The days when the X86 processor could do just about every kind of processing in the datacenter are gone
"It's hard to believe it is true, but one need only look at the rapidly expanding line of compute engines from Intel over the past few years to know that it is true," opines Timothy Prickett Morgan in
The Next Platform
"The end of Dennard Scaling (making clocks faster in the same power budget, thus boosting performance, thanks to voltage and current scaling downwards as transistors shrink) and the slowdown in Moore's Law (doubling the transistors on and therefore halving the cost of processors every two years or so) are upon us. And just as the amount of data everyone wants to chew on for some kind of monetary gain or scientific advance (they are really the same thing) is growing at a horrendous pace..."
Intel Corp. is turning its attention to the convergence of high-performance computing and artificial intelligence
"In addition," writes Mike Wheatley in SiliconANGLE
, "Intel announced its oneAPI initiative, which aims to provide a simpler programming model for developing HPC and AI applications that can run on any kind of architecture, including GPUs, central processing units, field-programmable gate arrays and neural network processors.
Announced at the Supercomputing 2019 event in Colorado today, the new Ponte Vecchio discrete GPUs are built on Intel's Xe architecture using its most advanced seven-nanometer process and have been designed especially for HPC and AI training workloads..."