We’ve worked with PCs since the introduction of the original IBM Personal Computer, and one thing we know: the evolution of computing is never-ending.
Recently, Hewlett Packard Enterprise Co. reached a milestone in efforts to deliver a new kind of computer. And its future business may be more driven by the components it developed in the process, than as a complete system.
According to the Wall Street Journal, HPE announced in November in London the working prototype of what it calls “the Machine.” The Machine flips long-held conventional PC wisdom on its ear, in that the new product “leans heavily on memory technology to boost calculating speed.” Programmers are just now starting to work on software needed to exploit the system, which is not expected to be widely available until 2018.
HPE’s near-term plan, according to Journal reporter Don Clark, is to use components developed for the system in its conventional server systems. One component already in production is a chip called X1 and some related components intended to sharply lower the cost of sending data over speedy fiber-optic connections. Engineers claim this could make data transmitted from one machine to another in a network as fast as data is currently transmitted within a single computer.
The technology would also permit vertical stacking of components to save space, breaking with yet another tradition (i.e., horizontal stacking). In traditional computing, the focal point of system design is processors, with memory considered a scarce commodity. With the Machine, HPE claims, hundreds of thousands of processors can simultaneously tap into data stored in a vast pool of memory contained in one server or multiple boxes in a data center. Noted CEO Meg Whitman, “It’s breaking every rule.”
The net effect of the Machine is a one hundredfold increase in computational speed – immensely useful in large chores like helping airlines respond to disruptions, preventing credit card fraud and assessing the risk of vast financial portfolios.
And if there’s one thing we’ve learned in the tech arena, most of the innovations that first solve big problems in large scale systems eventually move down to our local distributed business networks, much to the benefit of small businesses across the world.