Shared Memory Parallelism Can Be Simple Fast And Scalable
Download Shared Memory Parallelism Can Be Simple Fast And Scalable PDF/ePub or read online books in Mobi eBooks. Click Download or Read Online button to get Shared Memory Parallelism Can Be Simple Fast And Scalable book now. This website allows unlimited access to, at the time of writing, more than 1.5 million titles, including hundreds of thousands of titles in various foreign languages.
Shared-Memory Parallelism Can be Simple, Fast, and Scalable
Parallelism is the key to achieving high performance in computing. However, writing efficient and scalable parallel programs is notoriously difficult, and often requires significant expertise. To address this challenge, it is crucial to provide programmers with high-level tools to enable them to develop solutions easily, and at the same time emphasize the theoretical and practical aspects of algorithm design to allow the solutions developed to run efficiently under many different settings. This thesis addresses this challenge using a three-pronged approach consisting of the design of shared-memory programming techniques, frameworks, and algorithms for important problems in computing. The thesis provides evidence that with appropriate programming techniques, frameworks, and algorithms, shared-memory programs can be simple, fast, and scalable, both in theory and in practice. The results developed in this thesis serve to ease the transition into the multicore era. The first part of this thesis introduces tools and techniques for deterministic parallel programming, including means for encapsulating nondeterminism via powerful commutative building blocks, as well as a novel framework for executing sequential iterative loops in parallel, which lead to deterministic parallel algorithms that are efficient both in theory and in practice. The second part of this thesis introduces Ligra, the first high-level shared memory framework for parallel graph traversal algorithms. The framework allows programmers to express graph traversal algorithms using very short and concise code, delivers performance competitive with that of highly-optimized code, and is up to orders of magnitude faster than existing systems designed for distributed memory. This part of the thesis also introduces Ligra+, which extends Ligra with graph compression techniques to reduce space usage and improve parallel performance at the same time, and is also the first graph processing system to support in-memory graph compression. The third and fourth parts of this thesis bridge the gap between theory and practice in parallel algorithm design by introducing the first algorithms for a variety of important problems on graphs and strings that are efficient both in theory and in practice. For example, the thesis develops the first linear-work and polylogarithmic-depth algorithms for suffix tree construction and graph connectivity that are also practical, as well as a work-efficient, polylogarithmic-depth, and cache-efficient shared-memory algorithm for triangle computations that achieves a 2–5x speedup over the best existing algorithms on 40 cores. This is a revised version of the thesis that won the 2015 ACM Doctoral Dissertation Award.
Code Nation
Author: Michael J. Halvorson
language: en
Publisher: Morgan & Claypool
Release Date: 2020-04-22
Code Nation explores the rise of software development as a social, cultural, and technical phenomenon in American history. The movement germinated in government and university labs during the 1950s, gained momentum through corporate and counterculture experiments in the 1960s and 1970s, and became a broad-based computer literacy movement in the 1980s. As personal computing came to the fore, learning to program was transformed by a groundswell of popular enthusiasm, exciting new platforms, and an array of commercial practices that have been further amplified by distributed computing and the Internet. The resulting society can be depicted as a “Code Nation”—a globally-connected world that is saturated with computer technology and enchanted by software and its creation. Code Nation is a new history of personal computing that emphasizes the technical and business challenges that software developers faced when building applications for CP/M, MS-DOS, UNIX, Microsoft Windows, the Apple Macintosh, and other emerging platforms. It is a popular history of computing that explores the experiences of novice computer users, tinkerers, hackers, and power users, as well as the ideals and aspirations of leading computer scientists, engineers, educators, and entrepreneurs. Computer book and magazine publishers also played important, if overlooked, roles in the diffusion of new technical skills, and this book highlights their creative work and influence. Code Nation offers a “behind-the-scenes” look at application and operating-system programming practices, the diversity of historic computer languages, the rise of user communities, early attempts to market PC software, and the origins of “enterprise” computing systems. Code samples and over 80 historic photographs support the text. The book concludes with an assessment of contemporary efforts to teach computational thinking to young people.
Semantic Web for the Working Ontologist
Enterprises have made amazing advances by taking advantage of data about their business to provide predictions and understanding of their customers, markets, and products. But as the world of business becomes more interconnected and global, enterprise data is no long a monolith; it is just a part of a vast web of data. Managing data on a world-wide scale is a key capability for any business today. The Semantic Web treats data as a distributed resource on the scale of the World Wide Web, and incorporates features to address the challenges of massive data distribution as part of its basic design. The aim of the first two editions was to motivate the Semantic Web technology stack from end-to-end; to describe not only what the Semantic Web standards are and how they work, but also what their goals are and why they were designed as they are. It tells a coherent story from beginning to end of how the standards work to manage a world-wide distributed web of knowledge in a meaningful way. The third edition builds on this foundation to bring Semantic Web practice to enterprise. Fabien Gandon joins Dean Allemang and Jim Hendler, bringing with him years of experience in global linked data, to open up the story to a modern view of global linked data. While the overall story is the same, the examples have been brought up to date and applied in a modern setting, where enterprise and global data come together as a living, linked network of data. Also included with the third edition, all of the data sets and queries are available online for study and experimentation at data.world/swwo.