1 /*! \page publis Publications
3 \section pub_reference Reference publication about SimGrid
5 When citing SimGrid, the prefered reference paper is <i>Scheduling
6 Distributed Applications: the SimGrid Simulation Framework</i>, even if it's
7 a bit old now. We are actively working on improving this.
9 \li <b>Scheduling Distributed Applications: the
10 SimGrid Simulation Framework</b>\n
11 by <em>Henri Casanova and Arnaud Legrand and Loris Marchal</em>\n
12 Proceedings of the third IEEE International Symposium
13 on Cluster Computing and the Grid (CCGrid'03)\n
14 Since the advent of distributed computer systems an active field
15 of research has been the investigation of scheduling strategies
16 for parallel applications. The common approach is to employ
17 scheduling heuristics that approximate an optimal
18 schedule. Unfortunately, it is often impossible to obtain
19 analytical results to compare the efficacy of these heuristics.
20 One possibility is to conducts large numbers of back-to-back
21 experiments on real platforms. While this is possible on
22 tightly-coupled platforms, it is infeasible on modern distributed
23 platforms (i.e. Grids) as it is labor-intensive and does not
24 enable repeatable results. The solution is to resort to
25 simulations. Simulations not only enables repeatable results but
26 also make it possible to explore wide ranges of platform and
27 application scenarios.\n
28 In this paper we present the SimGrid framework which enables the
29 simulation of distributed applications in distributed computing
30 environments for the specific purpose of developing and evaluating
31 scheduling algorithms. This paper focuses on SimGrid v2, which
32 greatly improves on the first version of the software with more
33 realistic network models and topologies. SimGrid v2 also enables
34 the simulation of distributed scheduling agents, which has become
35 critical for current scheduling research in large-scale platforms.
36 After describing and validating these features, we present a case
37 study by which we demonstrate the usefulness of SimGrid for
38 conducting scheduling research.
40 Previous publication do not cover the GRAS part of the framework. So, if you
41 want to cite GRAS, please use this publication instead:
43 \li <b>Gras: A Research & Development Framework for Grid and P2P
45 by <em>Martin Quinson</em>\n
46 To appear: The 18th IASTED International Conference on
47 Parallel and Distributed Computing and Systems (PDCS 2006)\n
48 http://www.loria.fr/~quinson/articles/gras-iasted06.pdf
50 \section pub_simulation Other publications about the SimGrid framework
52 \li <b>The SimGrid Project - Simulation and Deployment of Distributed Applications</b>\n
53 by <em>A. Legrand, M. Quinson, K. Fujiwara, H. Casanova</em>\n
54 <b>POSTER</b> in Proceedings of the IEEE International Symposium on High Performance Distributed Computing (HPDC-15), Paris, France, May 2006.\n
56 <a href="http://navet.ics.hawaii.edu/~casanova/homepage/papers/simgrid_hpdc06.pdf"><img src="poster_thumbnail.png" /></a>
58 http://navet.ics.hawaii.edu/~casanova/homepage/papers/simgrid_hpdc06.pdf
60 \li <b>A Network Model for Simulation of Grid Application</b>\n
61 by <em>Henri Casanova and Loris Marchal</em>\n
63 In this work we investigate network models that can be
64 potentially employed in the simulation of scheduling algorithms for
65 distributed computing applications. We seek to develop a model of TCP
66 communication which is both high-level and realistic. Previous research
67 works show that accurate and global modeling of wide-area networks, such
68 as the Internet, faces a number of challenging issues. However, some
69 global models of fairness and bandwidth-sharing exist, and can be link
70 withthe behavior of TCP. Using both previous results and simulation (with
71 NS), we attempt to understand the macroscopic behavior of
72 TCP communications. We then propose a global model of the network for the
73 Grid platform. We perform partial validation of this model in
74 simulation. The model leads to an algorithm for computing
75 bandwidth-sharing. This algorithm can then be implemented as part of Grid
