Hybrid Performance Modeling and Prediction of Large-Scale Computing Systems

Abstract—Performance is a key feature of large-scale computing systems. However, the achieved performance when a certain program is executed is significantly lower than the maximal theoretical performance of the large-scale computing system. The model-based performance evaluation may be used to support the performance-oriented program development for large-scale computing systems. In this paper we present a hybrid approach for performance modeling and prediction of parallel and distributed computing systems, which combines mathematical modeling and discrete-event simulation. We use mathematical modeling to develop parameterized performance models for components of the system. Thereafter, we use discreteevent simulation to describe the structure of system and the interaction among its components. As a result, we obtain a highlevel performance model, which combines the evaluation speed of mathematical models with the structure awareness and fidelity of the simulation model. We evalua...

• Type: Conference
• Year: 2008
• Where: CISIS
• Authors: Sabri Pllana, Siegfried Benkner, Fatos Xhafa, Leonard Barolli
• Keywords: CISIS 2008, Large-scale Computing Systems, Mathematical Modeling, Performance, Software Engineering