arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Dylan Dewitt

2 papers · 2008–2014

2008: 12009: 02010: 02011: 02012: 02013: 02014: 1
2008peak 1 in 20082014

Papers

showing 1–2 of 2 RSS

Cultural Rights and Political Authority in Maya Guatemala

Dylan DeWitt

Political Science and International Relations · VCU Scholars Compass (Virginia Commonwealth University) · · cited 0× · OpenAlex ↗ · more like this ·

Following a civil war that engulfed the nation for thirty-six years, the Guatemalan state has taken steps to integrate previously remote territories into its broader political and economic system. This has led to the increased political inclusion and economic integration of Mayan communities that had remained on the outskirts of Ladino society. Unfortunately, not much attention has been given to understanding the effects of this process on indigenous political institutions. After traveling to the Western Highlands region in December 2013 and surveying research from political science, anthropology, and environmental science, I have concluded that the 1996 Peace Accords have not helped to empower local Mayan political institutions. In fact, this process of political and economic integration has delegitimized indigenous political authority through the state institutionalization of private…

Parallel Benchmarks Inspired by Berkeley Dwarfs

Dylan Dewitt

Artificial Intelligence · · cited 0× · OpenAlex ↗ · more like this ·

Abstract — As more and more systems switch from the traditional uniprocessor architectures to multicore and multiprocessor parallel architectures, a new wave of benchmarks have to be created to mimic computational behavior of real-world problems but in a more parallel fashion. Berkeley took on this challenge two years ago and just released a technical write-up of what they believe is the basis for new parallel processor architecture benchmarks, called the 13 dwarfs. These benchmarks stress everything from inter-processor communications to intense processor utilization. In this paper we implement two of these dwarfs, MapReduces and Spectral Methods, and benchmark them on current as well as past multithreaded architectures. We start out with background on the hardware we will be using for the benchmarks. We then go into some details about the benchmarks themselves, followed by our design…