-
Notifications
You must be signed in to change notification settings - Fork 0
/
rt-nsf-2016.tex
577 lines (409 loc) · 29.8 KB
/
rt-nsf-2016.tex
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
%%%%%%%%% MASTER -- compiles the 4 sections
\documentclass[10pt,letterpaper]{article}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\pagestyle{plain} %%
%%%%%%%%%% EXACT 1in MARGINS %%%%%%% %%
\setlength{\textwidth}{6.5in} %% %%
\setlength{\oddsidemargin}{0in} %% (It is recommended that you %%
\setlength{\evensidemargin}{0in} %% not change these parameters, %%
\setlength{\textheight}{8.67in} %% at the risk of having your %%
\setlength{\topmargin}{0in} %% proposal dismissed on the basis %%
\setlength{\headheight}{-40pt} %% of incorrect formatting!!!) %%
\setlength{\headsep}{65pt} %% %%
\setlength{\footskip}{0.3in} %% %%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%
\newcommand{\required}[1]{\section*{\hfil #1\hfil}} %%
\renewcommand{\refname}{\hfil References Cited\hfil} %%
\bibliographystyle{plain} %%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%PUT YOUR MACROS HERE
\usepackage{hyphenat}
\usepackage{layout}
\usepackage{verbatim}
\usepackage{cite}
\usepackage[pdftex]{graphicx}
\usepackage{setspace}
\usepackage[cmex10]{amsmath}
\usepackage{amsmath, amsthm, amssymb}
\usepackage{algorithmic}
\usepackage{algorithm}
\usepackage{subfig}
\usepackage{caption}
\usepackage{tabularx}
%\usepackage[caption=false]{subfig}
%\usepackage{subcaption}
%\captionsetup{compatibility=false}
\usepackage{mathpazo}
\linespread{1.05} % Palatino needs more leading (space between lines)
\usepackage{euler}
\usepackage{url}
%\usepackage{courier}
%\renewcommand{\familydefault}{\sfdefault}
%\usepackage[compact]{titlesec}
%\addtolength{\itemsep}{-0.05in}
%\usepackage{subcaption}
\usepackage{paralist}
\usepackage{tikz}
\usepackage{pgfplots}
% options for pgfplots
\pgfplotsset{compat=1.8,compat/show suggested version=false}
\usetikzlibrary{calc,trees,arrows,patterns,plotmarks,shapes,snakes,er,3d,automata,backgrounds,topaths,decorations.pathmorphing,decorations.markings}
%\pgfplotsset{compat=newest}
\pgfplotsset{
/pgfplots/bar cycle list/.style={/pgfplots/cycle list={%
{black,fill=black!30!white,mark=none},%
{black,fill=red!30!white,mark=none},%
{black,fill=green!30!white,mark=none},%
{black,fill=yellow!30!white,mark=none},%
{black,fill=brown!30!white,mark=none},%
}
},
}
\usetikzlibrary{external}
\tikzexternalize[prefix=out/]
\tikzexternalize
% don't externalize todonotes
%\makeatletter
%\renewcommand{\todo}[2][]{\tikzexternaldisable\@todo[#1]{#2}\tikzexternalenable}
%\makeatother
% end of externalization
\usetikzlibrary{patterns}
\usepgfplotslibrary{groupplots}
\pgfplotsset{
every axis label/.append style={font=\footnotesize},
tick label style={font=\footnotesize},
}
%\usepackage[normalem]{ulem}
%\setlength{\abovecaptionskip}{2pt plus 2pt minus 2pt}
%\setlist{noitemsep,topsep=0pt}
\usepackage{wrapfig}
\makeatletter
\renewcommand{\paragraph}{%
\@startsection{paragraph}{4}%
{\z@}{1.25ex \@plus 1ex \@minus .2ex}{-1em}%
{\normalfont\normalsize\bfseries}%
}
\makeatother
%\usepackage[utf8]{inputenc}
%---------------------------------------------------------------------------------%
%*********************************************************************************%
%---------------------------------------------------------------------------------%
\title{\textbf{CSR:Small:Collaborative Research:Exploring Timing-Energy Tradeoffs on Heterogeneous Computing Platforms}}
\def\changemargin#1#2{\list{}{\rightmargin#2\leftmargin#1}\item[]}
\let\endchangemargin=\endlist
\hyphenation{op-tical net-works semi-conduc-tor}
%\pagestyle{fancy}
%\includeonly{NSFsumm}
\date{\vspace{4mm}}
%\doublespace
\begin{document}
%\layout
%\maketitle
%\pagenumbering{arabic}
%---------------------------------------------------------------------------------%
%*********************************************************************************%
%---------------------------------------------------------------------------------%
\thispagestyle{empty}
\begin{center}
\Large{\textbf{CSR:Small:Collaborative Research:Exploring Timing-Energy Tradeoffs on Heterogeneous Computing Platforms}}
\end{center}
\paragraph{Project Summary.}
The physical constraints of thermal and power dissipation have led to
the emergence of \emph{dark silicon} in embedded processor designs.
