Visualization of computer architecture simulation data for system-level design space exploration

Abstract. System-level computer architecture simulations create large volumes of simulation data to explore alternative architectural solutions. Interpreting and drawing conclusions from this amount of simulation results can be extremely cumbersome. In other domains that also struggle with interpreting large volumes of data, such as scientific computing, data visualization is an invaluable tool. Such visualization is often domain specific and has not become widely studied and util-ized for evaluating the results of computer architecture simulations. In this paper, we describe an interactive visual tool for exploring and analyzing alternative architectural solutions at multiple levels of abstraction. As a proof of concept, we have used this tool to create a coordinated, multiple-view visualiza-tion for our computer architecture simulation and exploration environment, called Sesame, which aims at system-level performance analysis and design space ex-ploration of multi-core embedded systems. Our results show that our multivariate visualization support can help designers to more easily understand the reasons be-hind the differences in performance of different design choices, and thus gain more insight in the performance landscape of the design space