China
Africa
Explore chapters and articles related to this topic
Carbon Nanotube Electronics
Published in Ann Rose Abraham, Soney C. George, A. K. Haghi, Carbon Nanotubes, 2023
The celebrated Moore’s law, stated by Gordon E Moore, co-founder of Intel, predicts that the rate of advancement in processor technology doubles every 18–24 months or in other words the number of transistors on a processor doubles every 18 months. This observation was given in his article, “Cramming more components onto integrated circuits” published in 1965. During this time, the number of transistors in a single chip was just 32, which has now increased to half a billion. This shows the significance and exactness of the prediction made by this visionary. Moore’s law was given a mathematical form by Meindle.28N = F−2 D2 PE
It's Not Just the Algorithms, Really!
Published in James Luke, David Porter, Padmanabhan Santhanam, Beyond Algorithms, 2022
James Luke, David Porter, Padmanabhan Santhanam
With the advent of the internet age and the maturity of the IT infrastructure, the collection of large amounts of data became possible. The Moore’s law advances in computing architecture i.e. Central Processing Units (CPUs), Graphics Processing Units (GPUs), on-board memory and storage, made creation and use of much larger neural networks practical. All these led to the new age of Deep Neural Networks (DNN). A DNN is a Neural Network with many more hidden layers. In the 1990s, we were building Neural Networks with one, or sometimes two, hidden layers. We are now able to build networks with huge numbers of hidden layers. Consider the two networks in Figure 3.8. The conventional network on the left with just four neurons in the hidden layer has 26 parameters to adjust. One of the popular Deep CNNs for classifying 1000 objects from images had eight intermediate layers, 650,000 neurons and 60 million parameters to optimise! [18]. This sheer number of parameters that need to be configured means that we need significant processing power and a massive volume of training data.
Moore’s Law: In the 21st Century
Published in Niladri Pratap Maity, Reshmi Maity, Srimanta Baishya, High-K Gate Dielectric Materials, 2020
“If Moore’s law is simply a measure of the increase in the number of electronic devices per chip, then, Moore’s law has much more time to go, probably decades,” said Intel CTO Justin Rattner in recent times in an interview with Network World. Powel has shown the technology node from 130 nm to 22 nm versus the performance of the semiconductor chips (Powell, 2008). The gate terminal length possesses on shrinking as the technology node decreases. As what most researchers anticipated for previous few decades, the performance or the speed of the designed semiconductor chips must be increased as well.
KNOWM memristors in a bridge synapse delay-based reservoir computing system for detection of epileptic seizures
Published in International Journal of Parallel, Emergent and Distributed Systems, 2022
Dawid Przyczyna, Grzegorz Hess, Konrad Szaciłowski
To address the computational problems of the modern age, scientists are working to improve existing hardware and software technologies, as well as develop unconventional and/or hybrid approaches known as heteroic computing [12]. The growth of the first approach is commonly known to be subjected to Moore’s law, whereas the second approach attempts to overcome its limitations through embedded multifunctionality and interaction with the environment [13]. Moore’s law states that computational capabilities of conventional transistor technologies will double every few years. However, there is a lower limit to miniaturisation due to (i) the granularity of matter and (ii) the effects of quantum tunnelling of electrons through the gate of the transistor and (iii) heat management problems, which are the main reasons for looking for other computational technologies [14–19]. Second major issue is the so-called Von Neuman Bottleneck. It is a problem that limits the computing abilities of classic computers resulting from the separation of memory functions and processing that require communication between these components, which in turn limits the speed of the computing process. One of the novel technologies that aims to resolve some of modern computing problems is the application of memristors (and other memristive elements) and memristive circuits [20–24].
Why human factors science is demonstrably necessary: historical and evolutionary foundations
Published in Ergonomics, 2021
J. C. F. de Winter, P. A. Hancock
Now, in 2020, robots (i.e. intelligent self-controlling machines) are widespread in many facets of life, including driving (automated driving), households (e.g. robotic vacuum cleaners), warehouses, and agriculture. Moore’s law (Moore 1965), the notion that the number of transistors on a computer chip doubles about every two years, suggests that computers will enable increasingly intelligent applications. With new developments in reinforcement learning, an increase in machine intelligence and a more widespread use of computers/robots, is anticipated.4 Some have argued that technological developments are slowing down and that Moore’s law is coming to an end. This statement appears to be incorrect, as recent data show that the number of transistors on a chip continues to increase exponentially until the present day (Schwierz and Liou 2020; Sun et al. 2019).
Research progress of diamond/aluminum composite interface design
Published in Functional Diamond, 2022
Zengkai Jiao, Huiyuan Kang, Bo Zhou, Aolong Kang, Xi Wang, Haichao Li, Zhiming Yu, Li Ma, Kechao Zhou, Qiuping Wei
The rapid development of information technology has promoted the progress of electronic devices to integration, miniaturization, and lightweight. Back in 1965, Intel co-founder Gordon Moore formulated his famous Moore’s law: The number of transistors on a chip would double every 12 months. However, the high integration of electronic components makes the power density increase continuously, and the calorific value rises sharply. The problem of heat dissipation poses a severe challenge to the development of the electronic information industry. A new international roadmap for semiconductor technology will no longer follow Moore’s Law if the heat dissipation problem has not been solved [1–3]. The heat dissipation problem has increasingly promoted the development of thermal management materials. Common thermal management materials include polymers, ceramics, metals, and metal matrix composites. Among these, metal matrix composite has become one of the research hotspots. It uses metal or alloy as the matrix, utilizes the second phase with high quality to strengthen the body, displaying excellent performance of each component. Table 1 lists the main properties of common metals and metal-based thermal management materials [4]. Among them, metals such as Al and Cu are low-cost, easy to process, and exhibit high thermal conductivity, but their high thermal expansion coefficient limits their application. Invar and Kovar alloys have a low thermal expansion coefficient, but low thermal conductivity and high density. W-Cu, Mo-Cu, SiC/Al, and other materials have thermal expansion coefficients matching with semiconductor materials such as Si and GaAs, but their thermal conductivity is mostly below 200w/(m k), which is difficult to meet the heat dissipation requirements of high-power integrated circuits.