HP has announced that a working prototype of its forthcoming supercomputer system is finally ready after years of research and development.
Nicknamed ‘the Machine’, the new supercomputer was first announced more than a couple of years back in 2014. HP touts it as the world’s first ever memory-driven computing platform, meaning instead of relying on processors, it will be largely dependent on memory to perform tasks. One of the core motives behind the project was to develop a supercomputer that outperforms all existing technology by a long shot, and it seems like HP has achieved that goal beyond any doubt.
The Machine is capable of storing massive amounts of data in comparatively much smaller spaces. In fact, such is its storage efficiency that the supercomputer can dramatically reduce the space required an entire data center.
“Its simulations show that memory-driven computing can achieve improved execution speeds up to 8,000 times faster than conventional computers,” HP said.
Worth noting, despite all its potentials, the project is still under development and could be years away before a commercial version is ready. Also, its target market is high-end servers that big companies use to bring you stuff such as YouTube or Facebook.
However, just because the Machine has nothing to do with consumer PCs doesn’t mean that you should not be excited about this new achievement by HP. After all, as pointed out by the company, the new technologies used in the Machine — more specifically the memory driven computing — will eventually one day migrate to consumer products.
The Machine basically makes use of photonics to transmit data using light — thanks to its massive and super-fast memory pool. This, in turn, makes it possible for processors to access large amounts of data from large memory pools at much faster speed. Unlike current supercomputers where the transfer of data slows down while moving from one processor to the other, the Machine’s configuration enables its processors to access the same memory pool simultaneously, leading to a faster and much improved overall performance.
https://youtu.be/2VG59FYkPdM%20
[Source]