing, communication, and service systems. In this paper, we examine the effect of resource pooling and assess its impact on system performance. In particular, we provide performance bounds on the effectiveness of several pooling scenarios and discuss capacity and utilization tradeoffs between independent and pooled systems. We also propose a methodology for making optimal pooling decisions and describe the characteristics of this optimal solution. Limitations to the effectiveness of pooling are identified and conditions under which pooling may degrade performance are discussed.