Dark silicon refers to the fact that not all transistors can be
powered at full speed all the time --- doing so would destroy the chip
--- so some set of transistors must be kept dark; i.e., unpowered. To
address this issue, embedded processor designers have turned to
heterogeneous processor designs which combine powerful, but energy
inefficient cores, with slow, but energy efficient ones.
The combination of dark silicon and heterogeneous processor designs
creates both a challenge and an opportunity for embedded systems
designers and embedded operating systems in particular. Embedded
operating systems must support hard real-time requirements and
minimize energy consumption. The challenge of heterogeneous designs
is that real-time schedulers must now incorporate knowledge of
differing core types and reason about the diverse design space of
scheduling options that arise from multiple core types on the same
chip. The opportunity is that dark silicon effectively creates a
``reserve'' of additional computing power that can be brought to bear
in short bursts, but is not sustainable without damaging the hardware.
\emph{This proposal's goal is to create new real-time scheduling
algorithms for embedded systems that both address the challenge and
exploit the opportunity of heterogeneous processor designs and dark
silicon.}
\paragraph{Intellectual Merit.}
We seek to develop new real-time scheduling algorithms for
heterogeneous multicores that (1) are more energy efficient than the
current state-of-the-art and (2) can provide a new capability:
guaranteeing both timing and energy consumption for embedded
workloads. While prior work has addressed real-time support on
heterogeneous architectures, it has primarily focused on porting
existing techniques developed for homogeneous systems and ensuring
that they can still meet hard real-time guarantees. When ported to
heterogeneous systems, existing techniques are more energy efficient,
but this gain is entirely due to the processor architecture and not an
inherent feature of the algorithm. The key insight in our proposed
work will be to start from scratch and incorporate knowledge of
heterogeneity and energy into our proposed algorithms from the
beginning of the project. This approach promises to lead to new
algorithms that have better energy efficiency and offer new
capabilities.
We plan to evaluate these algorithms both analytically and
empirically. Our empirical test case will be automotive computing, a
rich embedded processing domain that includes safety critical tasks,
like collision detection, and best effort tasks, like infotainment. We
will test the energy savings and schedulability of our algorithms by
implementing them for an state-of-the-art ARM big.LITTLE heterogeneous
embedded multicore and using them to schedule a representative
automotive workload. We will demonstrate the proposed algorithms'
energy efficiency and ability to increase schedulability for the
automotive workload on real heterogeneous processors.
\paragraph{Broader Impacts.}
The proposed research agenda provides for significant broader impacts
related to (1) curriculum development and student training through
integration of modules in existing real-time, energy-aware, and
embedded system design courses, and (2) outreach through established
enrichment programs at the participating universities. The proposed
research benefits society by directly addressing the needs of smart
cyberphysical systems such as self-driving cars, embedded medical
devices, and intelligent robotics. Such systems play critical roles
in manufacturing, national security, and healthcare. The proposed
research benefits cyberphysical systems by (1) enabling greater
functionality and new capabilities for the same cost and (2) greatly
reducing system energy consumption, perhaps leading to medical devices
that last longer or systems that need to be replaced less often.
