top of page
Meditating with
Microprocessors
Hello My Family, Friends, Colleagues
We have some great news to share with you all after months and months of toiling. Our technical team would have me believe that a few weeks’ delays are no delay at all. To be fair to our team, the goal post was extended by our esteemed client. Last couple of months we have been really stomped for time.
Controversies around delays notwithstanding, we are extremely excited to announce our Artificial Intelligence based Microprocessor Tuning System. In laymen terms, it is a framework that predicts load and preconfigures the system for it to save precious time and money. This is by far the greatest work that we have done. At the moment, we target only Intel Microprocessors.
The article on Mohit's blog documents the working of the system. This will double up as our official documentation till such time we get our product documentation ready. And just so that we are clear, microprocessors don't meditate. It is just a metaphor to ease into the technical mumbo jumbo.
Also, We are looking for a name, and we would value your suggestions so keep them coming. Your support over the years can only be underemphasized. It keeps us going, to say the least.
Given the situation, because of the pandemic, it is a soft launch. But the moment the situation gets better we will celebrate and savor the moment.
Warm Regards,
Veena Mohit
CEO and Founder | STILLWATERS AI
Artificial Intelligence based Hardware
(Microprocessor) tuning:Implementing a very simple idea
In the current series titled Meditating-with-microprocessors, I demonstrate the use of Artificial Intelligence to tune microprocessors for Ultra-Low-Latency and Realtime loads. The techniques, in general, can be extended to other components of a computer system like storage devices, memory, etc. However, the article series and my work is currently restricted to Intel microprocessors only. In the future, we may extend this to other hardware components of a computer system. This is a very specialized and intense field and hence I intend to break it down using the first-principles approach into simpler pieces of technology that are easy to understand.
There are 5 parts to the series Artificial Intelligence based Hardware(Microprocessor) tuning:
Part 1. Implementing a very simple idea
Part 2. A crashcourse in Microarchitecture and Linux CPUIDLE interface
Part 3. Trading off power for UltraLowLatency
Part 4. Artificial Intelligence guided Predictive MicroProcessor tuning
Part 5. Appendix:Tools of the trade
In the balance then, this is a documentation of my journey navigating these utterly specialized fields (Microarchitecture and Artificial Intelligence), how to marry them together, the issues I faced, the respective solutions, what (how much) are the benefits if any, and what to have in the toolbox
bottom of page