Transcript Parallel Landscape Fish Model for South Florida Ecosystem Simulation

Parallel Landscape Fish Model for South Florida Ecosystem Simulation
Dali
1
Wang ,
Michael W.
2
Berry ,
Eric A.
1
Carr ,
2Department
of Computer Science,
University of Tennessee, Knoxville, TN 37996
[email protected]
1The
Institute for Environmental Modeling,
569 Dabney Hall, University of Tennessee, Knoxville, TN 37996
[email protected], [email protected], [email protected]
The Structure of ALFISH Model
Performance
ALFISH is an Intermediate Trophic Level Functional Groups model which includes two
main subgroups (small planktivorous fish and large piscivirous fish), structured by size.
Performance of Static and Dynamic Landscape Partitioning
Objectives
• To compare, in a spatially explicit manner, the relative
effects of alternative hydrologic scenarios on fresh-water
fish densities across South Florida.
• To provide a measure of dynamic, spatially-explicit food
resources available to wading birds.
• Speedup model performance to allow real-time data
generation and information sharing between models.
Louis J.
1
Gross
Initialization
Static Partition
Dynamic partition
Update water and lower trophic data
Density-independent fish movement
Diffusive fish movement
Mortality/Aging/Growth/Reproduction
The ATLSS Hierarchy of Models
End of
run?
Output
An ecological multimodel designed to assess the effects on key biota of alternative water
management plans for the regulation of water flow across the Everglades landscape.
yes Final
output
no
Parallel Computational Model
Master Processor
Cape Sable
Seaside Sparrow
Individual-Based
Models
Snail Kite
White-tailed Deer
Wading Birds
Florida Panther
Radio-telemetry
Tracking Tools
Computational Processors
Master Thread
Initialization
Initialization
Initialization
Create threads
Update water and lower trophic data
Age/Size Structure
Models
Fish Functional Groups
Alligators
Reptiles and Amphibians
Linked Cell
Models
(MPI_Broadcast)
Vegetation
Process Models
Spatially-Explicit
Species Index Models
Output
Long-legged
Wading Birds
Short-legged
Wading Birds
White-tailed Deer
Density-independent fish movement
Snail Kite
High Resolution Topography
High Resolution Hydrology
(MPI_SENDRECV)
Save output from
previous step
Join threads
repartition boundary
no
End of run?
End of run?
yes
no
no
Final output
Diffusive fish movement
(Barrier_wait)
14
CPUs
Proc
Proc
Proc
2(16K) Cache
2(16K) Cache
2(16K) Cache
8M Ext Cache
8M Ext Cache
8M Ext Cache
3.2 Gbs System Interconnection
Grid-based partitioning can be highly effective for age- and size-structured
spatially explicit landscape fish models. Adapting dynamic partitioning, the fast
model turnaround time is about 2.5 hours yielding a speedup factor of 12.
•
The MPI implementation requires substantial (main) memory and is suitable for
execution on clusters with high speed interconnection.
•
Compared to the MPI implementation, a Pthread implementation of the model
requires less (main) memory and more scalable, but is more workload sensitive.
Save date to memory
(pthread_signal)
Final output
Finalization
Computational Environments
Whole System
Processors
14(400MHz)
Architecture
UltraSPARC
L1 Cache:
16-KB / 16-KB
Ext Cache:
8 MB
System Interconnect 3 GB/sec
Main Memory
10 GB
Storage
1.5 TB
•
(Barrier_wait)
End of run?
yes
yes
Conclusions
(Barrier_wait)
Mortality/Aging/Growth/Reproduction
(MPI_Send)
Disturbance
© TIEM / University of Tennessee
Density-independent fish movement
(pthread_cond_wait)
Diffusive fish movement
Move data between boundaries
Abiotic Conditions
Models
(pthread_cond_wait)
(pthread_signal)
Mortality/Aging/Growth/Reproduction
Alligators
Update water data from memory
Update water data
for next step
(MPI_SENDRECV)
Cape Sable
Seaside Sparrow
Update lower trophic data
Receive Date
Repartition the
landscape
Lower Trophic Level Components
Computational Threads
Simulation Results
January 1
April 1
October 1
Selected References

D. Wang, M. W. Berry, E. Carr, L. J. Gross, Design and Implementation of Dynamic Fish Model in
South Florida on Parallel Architecture, Proceedings of 37th Hawaii International Conference on
System Sciences, 2004

H.Gaff, D. DeAngelis, L. Gross, R. Salinas, M. Shorrosh, A Dynamic Landscape Model for Fish in
the Everglades and its Application to Restoration, Ecological Modeling, 127 (2000), pp. 33--53.

H. Gaff, J. Chick, J. Trexler, D. L DeAngelis, L. J. Gross, R. Salinas, Evaluation of and Insights from
ALFISH: a Spatial-explicit, Landscape-level Simulation of Fish Populations in the Everglades,
Hydrobiologia (to appear).

D. Wang, M. W. Berry, E. Carr, L. J. Gross, A Parallel Fish Landscape Model for Ecosystem
Modeling on a Computing Grid, Parallel and Distributed Computing Practices (in review)
10 G Main Memory
Proc
16
nodes
Proc
Proc
Proc
Proc
Proc
System bus
System bus
System bus
Memory
Memory
Memory
1Gbs Ethernet Network with dedicated switches
Each Node
Processors
2(450MHz)
Architecture
UltraSPARC
L1 Cache:
16-KB / 16-KB
Ext Cache:
4 MB
Main Memory
0.5 GB
Storage
30 GB
System Connect
1 GB/sec
This research has been supported by the National Science Foundation
under grant No. DEB-0219269. This research used the resources of the
Scalable Intracampus Research Grid (SInRG) Project at the University
of Tennessee, supported by the National Science Foundation CISE
Research Infrastructure Award EIA-9972889.
Spatial averaged (over 31 years) fish density map in
Everglades. No discernable differences are observed in
spatial outputs form the different parallel implementations.
The sequential implementation was developed with support from
the U.S. Geological Survey, through Cooperative Agreement No.
1445-CA09-95-0094 with the University of Tennessee, and the
Department of Interior's Critical Ecosystem Studies Initiative.