\paragraph{Keywords:}
Real-time; operating systems; embedded; energy; heterogeneous multicore; scheduling
%---------------------------------------------------------------------------------%
%*********************************************************************************%
%---------------------------------------------------------------------------------%
\newpage
\pagenumbering{arabic}
\setcounter{page}{1}
\section{Introduction}
\label{sec:intro}
\input{introduction.tex}
%*********************************************************************************%
%*********************************************************************************%
\section{Background and Related Work}
\label{sec:motivation}
\input{background.tex}
%*********************************************************************************%
\subsection{Hardware Platform}
\label{sec:hardware}
\input{hardware.tex}
%*********************************************************************************%
\subsection{Real-Time Scheduling Algorithms and Schedulability Tests}
\label{sec:model}
\input{model.tex}
%*********************************************************************************%
\subsection{Related Work}
\label{sec:relatedwork}
\input{relatedwork.tex}
%*********************************************************************************%
%*********************************************************************************%
\section{Detailed Research Plan}
\label{sec:researchplan}
\input{researchplan.tex}
%*********************************************************************************%
\subsection{Step 1: Identifying Energy-Efficient Configurations that Guarantee Timing}
\label{sec:step1}
\input{step1.tex}
%*********************************************************************************%
\subsection{Step 2: Energy Optimization and Guarantees}
\label{sec:step2}
\input{step2.tex}
%*********************************************************************************%
\subsection{Step 3: Implementation and Overhead-Conscious Schedulability Studies}
\label{sec:step3}
\input{step3.tex}
%*********************************************************************************%
\subsection{Step 4: Conduct Case Studies}
\label{sec:step4}
\input{step4.tex}
%*********************************************************************************%
\subsection{Timeline}
\label{sec:timeline}
\input{timeline.tex}
%*********************************************************************************%
%*********************************************************************************%
\section{Broader Impact of the Proposed Work}
\label{sec:impact}
\input{impact.tex}
%*********************************************************************************%
%*********************************************************************************%
\section{Results From Previous and Current NSF Support}
\label{sec:prior}
\input{prior.tex}
%*********************************************************************************%
%*********************************************************************************%
\newpage
\pagenumbering{gobble}
\bibliographystyle{plain}
\bibliography{b1,b2,b3,b4,hank}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\newpage
\pagenumbering{gobble}
\thispagestyle{empty}
\required{Cong Liu}
\vspace{-3mm}
\begin{center}
Department of Computer Science, University of Texas at Dallas, Richardson, TX 75081
http://www.utdallas.edu/$\sim$cong/
\end{center}
\noindent \textbf{Professional Preparation}\
\begin{tabular}{l l l}
Wuhan Univ. of Technology & Computer Science & B.S. with honor, 2005 \\
Auburn Univ. & Computer Science & M.S., 2008 \\
Univ. of North Carolina at Chapel Hill & Computer Science & Ph.D., 2013\\ \\
\end{tabular}
\noindent \textbf{Appointments}\
\hspace{1mm} Assistant Professor, Department of Computer Science, University of Texas at Dallas, 9/13-present. \\
\noindent \textbf{Products}
\hspace{-4mm} \textbf{(i) Most Relevant Publications}\
\vspace{-1mm}
\begin{itemize}
\item
Husheng Zhou, Guangmo Tong, and Cong Liu, ``\textit{GPES: A Preemptive Execution System for GPGPU Computing}'', Proceedings of the 21st IEEE Real-Time and Embedded Technology and Applications Symposium (\underline{RTAS}), 2015.
\item
Guangmo Tong and Cong Liu, ``\textit{Supporting Soft Real-Time Sporadic Task Systems on Heterogeneous Multiprocessors with No Uilitzation Loss}'', IEEE Transactions on Parallel and Distributed Systems (\underline{TPDS}), 2015.
\item
Husheng Zhou and Cong Liu, ``\textit{Task Mapping in Heterogeneous Embedded Systems for Fast Completion Time}'', Proceedings of the 14th ACM International Conference on Embedded Software (\underline{EMSOFT}), 2014.