76 application simulations. We provide such an implementation for the
77 SimGrid simulation toolkit.\n
78 ftp://ftp.ens-lyon.fr/pub/LIP/Rapports/RR/RR2002/RR2002-40.ps.gz
81 \li <b>MetaSimGrid : Towards realistic scheduling simulation of
82 distributed applications</b>\n
83 by <em>Arnaud Legrand and Julien Lerouge</em>\n
84 Most scheduling problems are already hard on homogeneous
85 platforms, they become quite intractable in an heterogeneous
86 framework such as a metacomputing grid. In the best cases, a
87 guaranteed heuristic can be found, but most of the time, it is
88 not possible. Real experiments or simulations are often
89 involved to test or to compare heuristics. However, on a
90 distributed heterogeneous platform, such experiments are
91 technically difficult to drive, because of the genuine
92 instability of the platform. It is almost impossible to
93 guarantee that a platform which is not dedicated to the
94 experiment, will remain exactly the same between two tests,
95 thereby forbidding any meaningful comparison. Simulations are
96 then used to replace real experiments, so as to ensure the
97 reproducibility of measured data. A key issue is the
98 possibility to run the simulations against a realistic
99 environment. The main idea of trace-based simulation is to
100 record the platform parameters today, and to simulate the
101 algorithms tomorrow, against the recorded data: even though it
102 is not the current load of the platform, it is realistic,
103 because it represents a fair summary of what happened
104 previously. A good example of a trace-based simulation tool is
105 SimGrid, a toolkit providing a set of core abstractions and
106 functionalities that can be used to easily build simulators for
107 specific application domains and/or computing environment
108 topologies. Nevertheless, SimGrid lacks a number of convenient
109 features to craft simulations of a distributed application
110 where scheduling decisions are not taken by a single
111 process. Furthermore, modeling a complex platform by hand is
112 fastidious for a few hosts and is almost impossible for a real
113 grid. This report is a survey on simulation for scheduling
114 evaluation purposes and present MetaSimGrid, a simulator built
116 ftp://ftp.ens-lyon.fr/pub/LIP/Rapports/RR/RR2002/RR2002-28.ps.gz
118 \li <b>SimGrid: A Toolkit for the Simulation of Application
120 by <em>Henri Casanova</em>\n
121 Advances in hardware and software technologies have made it
122 possible to deploy parallel applications over increasingly large
123 sets of distributed resources. Consequently, the study of
124 scheduling algorithms for such applications has been an active area
125 of research. Given the nature of most scheduling problems one must
126 resort to simulation to effectively evaluate and compare their
127 efficacy over a wide range of scenarios. It has thus become
128 necessary to simulate those algorithms for increasingly complex
129 distributed, dynamic, heterogeneous environments. In this paper we
130 present SimGrid, a simulation toolkit for the study of scheduling
131 algorithms for distributed application. This paper gives the main
132 concepts and models behind SimGrid, describes its API and
133 highlights current implementation issues. We also give some
134 experimental results and describe work that builds on SimGrid's
136 http://grail.sdsc.edu/papers/simgrid_ccgrid01.ps.gz
138 \section pub_ext Papers that use SimGrid-generated results (not counting our owns)
140 This list is a selection of articles. We list only papers written by people
141 external to the development group, but we also use our tool ourselves (see
145 - <b>Hierarchical Scheduling of Independent Tasks with Shared Files</b>\n
146 by <em>H. Senger, F. Silva, W. Nascimento</em>.\n
147 Proceedings of the Sixth IEEE International Symposium on Cluster Computing and the Grid Workshop (CCGRIDW'06), 2006.\n
148 http://www.unisantos.br/mestrado/informatica/hermes/File/senger-HierarchicalScheduling-Workshop-TB120.pdf
149 - <b>Critical Path and Area Based Scheduling of Parallel Task Graphs on Heterogeneous Platforms</b>\n
