MODELS: Modellierung der Performance von Software-definierten Netzwerken in Rechenzentren (DFG) (03/2017-08/2019)

Description: Performance modeling and prediction techniques provide a powerful tool to analyze the performance of IT systems and applications running in a modern data center. Nowadays, data centers are becoming increasingly dynamic due to the common adoption of virtualization technologies. Virtual machines, data and services can be migrated on demand between physical hosts to optimize resource utilization while enforcing service-level agreements.  This makes an accurate and timely performance analysis a challenging problem.

For the analysis, analytical performance models or simulations can be used. The coarse-grained performance models abstract many details of the real network and therefore have limited predictive power. Thus, this project tries to work on a modeling approach that bridges the gap between coarse-grained analytical models and detailed simulation models. The proposed models are intended to keep the generic character of coarse-grained analytical models while at the same time enabling more accurate performance prediction by taking into account the major performance influencing factors of modern virtualized network infrastructures.

People involved: Samuel Kounev, Stefan Herrnleben

Project end: August 2019