We hear everyday about how the so-called Internet of Things - the growing interconnectivity of previously standalone devices - will revolutionise the way we handle even simple tasks. But it creates a problem: data overload. In an environment where everything is connected to the cloud, we will be creating so much data that we don't have the computing power to store it, let alone understand it.
HP have been in the computing business for a long time, but they believe they have the developed the first product which can handle what they call this “exponential increase in data”. In order to do this, they have invented a whole new computer architecture, with massively enhanced data processing capabilities.
Six times faster than a regular server, while using 80 times less energy.
Dubbed ‘The Machine’, this new computer is, according to HP, six times faster than a regular server, while using 80 times less energy. In more numeric terms, The Machine can process, 160 Petabytes (167,800,000GB) of data in just 250 Nanoseconds.
In addition, HP is confident that this technology will not just be used in large server computers, but also could be scaled down into personal computers, laptops and even smartphones. In fact, they believe that it might even be possible to create a smartphone with up to 100 Terabytes of memory.
In order to achieve these massive gains in computing power, storage, and energy efficiency, HP has tapped into a number of key technologies. Firstly, rather than using a small group of processing chips which are good at all tasks, The Machine has many groups of cores, with each group excelling at a specific task. Secondly, the computer uses photonic (laser light) information transfer, rather than through copper wire, greatly reducing power consumption and heat. Finally, the design architecture, makes use of new ‘memristor’ technology which allows data to be stored in a certain state even when power is off.
While this technology certainly has a lot of promise, it is sadly still a few years off. HP expects to have its first samples ready by 2015, and its first Machines ready for demonstration by 2018.
Cool Ad Here