150 by <em>Tchimou N'Takpé and Frédéric Suter</em>\n
151 Proceedings of the Twelfth International Conference on Parallel and Distributed Systems (ICPADS), 2006.
152 - <b>Evaluation of Knapsack-based Scheduling using the NPACI JOBLOG</b>\n
153 by <em>D. Vanderster, N. Dimopoulos, R. Parra-Hernandez and R. Sobie</em>.\n
154 20th International Symposium on High-Performance Computing in an
155 Advanced Collaborative Environment (HPCS'06)\n
156 http://doi.ieeecomputersociety.org/10.1109/HPCS.2006.23
157 - <b>Metascheduling Multiple Resource Types using the MMKP</b>\n
158 by <em>D. Vanderster, N. Dimopoulos, R. Sobie</em>\n
159 To Appear: 7th IEEE/ACM International Conference on Grid Computing,
160 Barcelona, September 28th-29th 2006
162 - <b>On Dynamic Resource Management Mechanism using Control
163 Theoretic Approach for Wide-Area Grid Computing</b>\n
164 by <em>Hiroyuki Ohsaki, Soushi Watanabe, and Makoto Imase</em>\n
165 in Proceedings of IEEE Conference on Control Applications (CCA 2005), Aug. 2005.\n
166 http://www.ispl.jp/~oosaki/papers/Ohsaki05_CCA.pdf
167 - <b>Evaluation of Meta-scheduler Architectures and Task Assignment Policies for
168 high Throughput Computing</b>\n
169 by <em>Eddy Caron, Vincent Garonne and Andrei Tsaregorodtsev</em>\n
170 Proceedings of 4th Internationnal Symposium on Parallel and
171 Distributed Computing Job Scheduling Strategies for Parallel
172 Processing (ISPDC'05), July 2005.\n
173 http://www.ens-lyon.fr/LIP/Pub/Rapports/RR/RR2005/RR2005-27.pdf
175 - <b>Deadline Scheduling with Priority for Client-Server Systems on the Grid</b>\n
176 by <em>E Caron, PK Chouhan, F Desprez</em>\n
177 in IEEE International Conference On Grid Computing. Super Computing 2004, oct 2004.
178 - <b>Efficient Scheduling Heuristics for GridRPC Systems</b>\n
179 by <em>Y. Caniou and E. Jeannot.</em>\n
180 in IEEE QoS and Dynamic System workshop (QDS) of International Conference
181 on Parallel and Distributed Systems (ICPADS), New-Port Beach California, USA,
182 pages 621-630, July 2004\n
183 http://graal.ens-lyon.fr/~ycaniou/QDS04.ps
184 - <b>Exploiting Replication and Data Reuse to Efficiently Schedule
185 Data-intensive Applications on Grids</b>\n
186 by <em> E. Santos-Neto, W. Cirne, F. Brasileiro, A. Lima.</em>\n
187 Proceedings of 10th Job Scheduling Strategies for Parallel Processing, June 2004.\n
188 http://www.lsd.ufcg.edu.br/~elizeu/articles/jsspp.v6.pdf
189 - <b>Resource Management and Knapsack Formulations on the Grid</b>\n
190 by <em>R. Parra-Hernandez, D. Vanderster and N. J. Dimopoulos</em>\n
191 Fifth IEEE/ACM International Workshop on Grid Computing (GRID'04)\n
192 http://doi.ieeecomputersociety.org/10.1109/GRID.2004.54
193 - <b>Scheduling BoT Applications in Grids using a Slave Oriented Adaptive
195 by <em>T. Ferreto, C. A. F. De Rose and C. Northfleet.</em>\n
196 Second International Symposium on Parallel and Distributed Processing
197 and Applications (ISPA), 2004, Hong Kong. Published in Lecture Notes in
198 Computer Science (LNCS), Volume 3358, by Springer-Verlag. p. 392-398.
200 - <b>Link-Contention-Aware Genetic Scheduling Using Task Duplication in Grid Environments</b>\n
201 by <em>Wensheng Yao, Xiao Xie and Jinyuan You</em>\n
202 in Grid and Cooperative Computing: Second International Workshop, GCC 2003, Shanghai, China, December 7-10, 2003 (LNCS)\n
203 http://www.chinagrid.edu.cn/chinagrid/download/GCC2003/pdf/266.pdf
204 - <b>New Dynamic Heuristics in the Client-Agent-Server Model</b>\n
205 by <em>Y. Caniou and E. Jeannot</em>\n
206 in IEEE 13th Heteregeneous Computing Workshop - HCW'03, Nice, France, April 2003.\n
207 http://graal.ens-lyon.fr/~ycaniou/HCW03.ps
208 - <b>A Hierarchical Resource Reservation Algorithm for Network Enabled Servers</b>\n
209 by <em>E. Caron, F. Desprez, F. Petit, V. Villain</em>\n
210 in the 17th International Parallel and Distributed Processing Symposium -- IPDPS'03, Nice - France, April 2003.
212 \section pub_self Our own papers that use SimGrid-generated results
214 This list is a selection of the articles we have written that used results
215 generated by SimGrid.
