what is ways to overcome power problem in network on chip and processor technology that make the gap between closer?
is the batteries the problem of power consumption in this field of computer science (battery capacity) or there is more options for solving it. (Really efficient method);
I am trying to research on solving this problem efficiently. can you please help me in this research?
and what other sciences is going to meet in this scope? (like chemistry for batteries and physics etc.)
thanks for attentions
Hi Yasin
There will be a number of issues that will need to be addressed here. Ideally, you would want to do it in as cost effective a manner as possible. First, the obvious option would be to see if you can find a nearby power source, either to drive the hardware, or to top up the charge in the battery. This is often not practical, such as in IoT at remote locations, or remote SCADA system elements. The next logical choice would then be to see if you can install a solar powered feed to continuously recharge the battery. A water powered generator would also fit the bill here. Assuming these options are impractical, then increasing battery capacity would be the next logical move, either by research to provide greater capacity (very expensive) or by installing a bigger battery. If these turn out not to be practical, then you need to look very carefully at your hardware to see whether it is matched to the demands being placed on it, and whether the software running on it is over bloated, or signals are being passed through it with too high a frequency. Matching CPU capability to load would be a useful start. Minimising the operating system and software code running on the hardware would also be a smart move. Most operating systems and software running thereon are heavily over-bloated. If you are running something with 150MB of code, but you are only using a very small portion of this, then by running exactly what you need, there will be a much smaller draw on your available battery capacity. Similarly, cutting down the frequency of signals being passed though the hardware would also reduce the load. You could create a temporary data store on a high speed SD card, then update once a day instead of doing live updates, subject to your needs. Usually, you can cut down on frequency without too much impact on the overall system functionality. These last three suggestions you can carry out now, with minimal cost outlay. All the previous suggestions will involve some element of capital spend and/or research which will take some time to gain any positive benefits.
Hope that helps.
Regards
Bob
Your question is squarely a thesis topic Yasin.. We can not answer it only in one simple sentence ...
- Badges
- Science topic
- Similar topics
- Computer Science