Processing Storage Information Batteries Instead Of Electricity Storage
Our Daily Use Of Devices Such As Mobile Phones Imposes A Lot Of Computational Burden On Data Centers, And Because, Of These Calculations, A Lot Of Electricity Is Consumed.
Electricity Storage, Although renewable sources can supply some of this electricity at intervals, such as at night when it is impossible to use a source such as a sun, the renewable energy solution is limited to batteries.
Introducing the idea of information batteries, researchers have proposed an answer to this challenge, which we will look at in the “Mobile World” of this issue. In addition, we will refer to a new method for predicting high-risk areas of Corona.
One of the challenges in renewable energy is that resources such as the sun or wind are not always available. As a result, unlike fossil fuels, electricity generation is not based on renewable sources and is interrupted for hours a day.
One way to address such challenges is to use batteries to generate electricity. Although battery manufacturing technology has grown significantly in recent years, it is still impossible to store energy on a large scale in batteries.
Researchers at the University of Southern California have proposed “information batteries” to solve this problem.
The idea is to process data at intervals that are not limited to power generation by predicting the workload of data centers and storing the results of these processes for later use.
Because data storage is considerably simpler than electricity storage, this would be an optimal way to use renewable resources better.
For example, instead of storing the energy generated by a solar panel in the battery so that a data center can use it at night, we can perform these processes by predicting the computations that the data center needs when the power is on.
It is produced. In this case, the data center can store this data for future use. This computational workload can be compared to a jigsaw puzzle, as each piece is a part of the puzzle. The critical point is that, in practice, a large part of the computational workload is predictable.
Figure 1 – Researchers have come up with the idea of information batteries for the constant use of renewable sources.
For example, most computations can predict without processing in a data center. In addition, a considerable part of the calculations’ results can use in various applications. In this case, by eliminating duplicate tasks in power consumption, more savings can make, and we do not need to repeat the calculations every time.