218 - <b>On the Harmfulness of Redundant Batch Requests</b>\n
219 by <em>H. Casanova</em>\n
220 Proceedings of the IEEE International Symposium on High Performance Distributed Computing (HPDC-15), Paris, France, May 2006.\n
221 http://navet.ics.hawaii.edu/~casanova/homepage/papers/hpdc_2006.pdf
222 - <b>An evaluation of Job Scheduling Strategies for Divisible Loads on Grid Platforms</b>\n
223 by <em>Y. Cardinale, H. Casanova</em>\n
224 in Proceedings of the High Performance Computing & Simulation Conference (HPC&S'06), Bonn, Germany, May 2006.\n
225 http://navet.ics.hawaii.edu/~casanova/homepage/papers/cardinale_2006.pdf
226 - <b>Interference-Aware Scheduling</b>\n
227 by <em>B. Kreaseck, L. Carter, H. Casanova, J. Ferrante, S. Nandy</em>\n
228 International Journal of High Performance Computing Applications (IJHPCA), to appear.\n
229 http://navet.ics.hawaii.edu/~casanova/homepage/papers/kreaseck_ijhpca_2005.pdf
231 - <b>From Heterogeneous Task Scheduling to Heterogeneous Mixed Data and Task Parallel Scheduling</b>\n
232 by <em>F. Suter, V. Boudet, F. Desprez, H. Casanova<em>\n
233 Proceedings of Europar, 230--237, (LCNS volume 3149), Pisa, Italy, August 2004.
234 - <b>On the Interference of Communication on Computation</b>\n
235 by <em>B. Kreaseck, L. Carter, H. Casanova, J. Ferrante</em>\n
236 Proceedings of the workshop on Performance Modeling, Evaluation, and Optimization of Parallel and Distributed Systems, Santa Fe, April 2004.\n
237 http://navet.ics.hawaii.edu/~casanova/homepage/papers/k_pmeo2004.pdf
239 - <b>RUMR: Robust Scheduling for Divisible Workloads</b>\n
240 by <em>Y. Yang, H. Casanova</em>\n
241 Proceedings of the 12th IEEE Symposium on High Performance and Distributed Computing (HPDC-12), Seattle, June 2003.\n
242 http://navet.ics.hawaii.edu/~casanova/homepage/papers/yang_hpdc2003.pdf
243 - <b>Resource Allocation Strategies for Guided Parameter Space Searches</b>\n
244 by <em>M. Faerman, A. Birnbaum, F. Berman, H. Casanova</em>\n
245 International Journal of High Performance Computing Applications (IJHPCA), 17(4), 383--402, 2003.\n
246 http://grail.sdsc.edu/papers/faerman_ijhpca04.pdf
248 - <b>Resource Allocation for Steerable Parallel Parameter Searches</b>\n
249 by <em>M. Faerman, A. Birnbaum, H. Casanova, F. Berman</em>\n
250 Proceedings of the Grid Computing Workshop, Baltimore, 157--169, November 2002.\n
251 http://grail.sdsc.edu/projects/vi_itr/grid02.pdf
253 - <b>Applying Scheduling and Tuning to On-line Parallel Tomography </b>\n
254 by <em>Shava Smallen, Henri Casanova, Francine Berman</em>\n
255 in Proceedings of Supercomputing 2001\n
256 http://grail.sdsc.edu/papers/tomo_journal.ps.gz
258 - <b>Heuristics for Scheduling Parameter Sweep applications in Grid environments</b>\n
259 by <em>Henri Casanova, Arnaud Legrand, Dmitrii Zagorodnov and Francine Berman</em>\n
260 in Proceedings of the 9th Heterogeneous Computing workshop (HCW'2000), pp349-363.\n
261 http://navet.ics.hawaii.edu/~casanova/homepage/papers/hcw00_pst.pdf
266 \li <b>Optimal algorithms for scheduling divisible workloads on
267 heterogeneous systems</b>\n
268 by <em>Olivier Beaumont and Arnaud Legrand and Yves Robert</em>\n
269 in Proceedings of the 17th International Parallel and Distributed Processing Symposium (IPDPS'03).\n
270 Preliminary version on ftp://ftp.ens-lyon.fr/pub/LIP/Rapports/RR/RR2002/RR2002-36.ps.gz
273 \li <b>On-line Parallel Tomography</b>\n
274 by <em>Shava Smallen</em>\n
275 Masters Thesis, UCSD, May 2001