\item
Cong Liu and Jianjia Chen, ``\textit{Bursty-Interference Analysis Techniques for Analyzing Complex Real-Time Task Models}'', Proceedings of the 35th IEEE Real-Time Systems Symposium (\underline{RTSS}), 2014.
\item Cong Liu, Jian Li, Wei Huang, Juan Rubio, and Evan Speight, ``\textit{Power-Efficient Time-Sensitive Mapping in CPU/GPU Heterogeneous Systems},'' Proceedings of the 21st International Conference on Parallel Architectures and Compilation Techniques (\underline{PACT}), pp. 23-32, 2012.
\end{itemize}
\hspace{-4mm} \textbf{(ii) Additional Products}\
\vspace{-1mm}
\begin{itemize}
%\vspace{-2mm}
%\item Liang He, Linghe Kong, Siyu Lin, Shaodong Ying, Yu Gu, Tian he, and Cong Liu, ``\textit{Reconfiguration-Assisted Charging in Large-Scale Lithium-ion Battery Systems},'' Proceedings of the 5th ACM/IEEE International Conference on Cyber-Physical Systems \sloppy(\underline{ICCPS}), 2014.
\item
Jianjia Chen, Wenhung Huang, and Cong Liu, ``\textit{K2U: A General Framework from k-Point Effective Schedulabiliy Analysis to Utilization-based Tests}'', Proceedings of the 36th IEEE Real-Time Systems Symposium (\underline{RTSS}), 2015
\item
Jianjia Chen and Cong Liu, ``\textit{Fixed-Relative-Deadline Scheduling of Hard Real-Time Tasks with Self-Suspen- sions}'', Proceedings of the 35th IEEE Real-Time Systems Symposium (\underline{RTSS}), 2014.
%\item Guangmo Tong and Cong Liu, ``\textit{Supporting Read/Write Applications in Embedded Real-time Systems via Suspension-aware Analysis}'', Proceedings of the 14th ACM International Conference on Embedded Software (\underline{EMSOFT}), 2014.
\item Cong Liu and James Anderson, ``An O(m) Analysis Technique for Supporting Real-Time Self-Suspending Task Systems,'' Proceedings of the 33th IEEE Real-Time Systems Symposium (\underline{RTSS}), pp. 373-382, 2012.
\item Cong Liu and James Anderson, ``\textit{Supporting Soft Real-Time DAG-based Systems on Multiprocessors with No Utilization Loss}," Proceedings of the 31st IEEE Real-Time Systems Symposium (\underline{RTSS}), pp. 3-13, 2010.
\item Cong Liu and James Anderson, ``Task Scheduling with Self-Suspensions in Soft Real-Time Multiprocessor Systems,'' Proceedings of the 30th IEEE Real-Time Systems Symposium (\underline{RTSS}), pp.425-436, 2009, \textbf{Best Student Paper Award.}
\end{itemize}
\newpage
\noindent \textbf{Synergistic Activities}\
\begin{changemargin}{4mm}{2mm}
\textbf{Service to the Scientific and Engineering Community}\
\vspace{-1mm}
\begin{changemargin}{2mm}{2mm}
- TPC, IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS), 2016. \\
- TPC, Euromicro Conference on Real-Time Systems (ECRTS), 2016. \\
- TPC, IEEE Real-Time Systems Symposium (RTSS), 2015. \\
- TPC, IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS), 2015. \\
- TPC, IEEE Real-Time Systems Symposium (RTSS), 2014. \\
- TPC, IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS), 2014.
% Technical Program Committee Member, International Conference on High Performance Compilation, Computing and Communications, 2014.
