Difference between revisions of "PhyloSoC: DIM SUM 2 GPU computing for individual based simulator of movement and demography"

From Phyloinformatics
Jump to: navigation, search
(Project Plan)
 
(3 intermediate revisions by the same user not shown)
Line 2: Line 2:
 
DIM SUM is a population demography and individual migration simulation. The goal of the project is to use the graphics card to improve the speed of DIM SUM significantly, so it is possible to simulate scenarios with much larger populations and larger landscapes.
 
DIM SUM is a population demography and individual migration simulation. The goal of the project is to use the graphics card to improve the speed of DIM SUM significantly, so it is possible to simulate scenarios with much larger populations and larger landscapes.
 
==Student==
 
==Student==
Peter Hoffmann
+
* [mailto:p-hoffmann@web.de Peter Hoffmann]
 
==Mentors==
 
==Mentors==
 
* Kevin Savidge (primary)
 
* Kevin Savidge (primary)
 
* [http://www.phyleauxgenetics.org/ Jeremy Brown]
 
* [http://www.phyleauxgenetics.org/ Jeremy Brown]
 
==Project Plan==
 
==Project Plan==
===Week 1 (May 23 - May 30):===
+
===Week 1 (May 23 - May 29):===
 
* Performance analysis of DIM SUM.
 
* Performance analysis of DIM SUM.
 
* Split populateAndMigrateThreaded into smaller function to better understand it.
 
* Split populateAndMigrateThreaded into smaller function to better understand it.
===Week 2 (May 30 - June 6):===
+
===Week 2 (May 30 - June 5):===
 
* Replace the ArrayLists of Objects in DispersalSettings, which are needed for the computations in the populateAndMigrateThreaded, with data structure, which can be copied in the graphic memory. The only acceptable data structures for CUDA are arrays of primitives. That means e.g. that all ArrayList<PFunction> must be replaced by a two dimension array for the outcomes and a two dimensional array for the probabilities.
 
* Replace the ArrayLists of Objects in DispersalSettings, which are needed for the computations in the populateAndMigrateThreaded, with data structure, which can be copied in the graphic memory. The only acceptable data structures for CUDA are arrays of primitives. That means e.g. that all ArrayList<PFunction> must be replaced by a two dimension array for the outcomes and a two dimensional array for the probabilities.
===Week 3 (June 6 - June 13):===
+
===Week 3 (June 6 - June 12):===
* Start replacing the ArrayLists, which stores the generation with CUDA compatible data structures. Even thought this are only one or two ArrayLists this task might be more time consuming than the first one, because there are elements added and deleted to this data structure during the simulation and I think the GPU can't resize an array.
+
* Replace the ArrayList, which stores the generation with a CUDA compatible data structure. Even thought it is only one ArrayList this task isn't trivial, because elements are removed from this array during the computation, which isn't possible on the GPU.
===Week 4 (June 13 - June 20):===
+
===Week 4 (June 13 - June 19):===
* Replace lldata with an Array, which stores the double values.
+
* Replace the class lldata with a data structure, which is compatible with CUDA.
===Week 5 (June 20 - June 27):===
+
* Remove variables from the Nodes, which are only needed for debugging.
* Finish replacing the ArrayLists, which stores the generation with CUDA compatible data structures.
+
===Week 5 (June 20 - June 26):===
===Week 6 (June 27 - July 4):===
+
* Replace the Nodes with a data structures, which can be used on the graphics card.
* Port the functions in XYFunction and DispersalSettings,which are needed by the migrate function, to CUDA.
+
* Code refactoring.
===Week 7 (July 4 - July 11):===
+
===Week 6 (June 27 - July 3):===
* Port the migrate function to CUDA. Even through this function is not used anymore. Porting this function to CUDA is a good preparation for porting populateAndMigrateThreaded. The parallelization of this function should be relative easy. The basic Idea is that the computation per child are independent from each other, so they can be executed in parallel. I think the only problem is that elements are removed from the children array, which can be solved by marking the elements (e.g. in a boolean array) and remove them on the CPU.  
+
* Replace decimal degrees with radians in migrate.
===Week 8 (July 11 - July 18):===
+
* Port the migrate part of populateAndMigrate to C++ using SWIG. This includes all function, which migrate depends on.
* Finish to port all the functions in DispersalSettings, Pfunction and XYFunction, which are needed during the computations, to CUDA.
+
===Week 7 (July 4 - July 10):===
===Week 9 (July 18 - July 25):===
+
* Port the migrate part of populateAndMigrate from C++ to CUDA. This includes all function, which migrate depends on. There might be still some limitations like a maximum image size of 2048x2048.
* Port all the functions in lldata to CUDA.
+
===Week 8 (July 11 - July 17):===
===Week 10 (July 25 - August 1):===
+
* Optimize performance of CUDA functions.
* Start porting the populateAndMigrateThreaded function to CUDA. The function is already parallelized. Everything which is executed in parallel in DIM SUM should now be executed on the GPU. That means that the run function in DispersalThread has to be ported to the GPU. Like with the migrate function the data structures are the main concern.
 
