MIT Libraries logoDSpace@MIT

MIT
View Item 
  • DSpace@MIT Home
  • MIT Libraries
  • MIT Theses
  • Doctoral Theses
  • View Item
  • DSpace@MIT Home
  • MIT Libraries
  • MIT Theses
  • Doctoral Theses
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Many-core architectures with time predictable execution Support for hard real-time applications

Author(s)
Kinsy, Michel A
Thumbnail
DownloadFull printable version (29.34Mb)
Other Contributors
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.
Advisor
Srinivas Devadas.
Terms of use
M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission. http://dspace.mit.edu/handle/1721.1/7582
Metadata
Show full item record
Abstract
Hybrid control systems are a growing domain of application. They are pervasive and their complexity is increasing rapidly. Distributed control systems for future "Intelligent Grid" and renewable energy generation systems are demanding high-performance, hard real-time computation, and more programmability. General-purpose computer systems are primarily designed to process data and not to interact with physical processes as required by these systems. Generic general-purpose architectures even with the use of real-time operating systems fail to meet the hard realtime constraints of hybrid system dynamics. ASIC, FPGA, or traditional embedded design approaches to these systems often result in expensive, complicated systems that are hard to program, reuse, or maintain. In this thesis, we propose a domain-specific architecture template targeting hybrid control system applications. Using power electronics control applications, we present new modeling techniques, synthesis methodologies, and a parameterizable computer architecture for these large distributed control systems. We propose a new system modeling approach, called Adaptive Hybrid Automaton, based on previous work in control system theory, that uses a mixed-model abstractions and lends itself well to digital processing. We develop a domain-specific architecture based on this modeling that uses heterogeneous processing units and predictable execution, called MARTHA. We develop a hard real-time aware router architecture to enable deterministic on-chip interconnect network communication. We present several algorithms for scheduling task-based applications onto these types of heterogeneous architectures. We create Heracles, an open-source, functional, parameterized, synthesizable many-core system design toolkit, that can be used to explore future multi/many-core processors with different topologies, routing schemes, processing elements or cores, and memory system organizations. Using the Heracles design tool we build a prototype of the proposed architecture using a state-of-the-art FPGA-based platform, and deploy and test it in actual physical power electronics systems. We develop and release an open-source, small representative set of power electronics system applications that can be used for hard real-time application benchmarking.
Description
Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2013.
 
Cataloged from PDF version of thesis.
 
Includes bibliographical references (p. 183-193).
 
Date issued
2013
URI
http://hdl.handle.net/1721.1/82351
Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Publisher
Massachusetts Institute of Technology
Keywords
Electrical Engineering and Computer Science.

Collections
  • Doctoral Theses

Browse

All of DSpaceCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

My Account

Login

Statistics

OA StatisticsStatistics by CountryStatistics by Department
MIT Libraries
PrivacyPermissionsAccessibilityContact us
MIT
Content created by the MIT Libraries, CC BY-NC unless otherwise noted. Notify us about copyright concerns.