%\textit{Reviewer}: IEEE Transactions on Computers, IEEE Transactions on Parallel and Distributed Systems; IEEE Transactions on Industrial Informatics; ACM Transactions on Embedded Computing Systems; Real-Time Systems; Journal of Systems Architecture; Journal of Parallel and Distributed Computing; IEEE Real-TIme and Embedded Technology and Applications Symposium; IEEE Real-Time Systems Symposium; IEEE Real-TIme and Embedded Computing Systems and Applications; Euromicro Conference on Real-Time Systems; IEEE International Conference on Computer Communications;
\end{changemargin}
\end{changemargin}
\noindent \textbf{Collaborators and Other Affiliations}\\
\vspace{-2mm}
\hspace{-2mm} \textbf{(i) Collaborators}\
\begin{changemargin}{8mm}{2mm}
\vspace{-2mm}
Dr. Evan Speight, IBM Research Austin Lab; \\
Dr. Jian Li IBM Research Austin Lab;\\
Dr. Tian He, University of Minnesota, Twin City; \\
Dr. Jianjia Chen, TU Dortmund (German);\\
Dr. Yu Gu, Singapore University of Technology and Design (Singapore);\\
Dr. Wei Gao, University of Tennessee, Knoxville; \\
Dr. Shinpei Kato, Nagoya University (Japan);\\
Total number of collaborators: 7
\end{changemargin}
\hspace{-4mm} \textbf{(ii) Graduate and Post-doctoral Advisors}\
\begin{changemargin}{8mm}{2mm}
Dr. James Anderson, Professor, Department of Computer Science, University of North Carolina at Chapel Hill.\\
Total number of graduate and post-doctoral advisors: 1
\end{changemargin}
\hspace{-4mm} \textbf{(iii) Thesis Advisor and Post-graduate Scholar Sponsor}\
\begin{changemargin}{8mm}{2mm}
Mr. Guangmo Tong, PhD Student, Computer Science, UTD (since 2014)
Mr. Husheng Zhou, PhD Student, Computer Science, UTD (since 2014)
Ms. Xia Zhang, PhD Student, Computer Science, UTD (since 2014)
Mr. Soroush Bateni, PhD student, Computer Science, UTD (since 2015)
Mr. Gbadebo Ayoade, Masters Student, Computer Science, UTD (Since 2014)
Total number of students: 5
%Mr. Gbadebo Ayoade, Masters Student, Computer Science, UTD (since 2013)
\end{changemargin}
%---------------------------------------------------------------------------------%
%---------------------------------------------------------------------------------%
\newpage
%\input{NSFbio-hank}
\pagenumbering{gobble}
\newpage
\pagenumbering{gobble}
\begin{center}
\Large{\textbf{\\Facilities, Equipment, and Other Resources}}
\end{center}
\paragraph{General computing environment.} The Department of Computer Science at UT Dallas hosts more than 40 research labs with access 24 hours a day, 7 days a week; 2 open labs for graduate students; and a 128-seat open access lab for undergraduate students. These labs consist of over 500 state-of-the-art, high-performance workstations and high-end PCs, all connected via gigabit Ethernet switches with a redundant fiber up-link to provide fast access to the campus network and the Internet. Nine classrooms and one large lecture hall with the latest computer and audio-visual equipment are available. Academic coursework, project, and computing systems are comprised of Linux x86\_64, Windows Server, and Solaris 10 executing on a collection of physical servers and virtual server private clouds. The servers are connected via fiber channel and iSCSI to a thin-provisioned 14TB 3PAR mesh-active storage array.
\paragraph{Specific computing facilities.} Computer Science Open Access Lab: This lab is accessible to all students currently enrolled in any course in the department of computer science (128 seats). The lab consists of 128 Dell OptiPlex 980 machines with i7-870 Processor (8M Cache, 2.93 GHz), 4GB RAM, 22'' (Dell P2210) flat-panel display, Windows 7, 64-bit; Two HP Laserjet 9040 printers are also located in the lab. The lab is open from 8am to 11pm, M-F, and 11pm-7pm Sat/Sun. The Lab is maintained by the CS Tech Staff. A monitor (typically a graduate student hired half-time) is on premises all the time.
