The demonstration was intended to show Prodigy can run customers’ legacy applications transparently at launch. Prodigy will also run those applications with better performance than contemporary or future ARM or RISC-V processors. When running hyperscale data centre workhorse programs like Hadoop and Apache, this capability will ensure that a broad range of applications can run on Prodigy out of the box. Tachyum is currently recompiling those workhorse programs to Prodigy native code.
Tachyum is also working on native Linux distribution with many applications for the processor’s launch in 2021. The company will also provide a dynamic binary translator that converts x86, ARM or RISC-V code to Prodigy native ISA. To prove that Prodigy can run heterogenous system applications efficiently on the same universal silicon, one demonstration ran a web server application on Prodigy using a mix of native and x86 code.
Tachyum’s Prodigy is capable of running HPC applications, convolution AI, explainable AI, general AI, bio AI, spiking neural networks and normal data centre workloads on a single homogeneous processor platform with a simple programming model. Tachyum claims that Prodigy has improved computational performance, energy consumption, hardware utilization and space requirements over the chips currently used in hyperscale data centres.
“Having a readily available solution and easy to use with massive amounts of software, demonstrates the foundation for success of a platform,” said Dr. Radoslav Danilak, Tachyum founder and CEO. “This demonstration of Prodigy’s ability to run software correctly – even legacy code from x86, ARM or RISC-V processors – shows that we will enable customers to seamlessly use the applications they are using today from day one of Prodigy’s launch.”
Prodigy will enter volume production in 2021.