===Week 11 (August 1 - August 8):===
 
* Finish porting the populateAndMigrateThreaded function to CUDA. The function is already parallelized. Everything which is executed in parallel in DIM SUM should now be executed on the GPU. That means that the run function in DispersalThread has to be ported to the GPU. Like with the migrate function the data structures are the main concern.
 
===Week 12 (August 8 - August 15):===
 
 
* Bug-fixing.
 
* Bug-fixing.
 +
===Week 9 (July 18 - July 24):===
 +
* Optimize the performance of other functions.
 +
* Further improve migrate performance.
 +
* Bug-fixing.
 +
===Week 10 (July 25 - July 31):===
 +
* Improve output performance.
 +
* Performance evaluation.
 +
* Bug-fixing.
 +
===Week 11 (August 1 - August 7):===
 +
* Code cleanup.
 +
* Improve documentation.
 +
===Week 12 (August 8 - August 14):===
 
* Code cleanup.
 
* Code cleanup.
 
* Improve documentation.
 
* Improve documentation.

Latest revision as of 12:36, 21 June 2011

Abstract

DIM SUM is a population demography and individual migration simulation. The goal of the project is to use the graphics card to improve the speed of DIM SUM significantly, so it is possible to simulate scenarios with much larger populations and larger landscapes.

Student

Mentors

Project Plan

Week 1 (May 23 - May 29):

  • Performance analysis of DIM SUM.
  • Split populateAndMigrateThreaded into smaller function to better understand it.

Week 2 (May 30 - June 5):

  • Replace the ArrayLists of Objects in DispersalSettings, which are needed for the computations in the populateAndMigrateThreaded, with data structure, which can be copied in the graphic memory. The only acceptable data structures for CUDA are arrays of primitives. That means e.g. that all ArrayList<PFunction> must be replaced by a two dimension array for the outcomes and a two dimensional array for the probabilities.

Week 3 (June 6 - June 12):

  • Replace the ArrayList, which stores the generation with a CUDA compatible data structure. Even thought it is only one ArrayList this task isn't trivial, because elements are removed from this array during the computation, which isn't possible on the GPU.

Week 4 (June 13 - June 19):

  • Replace the class lldata with a data structure, which is compatible with CUDA.
  • Remove variables from the Nodes, which are only needed for debugging.

Week 5 (June 20 - June 26):

  • Replace the Nodes with a data structures, which can be used on the graphics card.
  • Code refactoring.

Week 6 (June 27 - July 3):

  • Replace decimal degrees with radians in migrate.
  • Port the migrate part of populateAndMigrate to C++ using SWIG. This includes all function, which migrate depends on.

Week 7 (July 4 - July 10):

  • Port the migrate part of populateAndMigrate from C++ to CUDA. This includes all function, which migrate depends on. There might be still some limitations like a maximum image size of 2048x2048.

Week 8 (July 11 - July 17):

  • Optimize performance of CUDA functions.
  • Bug-fixing.

Week 9 (July 18 - July 24):

  • Optimize the performance of other functions.
  • Further improve migrate performance.
  • Bug-fixing.

Week 10 (July 25 - July 31):

  • Improve output performance.
  • Performance evaluation.
  • Bug-fixing.

Week 11 (August 1 - August 7):

  • Code cleanup.
  • Improve documentation.

Week 12 (August 8 - August 14):

  • Code cleanup.
  • Improve documentation.