Bruce Klimpke, Technical Director, INTEGRATED Engineering Software
Like with all new hardware developments, there is always a range of technical issues that can limit their use for different algorithms. Cluster computing has limitations to getting data in and out. A lot of basic software capabilities need to be available to take advantage of hardware improvements.
The major market trend will be the continued advancement in algorithms to make use of parallel computing for both CPUs and GPUs. The limit to optimization and testing new concepts is still mostly limited to the horsepower available.
Dr. Jon Friedman, Aerospace & Defense and Automotive Industry Marketing Manager, The MathWorks, Inc.
In tough economic environments, aerospace companies look to increase the efficiency of their engineers. Modeling and simulation helps engineers explore design alternatives quickly without the need for as much prototype hardware, which often requires a significant amount of time to produce.
Analysis software needs to help engineers seamlessly take advantage of the availability of multi-core machines and cloud computing to quickly analyze large sets of data using multi-threaded algorithms. Consider the case where a systems engineer needs to determine the physical architecture of the next-generation aircraft and must perform tradeoff studies to determine the number and configuration of the engines, the number of passengers and routes that could be flown, the fuel economy, cost per seat mile, and aircraft emissions. This type of analysis can very easily be distributed over many different cores or to the cloud.
Boris Marovic, Product Marketing Manager Aerospace & Defense, Mentor Graphics Mechanical Analysis Division
Of course, an economic crisis is influencing those who perform simulations but, in general, it is usually production that suffers more than development of the products. One would not want to fall behind the competition when the economy starts picking up again. Currently, we see an increase in the requests for simulation again.
With increasing hardware performance, simulation tasks will grow to more and more complex tasks, considering even more physics than they did previously; hence, the simulation results will grow. We will need to cope with this amount of data and be able to post-process such an amount in a fast way without it taking hours to visualize a simple surface plot showing the temperature or pressure.
David Weinberg, President & CEO, NEi Software
We see the global demand for CAE growing, with FEA growing at a faster pace within the category. Industry reports from Research and Markets forecasts a 6.5% growth overall for CAE and a 16.6% growth for FEA over the next five years.
Large manufacturing countries, led by the BRIC countries, will continue to adopt CAE technology to meet the growing demand for localized product design through onshoring or nearshoring of both design and manufacturing. The use of CAE technology by original design manufacturers (ODMs) in these economies will also enable product differentiation to help to control price undercutting by original equipment manufacturers (OEMs).