May 23, 2024
Computer Processing Speeds

Revolutionary Method to Enhance Computer Processing Speeds Using Existing Hardware

A groundbreaking method that could potentially double the processing power of smartphones, tablets, personal computers, and servers has been unveiled. Hung-Wei Tseng, an associate professor of electrical and computer engineering at UC Riverside, introduced a novel approach to computer architecture in a paper titled “Simultaneous and Heterogeneous Multithreading.”

Tseng highlighted the prevalent use of graphics processing units (GPUs), hardware accelerators for artificial intelligence (AI) and machine learning (ML), and digital signal processing units in modern computer devices. These components operate independently, leading to the inefficient movement of information between processing units, causing a bottleneck.

In their research, Tseng and UCR computer science graduate student Kuan-Chieh Hsu introduced the concept of simultaneous and heterogeneous multithreading (SHMT). They demonstrated the implementation of an SHMT framework on an embedded system platform that leveraged a multi-core ARM processor, an NVIDIA GPU, and a Tensor Processing Unit hardware accelerator simultaneously.

The results were impressive, with the system achieving a 1.96 times speedup and a 51% reduction in energy consumption. Tseng emphasized that the existing processors within devices could be harnessed without the need for additional hardware.

The potential benefits of this method are vast. By utilizing the existing processing components concurrently, it could lead to cost savings on computer hardware and a reduction in carbon emissions associated with powering servers in large data processing centers. Additionally, it may decrease the demand for freshwater used for cooling server facilities.

While the research presents promising outcomes, Tseng’s paper underscores the necessity for further exploration into system implementation, hardware compatibility, code optimization, and the specific applications that could derive the most significant advantages.

The research findings were shared at the 56th Annual IEEE/ACM International Symposium on Microarchitecture held in Toronto, Canada, in October.

1. Source: Coherent Market Insights, Public sources, Desk research
2. We have leveraged AI tools to mine information and compile it