Skip to main content
    Brytlyt back from world first at Nvidia GTC 2016

    Brytlyt back from world first at Nvidia GTC 2016

    Having touched down back in the UK after spending the last three two days in Amsterdam at the NVIDIA GTC conference it seems amazing to me what has been achieved that just in the space of just over two years since we took our first tentative steps as one of ten finalists at NVIDIA’s very first Early Stage Challenge at the 2014 GTC conference in the heart of Silicon Valley. Fast forward a year and now we’ve been in Amsterdam we were attending the GTC as a company that is starting to move on a pace in terms of funding and security. But Just as importantly, having our work included in a coalition of hardware and software companies who worked with the University of Bielefeld to roll out and demonstrate a prototype that heralds a real breakthrough in GPU processing speeds and ease of use.

    It was genuinely a world first we showcased in a packed Expo Hall in the Moscow room at the conference, demonstrating the world’s fastest SQL database.

    We believe the world of data processing is changing because of GPUs and changing for the better. We are going to see GPU units in the future that promise even faster speeds, more energy efficiency and the elimination of data transfer bottlenecks. Our collaboration with Toradex, Antmicro, Christmann and the University of Bielefeld means that for the first time companies using a specialised Tegra GPU carrier will be able to deploy to data centres much more processing power and speed while reducing energy consumption and the knock on ability of being able to utilise much more data in the future.

    The new prototype extends the GPU memory footprint and allows the removes the data transfer bottlenecks in standard GPU (Tesla) systems. This means that SQL databases will now also be able to use Tegra effectively, some that previously was impossible. Standing alongside Research Associate Jens Hagemeyer we were able to show how our collaboration was able to build a unit that allows gives the user to have an easy to consume interface platform by adopting PostgreSQL as the userinterface and allowing legacy code and reporting infrastructure to be able to run out the box using the new hardware.

    I really believe this faster, more efficient processing will opening up a new world for those enterprises and data centres who until now have not been able to access this kind of processing power, speed and energy efficiency. Roll on NVIDIA GTC 2017.

    Back to Blog