Graduate Windows Lab: This 15 seat lab is accessible to all graduate students. Equipment consists of Dell Precision T1650 computers with Intel XEON CPU E3 1225 v2, @3.20GHz, 8GB RAM, 24'' flat-panel monitors, Windows 7 Enterprise and an HP Laserjet P4015n printer. The lab is accessible 24/7 with smart card access.
Project Design Lab: This 20 seat lab is accessible to all CS undergraduate/graduate students who take senior design project class. It consists of Dell Optiplex 760 machines with Intel Core2 Duo, [email protected], 2GB RAM, 17''flat-panel monitors, Windows 7 Enterprise. The lab is accessible 24/7 with smart card access.
General Access Systems: The Department also maintains several systems in its server room to which remote access is provided. These include the Network Programming System that are ccessible to all CS students enrolled in network programming and parallel processing courses (45 systems); Students access these systems via SSH; All these 45 systems run Linux CentOS 6.2 x86\_64. The Department also maintains two large Linux compute servers (Linux CentOS 6.2 x86\_64, Dell PowerEdge R710 w/2 six-core 2.53GHz Intel Xeon, 32GB RAM) and two large Solaris computer servers (Sun Fire V440 w/4x1GHz UltraSPARC-IIIi, 8GB RAM).
Server clouds: The Department also maintains a general purpose server cloud, a VMware vSphere 4 operating on 5 ESX hosts with dual 6-core processor 128GB RAM servers, used for teaching the cloud computing related courses as well as a network security courses server cloud, a VMware vSphere 4 executing on 4 processors of 6-core Opterons, 128GB RAM, and four 4Gb redundant fiber channel SAN connections.
In addition to the above labs and facilities, CS students also have access the UTDesign Studio that is equipped with modern computing facilities and equipment with more than 30,000 square feet of dedicated space, where students and corporate partners can create, innovate, design, build and learn. Further information can be found at $http://www.utdallas.edu/utdesign/corporate/studio/$. The UTDesign studio is primarily used by undergraduate seniors taking the capstone project course.
A 30' x 20' lab is available for Dr. Liu's students. The lab can comfortably accommodate 8 students and the computing equipment. Eight workstations equipped with state-of-the-art NVIDIA GPUs currently in the PI's lab will be available to the project.
%--------------------------------------------------------------------------------------------%
%xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx%
%--------------------------------------------------------------------------------------------%
\newpage
\pagenumbering{gobble}
\begin{center}
\Large{\textbf{\\Data Management Plan}}
\end{center}
This data management plan covers all of the data products to be produced in this project. It includes all source code that implements the energy-efficient real-time resource allocation algorithms, any source code added to LITMUS$^{\textrm{RT}}$, the suite of case-study programs, and scripts used to test and evaluate the implementation of the proposed components, along with the data collected during evaluation efforts.
\vspace{1mm}
\noindent \textbf{Source code.} All source code will be freely available for direct download from public web servers maintained by the University of Texas at Dallas Computer Science Department under an open source GPL license. Additional case-study test programs and scripts will be freely available under the open source BSD license and directly downloadable from public web servers maintained by the UTD Computer Science Department. Our intend in making these data products freely available is that other researchers can verify evaluation results, and relevant research conducted by other interested groups can be facilitated.
\vspace{1mm}
\noindent \textbf{Evaluation data.} Evaluation data (including raw data and resulting graphs) can be several terabytes in size. It is thus currently not feasible to make this data available to the public through direct download. We will store and back up such data on secured servers maintained by the University of Texas at Dallas Computer Science Department. Interested parties may request copies of this data through making arrangement for private direct download or using the postal system.
\vspace{1mm}
\noindent \textbf{Data Format and Metadata Standards: } We intend to share the measurement/trace data using comma-separated values (CSV) format, which is a set of file formats used to store tabular data in which numbers and text are stored in a plain-text form that can be easily written and read in a text editor. It is very easy for others to convert this format into more advanced data format at user's discretion. We will provide readme files to explain the data format and contextual details (metadata and instructions) for understanding and processing these data files. Source code and simulation programs will be stored in their respective native
formats, for example ��.m�� for Matlab code, ��.c/cpp�� for C/C++ code. Simulation and experimental results
will be stored in ��.dat�� or ��.fig�� formats. Project reports, conference, and journal papers will be in either
Microsoft Word (��.doc�� or ��.docx��) format, or Latex format.
