Systematic data analysis-based validation of simulation models with heterogeneous data sources.

FOGUELMAN, DANIEL; BONAVENTURA, MATIAS; CASTRO, RODRIGO

Córdoba

Simposio; Simposio Argentino de GRANdes DAtos (AGRANDA)-JAIIO 46; 2017

Complex networked computer systems are subjected to up-grades on a continuous basis. Modeling and simulation (M&S) of suchsystems helps with guiding their engineering processes when testing designoptions on the real system is not an option. Too often many system?soperational conditions need to be assumed in order to focus on the questionsat hand, a typical case being the exogenous workload. Meanwhile, soaringamounts of monitoring information is logged to analyze the system?s per-formance in search for improvement opportunities. Concurrently, researchquestions mutate as operational conditions vary throughout its lifetime.This context poses many challenges to assess the validity of simulationmodels. As the empirical knowledge base of the system grows, the questionarises whether a simulation model that was once deemed valid could beinvalidated in the context of unprecedented operation conditions.This work presents a conceptual framework and a practical prototypethat helps with answering this question in a systematic, automatedway. MASADA parses recorded operation intervals and automaticallyparameterizes, launches, and validates simulation experiments. MASADAhas been tested in the data acquisition network of the ATLAS particlephysics experiment at CERN. The result is an efficient framework forvalidating our models on a continuous basis as new particle collisionsimpose unpredictable network workloads.