If one big complex thread being executed too fast is risky as we saw, because there’s a physical limit to sheer clock speed and cheating optimisations are unsafe, then, we all know the other option: what about doing many simple and small things independently? To put it differently: if we cannot make the clock any […]
modern hardware
Processors, Part Two: the Present. Meltdown and Spectre
Imagine a restaurant. A professional chef, in charge of the best pizzas in town, and nobody knows how’re they done. It’s such a secret, those magic ingredients. And imagine a bunch of apprentices, supposed to help him. Imagine them following the chef’s orders. And then see, the chef going to the fridge to take some […]
Processors, Part One: the Past. The Architecture
How to widen the bottleneck of processing? If the solution were trivial, it wouldn’t merit a post – in fact the workarounds are of the kind that’s easy to invent but hard to formalise. Like the megapixels are not a true indicator of the quality of a camera, or the decibels are not a true […]
On so called low-level and performant languages
Last time, we dug into a new debate: are languages really decoupled from the hardware that runs them? Are the semantics of a language modelling mainstream hardware, and should they? Can we escape to new models of hardware, and new models of semantics? We saw that Backus made a good point arguing that languages have […]