P. R. Morin, \Two Topics in Applied Algorithmics," M.S. Thesis, School of CS., Carleton University, CA, 1998, Available through URL http://www.scs.carleton.ca/ morin. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
A General-Purpose Model for Heterogeneous Computation - Williams (2000) (Correct)
....size ranges from 10 4 to 10 5 integers. Each data point represents the average of 10 runs on a cluster comprised of 2, 4, 6, 8, and 10 heterogeneous processors. the p Gamma 1 splitters requires gp(1 r s ) 2L time. Since each processor is expected to receive approximately c j n keys [Mor98b], Step 6 uses O(c j n) computation time and g Delta maxfr j c j ng communication time, where 0 j p. Once each processor receives their keys, sorting them requires O(c j n lg c j n) time. Thus, the total time of the algorithm is O(c s n lg c s n) g(n p p pr s r s c s n) 5L. The ....
Pat Morin. "Two Topics in Applied Algorithmics.". Master's thesis, Carleton University, 1998.
The Heterogeneous Bulk Synchronous Parallel Model - Williams, Parsons (2000) (1 citation) (Correct)
....performance on a wide range of parallel machines. HBSP enhances the applicability of the BSP model by incorporating parameters that reflect the relative speeds of the heterogeneous computing components. Our starting point for the development of algorithms for HBSP are efficient BSP or HCGM [10, 11] applications. Specifically, we develop three HBSP algorithms prefix sums, matrix multiplication, and randomized sample sort that distribute the computational load according to processor speed without sacrificing performance. In fact, the cost model indicates that wall clock performance is ....
....languages runtime systems without a solid theoretical foundation. For an overview of these approaches, we refer the reader to the surveys by Siegel et al. 12] and Weems et al. 20] One notable exception is the the Heterogeneous Coarse Grained Multicomputer (HCGM) model, developed by Morin [10, 11]. HBSP and HCGM are similar in structure and philosophy. The main difference is that HGCM is not intended to be an accurate predictor of execution times whereas HBSP attempts to provide the developer with predictable algorithmic performance. 3 Heterogeneous BSP The Heterogeneous Bulk Synchronous ....
[Article contains additional citation context not shown here]
P. Morin. Two topics in applied algorithmics. Master's thesis, Carleton University, 1998.
ParLeda: A Library for Parallel Processing in Computational .. - Ghodsi, Sharifzadeh (Correct)
....of the interfaces between layers can result in an efficient design of the solution. 4 Computational Model ParLeda is based on a heterogeneous parallel computational model, named HBSP which is an extension to BSP model. The idea of design and using such model has been originated from [3] that used a similar model (HCGM) based on CGM. HBSP uses heterogeneous processors with different speeds. The speed of processors are considered as model parameters. This speed is a function of all software and hardware parameters like its virtual memory and processor speed involved in processor s ....
P. R. Morin, "Two Topics in Applied Algorithmics", M.S. Thesis, School of CS., Carleton University, CA, 1998, Available through URL http://www.scs.carleton.ca/ morin.
Mesh Simplification in Parallel - Langis Roth And (1999) (Correct)
....[7.4] 4 8 1.4 [0.8] 66.0 [7.6] 31.0 [11.1] 107.6 [6.2] 5 16 2.6 [2.7] 39.0 [2.6] 25.4 [2.5] 80.8 [5.2] 6 1 3813 7 2 5.4 [1.2] 1878 [310] 560 [362] 2079 [308] 8 23K 4 6.0 [2.4] 949 [38] 328 [71] 1141 [39] 9 8 6.8 [3. 6] 462 [22] 178 [36] 657 [28] 10 16 11.2 [12.4] 233 [13] 108 [17] 425 [12] 11 1 34811 12 2 52.8 [28.4] 15617 [452] 4057 [161] 16577 [429] 13 100.3K 4 68.5 [56.5] 7158 [70] 2086 [126] 8211 [174] 14 8 82.8 [1.8] 3449 [68] 1260 [127] 4427 [96] 15 16 162.8 [3.1] 1776 [75] 709 [99] 2874 [130] 16 1 17 2 18 198.3K 4 103 [12] 29640 [1367] 9483 [2768] ....
....[12.4] 233 [13] 108 [17] 425 [12] 11 1 34811 12 2 52.8 [28.4] 15617 [452] 4057 [161] 16577 [429] 13 100.3K 4 68.5 [56.5] 7158 [70] 2086 [126] 8211 [174] 14 8 82.8 [1.8] 3449 [68] 1260 [127] 4427 [96] 15 16 162.8 [3. 1] 1776 [75] 709 [99] 2874 [130] 16 1 17 2 18 198.3K 4 103 [12] 29640 [1367] 9483 [2768] 32103 [1638] 19 8 184 [20] 11338 [58] 3749 [307] 13714 [59] 20 16 369 [73] 5620 [13] 1873 [131] 8315 [353] Table 2: Parallel Simplification Statistics: Stanford Dragon. In summary, we conclude that the total time (T) observed for our method shows close to optimal ....
Patrick R. Morin. Two Topics in Applied Algorithmics. MCS thesis, Carleton University, 1998.
Outline of Proposed Research Activities - The Proposed Research (1998) Self-citation (Morin) (Correct)
No context found.
Pat Morin. Two topics in applied algorithmics. Master's thesis, Carleton University, 1998. [MCS]
ParLeda: A Library for Parallel Processing in Computational .. - Ghodsi, Sharifzadeh (Correct)
No context found.
P. R. Morin, \Two Topics in Applied Algorithmics," M.S. Thesis, School of CS., Carleton University, CA, 1998, Available through URL http://www.scs.carleton.ca/ morin.
Outline of Proposed Research Activities - The Proposed Research (1998) (Correct)
No context found.
Pat Morin. Two topics in applied algorithmics. Master's thesis, Carleton University, 1998. [MCS]
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC