The insights contained in Gordon Moore's now famous 1965 and 1975 papers have broadly guided the development of semiconductor electronics for over 50 years. However, the field-effect transistor is approaching some physical limits to further miniaturization, and the associated rising costs and reduced return on investment appear to be slowing the pace of development. Far from signaling an end to progress, this gradual "end of Moore's law" will open a new era in information technology as the focus of research and development shifts from miniaturization of long-established technologies to the coordinated introduction of new devices, new integration technologies, and new architectures for computing.
The Discrete Event Systems Specification (DEVS) formalism has been widely disseminated in this magazine and elsewhere for its applicability to computational science and engineering. While research in parallel and distributed simulation has been active in the past several decades, the utility of many techniques for high-performance DEVS simulation has been limited. Recent research has shown that that a reconstruction of Amdahl's and Gustafson's laws for parallelizing sequential code can afford a better understanding of the underlying principles and their application to simulation of DEVS models. In this article, the author shows, in comparison to the ideal case, that a relatively simple protocol based on the DEVS abstract simulator is both generally applicable and able to achieve significant speedup in parallel distributed simulation.
On the eve of exascale computing, traditional wisdom no longer applies. High-performance computing is gone as we know it. This article discusses a range of new algorithmic techniques emerging in the context of exascale computing, many of which defy the common wisdom of high-performance computing and are considered unorthodox, but could turn out to be a necessity in near future.
Previous research has suggested that access and exposure to computing, social supports, preparatory privilege, a sense of belonging in computing, and a computing identity all contribute to women pursuing computing as a field of study or intended career. What we know less about is what keeps young women persisting in computing despite the obstacles they encounter. This article describes findings from analysis of 64 in-depth interviews with young women who in high school expressed interest in computing by looking into National Center for Women and IT's Aspirations in Computing Award. The dataset includes award winners and nonwinners, some of whom have persisted in computing and some who have not. The authors' findings suggest that multiple, redundant supports, including community reinforcement, as well as a bolstered sense of identity/belonging, may make the difference in who persists and who does not.
The growth of computing and STEM-related jobs continues, but the diversity of those inhabiting those jobs has not. Diversity in the workforce can be measured in many ways, including gender, ethnicity, sexual orientation, socioeconomic status, and persons with disabilities (physical, learning, psychological). The guest editors of this special issue discuss the articles under focus, which touch on these different types of diversity and the unique challenges that exist in broadening the pipeline for that spectrum of the population.
Associate editor-in-chief Steven Gottlieb describes a recent experience with the Partnership for Integration of Computation into Undergraduate Physics (PICUP) program.
Columnist Charles Day describes how computer software could enable fashion construction.
Blaine Willhoft and Rob Willhoft review "Software Design Decoded: 66 Ways Experts Think" by Marian Petre and André van der Hoek, a collection of short, one-paragraph observations on the habits that successful engineers have developed or learned over time.
Frustrated by another failed software installation? Wondering why you can't reproduce your colleagues' computations? This story will tell you why. It won't magically solve your problems, but it does point out a glimpse of hope for the future.
Exploratory Research to Expand Opportunities in Computer Science for Students with Learning Differences
The computer science (CS) education field is engaging in unprecedented efforts to expand learning opportunities in K-12 CS education, but one group of students is often overlooked: those with specific learning disabilities and related attention deficit disorders. As CS education initiatives grow, K-12 teachers need research-informed guidance to make computing more accessible for students who learn differently. This article reports on the first phase of a National Science Foundation-supported exploratory research study to address this problem. The authors present their education research-practice partnership, initial findings, and highlights of a collaborative process that has furthered their work to support more equitable learning in CS.
Seven instructors at five institutions adopted process-oriented guided inquiry learning (POGIL) activities for their first-year courses. These POGIL activities were designed to prompt students to reflect on the relevance of the curriculum to their own lives. Students were significantly more comfortable with computers after taking the POGIL courses, even compared to students taking the same course from the same instructors. However, there was no overall effect on students' interest in taking more CS classes. Based on these findings, the authors developed "Levels of Student Participation and Stages of Relevant Curriculum" to help POGIL faculty make their classrooms more inclusive and their curriculum more relevant. Follow-up interviews with seven instructors demonstrated a marked increase in their plans to make their course content relevant to students.
The reported study investigated the impact of the Exploring Computer Science (ECS) program on the likelihood that students of all races and genders would pursue further computer science coursework in high school. ECS is designed to foster deep engagement through equitable inquiry around computer science concepts. The course provides experiences that are personally relevant. Using survey research, the authors sought to measure whether the personal relevance of students' course experiences influenced their expectancies of success in and value for the field of computer science and whether those attitudes predicted the probability that students pursued further computer science coursework. The results indicate that students find ECS courses personally relevant, are increasing their expectancies of success and perceived value for the field of computer science, and are more likely to take another computing course.
The combination of Moore's law and Dennard's scaling laws constituted the fundamental guidelines that provided the semiconductor industry with the foundation of a technically and economically viable roadmap until the end of the previous century. During this time, the transistor evolved from bipolar to PMOS to NMOS to CMOS. However, by the mid-1990s it was clear that fundamental physical limits of the MOS transistor were going to halt Dennard's scaling by 2005, at best. Eventually the power consumed by a single IC, under the relentless growth in the number of transistors and the continuous increase in operating frequency, pushed the IC temperature beyond reliable limits. Several new memory devices operating on completely new physics principles have been invented in the past 10 years and have already demonstrated that computing performance can be substantially improved by monolithically integrating several of these new heterogeneous memory layers on top of logic layers powered by a combination of CMOS and "new switch" transistors. This new phase, called 3D power scaling, will continue to support increase in transistor count at Moore's law pace well into the next decade.
As transistors shrink to nanoscale dimensions, trapped electrons - blocking "lanes" of electron traffic - are making it difficult for digital computers to work. In stark contrast, the brain works fine with single-lane nanoscale devices that are intermittently blocked (ion channels). Conjecturing that it achieves error-tolerance by combining analog dendritic computation with digital axonal communication, neuromorphic engineers (neuromorphs) began emulating dendrites with subthreshold analog circuits and axons with asynchronous digital circuits in the mid-1980s. Three decades in, researchers achieved a consequential scale with Neurogrid - the first neuromorphic system that has billions of synaptic connections. Researchers then tackled the challenge of mapping arbitrary computations onto neuromorphic chips in a manner robust to lanes intermittently - or even permanently - blocked by trapped electrons. Having demonstrated scalability and programmability, they now seek to encode continuous signals with spike trains in a manner that promises greater energy efficiency than all-analog or all-digital computing across a five-decade precision range.
The end of Moore's law may be the best thing that has happened in computing since the beginning of Moore's law. Confronting the end of an epoch should enable a new era of creativity by encouraging computer scientists to invent new biologically inspired paradigms, implemented on emerging architectures, with hybrid circuits and systems that combine the best of scaled silicon CMOS with new devices, physical interactions and materials.
Physics, medicine, astronomy — these and other hard sciences share a common need for efficient algorithms, system software, and computer architecture to address large computational problems. And yet, useful advances in computational techniques that could benefit many researchers are rarely shared. To meet that need, Computing in Science & Engineering (CiSE) presents scientific and computational contributions in a clear and accessible format.Subscribe to IEEE NEWS feed