If there has been one constant throughout the decades of change that has produced the data environment we have today, it’s that basic technologies continue to strive for greater performance.
Nowhere is this more evident than on chip-level architectures. Even as the physical limitations of Moore’s Law come into play, designers are busy creating new interconnects, interfaces and software constructs to keep the bits flowing, and even coming up with new structural technologies that could very well restart the field of digital signal processing on a whole new trajectory.https://o1.qnsr.com/log/p.gif?;n=203;c=204663295;s=11915;x=7936;f=201904081034270;u=j;z=TIMESTAMP;a=20410779;e=i
Intel, of course, is at the forefront of these efforts, says NewsFactor Network’s Jef Cozza, having just released the latest in the Core i7 line, including new desktop versions of the Kaby Lake family that should help boost performance, security and graphics capabilities for enterprise desktops and laptops. Aside from removing the need for a dedicated GPU for video conferencing and other rich media applications, the new devices double the battery life to 10 hours over existing Skylake machines, and provide a 65 percent data productivity improvement for notebooks running 4k workloads.
AMD, meanwhile, is drawing kudos for its new Ryzen CPU that sports a new “branch prediction” technology built around neural networking and artificial intelligence that streamlines the flow of data through the processor core. According to The Founders Daily’s Scott Taylor, the chip has already received orders from leading motherboard designers and is coinciding with a renewed bullishness for the company’s stock on Wall Street. The chip also features a new fabric architecture that manages clock speed, temperature, voltages and other functions in order to drive performance and minimize power consumption.
Elsewhere in the CPU sphere, new architectures are focusing on helping the enterprise deal with increased edge computing and the massively parallel workflows that are expected in Big Data and IoT environments. As Moor Insights and Strategy’s Jimmy Pike noted to Forbes recently, the ARM architecture is probably not going to see much action in the data center, but it should see opportunities where the edge interacts with mobile and embedded data streams. At the same time, IBM’s OpenPOWER is working to incorporate Nvidia’s NV-Link interface that, along with various process upgrades, may find its way into hyperscale architectures by Google and others.
And while Hewlett Packard Enterprise isn’t a name that usually comes up in conversations about chip development, the company is actually moving forward on new light-based designs that would essentially redraw the lines for CPU performance. The company has developed the largest such processor of its kind, says Digital Trends’ Mark Coppock – a DARPA-based device that features more than 1,000 optical components to derive digital information from the magnetic properties of photons rather than electrons. The design is still very much experimental and is subject to vibration, noise and other factors, but if successful, it could end up pushing Moore’s Law into the realm of quantum computing.
Few of these chip-level developments will have direct bearing on the decisions that CIOs need to make in this era of digital transformation, but it is comforting to know that advances in processing power are not at an end just because today’s silicon architectures are approaching their performance limits.
There is more than one way to process a bit, and the needs of an increasingly data-hungry world all but guarantee that new solutions will continue to supplant existing ones in the eternal quest for greater digital, and human, productivity.
Arthur Cole writes about infrastructure for IT Business Edge. Cole has been covering the high-tech media and computing industries for more than 20 years, having served as editor of TV Technology, Video Technology News, Internet News and Multimedia Weekly. His contributions have appeared in Communications Today and Enterprise Networking Planet and as web content for numerous high-tech clients like TwinStrata and Carpathia. Follow Art on Twitter @acole602.