Our demand for the software to be fast. We don’t like to wait until computer will perform all actions and produce results. Depends on the situation and use case, speed can be something very disturbing or even more – change the overall working process. Time ago, simulation and analysis software embedded into CAD systems changed the way engineers designed mechanical parts. Another, more PLMish example might be related to visualization of complete product configuration and performing clash analysis.
Nowadays, the computing mindset is taking us towards new approaches in data processing using new processor technologies as well as network computing opportunities. MIT Tech review article Graphics Chips Help Process Big Data Sets in Milliseconds is talking about how to visualize big data on cheap computers in milliseconds. The writeup is mentioning massively parallel database that allows to achieve big speed gain by storing data on the onboard memory of GPUs, which allows 70x performance gain. You can use the following link to use public interface to product tweetmap. Here is an interesting passage I captured:
The can be used to visualize 50 million geocoded tweets posted between September 28 and October 6. The tool allows users to explore different search terms, examine broad geographical trends, and zoom in on each tweet. For each of the 30 frames per second it generates when animating Twitter, Map-D scans all the tweets that have been loaded on the GPUs, constructing visualizations such as maps of how word usage—which could include mentions of a product name or news item—is propagating across a region or around the world in real time.
Watch the following video, which is quite self explaining.
What is my conclusion? Fast data processing can be a key technology to realize the opportunity to change design and engineering process. To visualize design options, analyzes variants and make a final assessment to validate potential mistakes – this is only a very short list of options. Moving forward you can think about analyzing of huge databases for potential failures and customer issues, potential supplier components. Theses days, technologies developed for web, social networks and mobile domain can be leveraged tomorrow to change future landscape of engineering software. Just my thoughts…
Best, Oleg
