Researchers from the University of Wisconsin-Madison have announced at the recent Hot Chips Event in Cupertino, California, that they have created the first open source general-purpose graphics processor (GPGPU).

Dubbed MIAOW, which stands for Many-core Integrated Accelerator Of the Waterdeep, is an open source implementation of AMD's Southern Islands instruction set architecture. MIAOW is the latest in a series of steps meant to keep processor development in step with Moore's Law. The researchers published a white paper on the device, which they outlined the architecture of the open source MIAOW GPU.

MIAOW was designed by a team of computer researchers at the University of Wisconsin led by Karu Sankaralingam, an associate professor of computer science at the University of Wisconsin-Madison. One of Sankaranlingam's students was working on an out-of-order processor design when chipmaker AMD released the instruction set architecture for one of its graphics chips- AMD Southern Islands ISA. This sparked the idea to build something new, an open source GPU.

The12-man research team developed the MIAOW GPU in 36 months, their goal was simply to create a functional GPU without setting any specific areas or performance goals. The goal was to emulate a full system and not to compete with commercial designs. The team also claimed that the device is capable of running full, unmodified OpenCL applications. OpenCL or Open Computing Language is a framework for writing programs that run across heterogeneous platforms consisting of central processing units (CPUs), graphics processing units (GPUs), Field-programmable gate arrays (FPGAs) and other processors.

The MIAOW GPU is basically a stripped-down version of commercial GPUs. It implements a subset of the AMD GPU instruction set. The current GPU design includes a host CPU that assign a kernel to the GPGPU, a memory controller, and a cached memory hierarchy. The device uses only 95 instructions and 32 compute units. But research team are reportedly in the process of adding a graphics pipeline to the current GPU design, a process expected to take about six months.

Graphics processors or GPUs are increasingly replacing CPUs for tackling high-performance computing and crunching big data. They are very power efficient and can provide high-performance computing. Potential applications could be in gaming, driverless cars, IoTs (Internet of Things) and deep learning.

Currently, MIAOW is strictly an academic research project, it does not posses the auxiliary logic required to create actual graphical output and lacks the logic needed to connect it to a specific memory or system bus.

Image Courtesy of: Place Components