It’s not quite time yet to be doing “the year in review” articles, but here’s one thing that flew below my radar: Zimbabwe officially launched a 36-teraflop supercomputer in February 2015.
The supercomputer, housed at the University of Zimbabwe’s Zimbabwe Centre for High Performance Computing (Zim-CHPC), was installed by Inspur Group of China funded by a $5,4 million loan courtesy of the Chinese government. Zimbabwe became the third (or fifth – depending on who you ask) African country to host a supercomputer, and number 2 on the continent in terms of raw computing performance.
Wikipedia describes a supercomputer as “a computer with a high-level computational capacity compared to a general-purpose computer”. The speed of a supercomputer is measured in FLoating-point Operations Per Second (FLOPS or flops), and the designation of ’36 teraflops’ means at its peak speed, the supercomputer can execute 36 trillion floating point operations.
For a rough benchmark of how that stacks up against the best in the world: the current number 1 supercomputer (China’s Tianhe-2) clocks in at 33,867 petaflops (33 867 quadrillion FLOPS, or 33 867 teraflops – though I probably don’t need to convert between trillions and quadrillions for our Zimbabwean readers). Supercomputer number 500 runs at less impressive 164 teraflops. More on the Tianhe-2: it has just over 3,1 million Intel Xeon processor cores and has a power draw of 17,8 Megawatts – which is more than half of what ZESA is currently producing at its Harare power station (30 Megawatts).