high-performance multi-user code development with google code current status (...just google for...

33
High-performance multi- High-performance multi- user code development with user code development with Google Code Google Code Current status (...just google for Pencil Code)

Upload: estella-mason

Post on 31-Dec-2015

234 views

Category:

Documents


1 download

TRANSCRIPT

High-performance multi-user code High-performance multi-user code development with Google Codedevelopment with Google Code

Current status

(...just google for Pencil Code)

PencilPencilcodecode

• Started in Sept. 2001 with Wolfgang Dobler• High order (6th order in space, 3rd order in time)• Cache & memory efficient• MPI, can run PacxMPI (across countries!)• Maintained/developed by ~40 people (SVN)• Automatic validation (over night or any time)• 0.0013 s/pt/step at 10243 , 2048 procs• http://pencil-code.googlecode.com

• Isotropic turbulence– MHD, passive scl, CR

• Stratified layers– Convection, radiation

• Shearing box– MRI, dust, interstellar– Self-gravity

• Sphere embedded in box– Fully convective stars– geodynamo

• Other applications– Chemistry, combustion– Spherical coordinates

Scaling Scaling propertiesproperties

3

Wlad Lyra on Kraken Babkovskaia et al (2011)

PencilPencilCodeCode

• Started in Sept. 2001 with Wolfgang Dobler• High order (6th order in space, 3rd order in time)• Cache & memory efficient• MPI, can run PacxMPI (across countries!)• Maintained/developed by ~40 people (SVN)• Automatic validation (over night or any time)• Max resolution so far 10243 , 4096 procs• http://pencil-code.googlecode.com

• Isotropic turbulence– MHD, passive scl, CR

• Stratified layers– Convection, radiation

• Shearing box– MRI, dust, interstellar– Self-gravity

• Sphere embedded in box– Fully convective stars– geodynamo

• Other applications– Homochirality– Spherical coordinates

5

Google Code webpageGoogle Code webpage

6

Continuous updates onlineContinuous updates online

7

Automatic validation testsAutomatic validation tests

8

Evolution of code sizeEvolution of code size

User meetings:User meetings:2005 Copenhagen2006 Copenhagen2007 Stockholm2008 Leiden2009 Heidelberg2010 New York2011 Toulouse2012 Helsinki

9

Increase in # of auto testsIncrease in # of auto tests

10

Regularity of check-insRegularity of check-ins

11

More check-ins since 2010More check-ins since 2010

12

Return to previous rateReturn to previous rate

13

14

15

16

17

TutorialTutorial

18

Tut

oria

l pag

esT

utor

ial p

ages

19

OlholOlhol

20

Olhoh.net analysisOlhoh.net analysis

21

CreditCredit

22

Free licence, but giving credit to researchFree licence, but giving credit to research

23

Faster and bigger machinesFaster and bigger machines

24

256 processor run at 1024256 processor run at 102433

25

Hyperviscous, Smagorinsky, normalHyperviscous, Smagorinsky, normal

Inertial range unaffected by artificial diffusionHau

gen

& B

rand

enbu

rg (

PR

E, a

stro

-ph/

0402

301)

height of bottleneck increased

onset of bottleneck at same position

26

Online data reduction and visualizationOnline data reduction and visualization

non-helically forced turbulence

27

Scalars on periphery of the boxScalars on periphery of the box

28

MRI turbulenceMRI turbulenceMRI = magnetorotational instabilityMRI = magnetorotational instability

2563

w/o hypervisc.t = 600 = 20 orbits

5123

w/o hypervisc.t = 60 = 2 orbits

29

Vorticity and DensityVorticity and Density

See poster by Tobi Heinemann on density wave excitation!

30

Transfer equation & parTransfer equation & paraallelizationllelization

Analytic Solution:

Ray direction

Intrinsic Calculation

Processors

SII

d

d

31

The Transfer Equation & The Transfer Equation & ParParaallelizationllelization

Analytic Solution:

Ray direction

Communication

Processors

32

The Transfer Equation & The Transfer Equation & ParallelizationParallelization

Analytic Solution:

Ray direction

Processors

Intrinsic Calculation

33

Future developmentsFuture developments

• Adaptive/multiple mesh

• More 3-D online graphics output

• Poisson solver for non-Cartesian geometries

• Parallel developments with graphics card