\vspace{1mm}
\noindent \textbf{Data Sharing, Distribution and Preservation Plan: }
We plan to make publicly available the electronic raw trace data produced under this project as well as the associated metadata that describes the
experimental setup, modeling methodology and data analysis methods. We also plan to make publicly
available the analyzed data in the forms of peer-reviewed conference, journal articles, thesis, and book
chapters. The electronic data will be preserved in multiple on-site RAID hard drive storages at the
University of Texas at Dallas and and University of Minnesota.
The data will be shared through the web server provided by the University of Texas at Dallas or through direct
contact with the PIs. Data will be available for access and sharing as soon as the proposed research has
been accomplished, or no more than two years after the acquisition of the data.
We will also share our tools and publications with the CPS research community through the CPS Virtual Organization (VO). The data will be preserved for at least three years beyond the award period, as required by NSF guidelines. Since the data format we
choose is CVS, which can be easily edited and converted, we expect no problem to archive and transform
our electronic trace data in the foreseeable future after the project has been completed. All data products produced in this project will also be archived in the UTD computer science computing repository for permanent storage.
\vspace{1mm}
\noindent \textbf{Privacy.} No personal data will be acquired in this project. There are no privacy concerns for the storage of data.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\newpage
\begin{center}
\Large\textbf{Collaboration Plan}
\end{center} \normalsize
\vspace{6mm}
\large \noindent\textbf{1 \ \ Regular Communication and Reporting}\normalsize
\vspace{4mm}
\noindent The proposed research activities will be carried out at two different universities: The University of Texas at Dallas and University of Chicago. The entire team will participate in a bi-weekly on-line meeting through free online teleconference services. In addition, as one of the budget line items, one domestic trip per year is arranged for bringing Dr. Hoffman and any of his supported students in this project to the University of Texas at Dallas together. A written progress report on all supported activities will be produced on a quarterly basis.
\vspace{6mm}
\large \noindent\textbf{2 \ \ Team Formation and Specific Roles}\normalsize
\vspace{4mm}
\noindent Our team consists of two experts in the general fields of real-time systems, operating systems, heterogeneous processor design, and power optimization. PI Liu has rich experience in operating systems support for heterogeneous real-time systems~\cite{Liu12, GPES, zhou2015supporting, Zhou2014a} and real-time resource allocation on heterogeneous computing platforms~\cite{chen2015k2u, Tong14a, LiuRTSS14a, Liu1, Liu2, Liu6, Liu7, Liu10, liu2014supporting, LiuRTSS14b, elliott1minimizing, Liu3, Liu4, Liu5, Liu9, Liu11, Liu13}.
PI Hoffmann has rich experience in homogeneous\cite{raw1,raw2,raw3,tilera1,tilera2} and heterogeneneous \cite{ASAP,HPEC,ASAP2,ISSoC} processor design as well as operating systems support for managing performance and power tradeoffs \cite{LEO,POET,DynamicKnobs,JouleGuard,PTRADE,PCP,TCST,HotPower,kim-cpsna}.
In this project, the University of Texas at Dallas will assume a management and coordination role. The University of Texas at Dallas will be responsible for maintaining a central project repository where tools will be integrated and data will be collected. These tools and data will be made available to the community. The specific roles of the project participants in all organizations involved are described as follows.
\begin{itemize}
\item Dr. Liu will be responsible for designing, analyzing, and evaluating energy-efficient configurations for heterogeneous platforms that simultaneously guarantee timing correctness.
\item Dr. Hoffmann will lead the research on the design and analysis of the techniques to aggressively address energy optimization and guarantees.
\item The two PIs intend to work closely together on the implementation and evaluation of the proposed approaches.
\end{itemize}
%--------------------------------------------------------------------------------------------%
%xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx%
%--------------------------------------------------------------------------------------------%
\end{document}