Search

This article discusses system crashes and how each level of the system needs to be implemented correctly and the system component interfaces need to be used correctly by applications in order to prevent crashes. Several papers are cited within this article exploring the file system, a lower level component within the system, an exploration of interface-level guarantees with bug-finders, crash-consistent programs, and how the newer concept of persistent memory interacts with system crashes.

The authors look at the development of machine learning (ML) models, explaining their view that research studies on the training of ML models should report the quantity of computer usage, or compute, and discussing the need for standardization in the measurement of compute. They cite the number of floating point operations (FLOP) as a basic measurement of compute. They offer reasons for the importance of measuring compute including enabling reproducibility of ML research and aiding in the forecasting of new artificial intelligence (AI) capabilities.

This article details Charles Babbage’s theoretical computer program, the first attempt to specific how to mechanize complex algorithms with a computer, written in 1837. The article also discusses his theoretical computer, the Analytical Engine, that his program would have run on. Topics include Babbage’s first and second code tables and the design of the Analytical Engine.

The author discusses the development of computer hardware and its isolation from computer software and algorithms. She mentions the early history of computers when computers were single use, how the time and cost of hardware development present barriers to software development, and the need for collaborative research.

This article explores the concept of human-centered design for computing devices. The article focuses on heads-up computing which would provide transparent information display on wearable devices. Topics include the utilization of body-compatible hardware, multi-modal voice and gesture interaction, and resource-aware interaction in heads-up devices as well as a hypothetical case study of how a heads-up device can aid in a busy working mother’s morning.

The article describes solutions to bridging the performance gap between CPUs (central processing units) and networks. These solutions can be either software or hardware based and include Linux’s eXpress Data Path—XDP and Field Programmable Gate Arrays—FPGA. The author suggests that by using XDP programmers will be free from learning new programming languages, design paradigms or hardware architectures.

The network interface cards (NICs) of modern computers are changing to adapt to faster data rates and to help with the scaling issues of general-purpose CPU technologies. Among the ongoing innovations, the inclusion of programmable accelerators on the NIC's data path is particularly interesting, since it provides the opportunity to offload some of the CPU's network packet processing tasks to the accelerator. Given the strict latency constraints of packet processing tasks, accelerators are often implemented leveraging platforms such as Field-Programmable Gate Arrays (FPGAs). FPGAs can be re-programmed after deployment, to adapt to changing application requirements, and can achieve both high throughput and low latency when implementing packet processing tasks. However, they have limited resources that may need to be shared among diverse applications, and programming them is difficult and requires hardware design expertise. We present hXDP, a solution to run on FPGAs software packet processing tasks described with the eBPF technology and targeting the Linux's eXpress Data Path. hXDP uses only a fraction of the available FPGA resources, while matching the performance of high-end CPUs. The iterative execution model of eBPF is not a good fit for FPGA accelerators. Nonetheless, we show that many of the instructions of an eBPF program can be compressed, parallelized, or completely removed, when targeting a purpose-built FPGA design, thereby significantly improving performance. We implement hXDP on an FPGA NIC and evaluate it running real-world unmodified eBPF programs. Our implementation runs at 156.25MHz and uses about 15% of the FPGA resources. Despite these modest requirements, it can run dynamically loaded programs, achieves the packet processing throughput of a high-end CPU core, and provides a 10× lower packet forwarding latency. [ABSTRACT FROM AUTHOR]

The author argues that computational models do not need to change to adjust for cost and technological change. He claims that random access machine, or model (RAM) is the best for algorithms and data structures, discusses the software spiral (SWS) for single central processing units (CPUs), and notes that cost-effective programming is critical for performance programming.

The article discusses the benefits and drawbacks of edge computing, particularly the hurdles in deploying edge computing. It examines edge computing and deadlines, authentication and physical security, and standardization.

An introduction to an article in the journal on the topic of computing geodesic paths, circles, and distances on a digital computer is presented.

The article examines potential reductions in energy consumption of computer hardware in machine learning systems. The design of devices used in artificially intelligent systems is scrutinized and information is provided on industry innovations for the development of energy-efficient hardware and software.

A calendar of events related to computers and technology around the world for March 2016, including a symposium on interactive three-dimensional (I3D) graphics and games, a symposium on computer science education, and a conference on intelligent user interfaces.

The article discusses the lessons learned from the development of the computer operating system called MINIX. Topics covered include how MINIX was created in 1984, how the author distributed the codes in developing MINIX, and the importance of determining how to market or distribute MINIX. It also discusses the different modes of the MINIX, how MINIX was made appropriate for hardware, and details relating to the porting system of MINIX.

The article discusses the generators of imperfect sources to develop random numbers, including through referencing the computer scientist Avi Wigderson's perspective on generating random numbers. The use of random numbers in computation, cryptography, computer science and computing is discussed. The article references research on random number generators by computer scientist David Zuckerman and graduate student Eshan Chattopadhyay.

The article reports that on many occasions since 1972, clocks around the world have stopped for one second in order to adjust for the difference between the rotation of the Earth and Coordinated Universal Time. These adjustments indicate that the sun is positioned over the Prime Meridian at noon, but they also cause different problems for computers in the world.

Several letters to the editor are presented in response to articles in previous issues including "Tom Kilburn: A Tale of Five Computers," by David Anderson, published in the May 2014 issue, "Toward a Pedagogy of Ethical Practice," by Chuck Huff and Almut Furchert in the July 2014 issue, and "Responsible Programming," by Vinton G. Cerf in the July 2014 issue.