Improved Chain Calculation for Sub-chain Dependencies in Layered Queueing Networks
Lianhua Li, Greg Franks
Abstract
Lianhua Li, Greg Franks
Abstract
Often, many software systems fail to meet requirements because of a lack of performance. A proven method for preventing or for diagnosing performance problems is through modeling. Layered Queueing Networks (LQN) are one popular technique for solving performance models. However, if a LQN is solved through decomposition and Mean Value Analysis (MVA), erroneous results can arise because of traffic dependencies in the decomposed models. This paper addresses one traffic dependency, called sub-chains, where customers from one chain "bleed into" another chain causing "extraneous" queueing delays. The new approach described here changes approximate MVA by adjusting the population in a routing chain depending on the originating sub-chain. This new approach substantially reduces, or even eliminates, the extraneous queueing delay caused by the sub-chain dependent traffic. The approach was applied to a substantial model of an on-line bookstore, and reduced the overall error in queueing time by a factor of 20 times, when compared to simulation. The more accurate queueing estimates yield better results for the other outputs of the LQN solver.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Often, many software systems fail to meet requirements because of a lack of performance. A proven method for preventing or for diagnosing performance problems is through modeling. Layered Queueing Networks (LQN) are one popular technique for solving performance models. However, if a LQN is solved through decomposition and Mean Value Analysis (MVA), erroneous results can arise because of traffic dependencies in the decomposed models. This paper addresses one traffic dependency, called sub-chains, where customers from one chain "bleed into" another chain causing "extraneous" queueing delays. The new approach described here changes approximate MVA by adjusting the population in a routing chain depending on the originating sub-chain. This new approach substantially reduces, or even eliminates, the extraneous queueing delay caused by the sub-chain dependent traffic. The approach was applied to a substantial model of an on-line bookstore, and reduced the overall error in queueing time by a factor of 20 times, when compared to simulation. The more accurate queueing estimates yield better results for the other outputs of the LQN solver.
Key concepts: Queueing theory, Mean value analysis, Layered queueing network, Computer science, Chain (unit), Dependency (UML), Solver, Mathematical optimization