2024-03-24 07:00:10
Researchers have developed a new method to accelerate the processing speed of electronic devices, such as smartphones or computers, without the need to replace existing components. This advance could double the speed (We distinguish:) of processing, while consuming less energy (In the common sense energy designates anything that allows you to carry out work, make food, etc.).
Modern devices are equipped with several types of microchips, including the central processing unit (CPU), graphics processing unit (GPU), hardware accelerators for artificial intelligence (AI), and data processing units. digital signal (General terms A signal is a simplified and generally coded message. It exists…) for audio. Usually, these components process data separately and sequentially, which can slow down overall processing.
To overcome this challenge, a team of scientists proposes an innovative framework where processing units operate in parallel, rather than one after the other. This method, called “simultaneous and heterogeneous multi-threading” (SHMT), allows different units to work on the same region of computer code at the same time (Time is a concept developed by human beings to understand… ), thus significantly improving the effectiveness of the treatment. Unlike “software pipelining”, a method that allows different components to work on different tasks simultaneously, SHMT allows for more flexible distribution of tasks between components. This means that different processing units can tackle the same part of the code simultaneously, then move on to new tasks once their part is completed.
In addition to speeding up processing, SHMT is more energy efficient. Many tasks normally assigned to power-hungry components, like the GPU, can be delegated to less power-hungry hardware accelerators.
This approach was tested on a prototype system consisting of an ARM multi-core CPU, an Nvidia GPU and a TPU hardware accelerator, showing almost twice the performance and 51% lower power consumption. compared to a traditional system.
Adopting this software framework over existing systems could not only reduce hardware costs, but also decrease carbon emissions and water demand. (Water is a ubiquitous chemical compound on Earth, essential for all… .) necessary for cooling large data centers, thanks to more efficient and environmentally friendly management of workloads.
However, the researchers point out that their study, based on a prototype, requires further work to assess how this method can be applied in practical contexts and what types of applications could benefit most from it.
1711267794
#innovation #doubles #speed #existing #smartphones #computers