Closed mlzhangli closed 2 days ago
xji3
commented
3 days ago @mlzhangli We recommend using the cuda build of BEAGLE on NVIDIA cards. You are currently using the openCL build and it seems that you didn't build the CUDA version of beagle. Please follow the instructions to build beagle with cmake and let us know (https://github.com/beagle-dev/beagle-lib/wiki/LinuxInstallInstructions).
mlzhangli
commented
2 days ago Oh I forgot to mention the cuda part, it didn't work at first so I changed to openCL, here is the info about it
$ ~/zl/BEASTv1.10.4/bin/beast -beagle_info
BEAST v1.10.4 Prerelease #bc6cbd9, 2002-2018
Bayesian Evolutionary Analysis Sampling Trees
Designed and developed by
Alexei J. Drummond, Andrew Rambaut and Marc A. Suchard
Department of Computer Science
University of Auckland
alexei@cs.auckland.ac.nz
Institute of Evolutionary Biology
University of Edinburgh
a.rambaut@ed.ac.uk
David Geffen School of Medicine
University of California, Los Angeles
msuchard@ucla.edu
Downloads, Help & Resources:
http://beast.community
Source code distributed under the GNU Lesser General Public License:
http://github.com/beast-dev/beast-mcmc
BEAST developers:
Alex Alekseyenko, Guy Baele, Trevor Bedford, Filip Bielejec, Erik Bloomquist, Matthew Hall,
Joseph Heled, Sebastian Hoehna, Denise Kuehnert, Philippe Lemey, Wai Lok Sibon Li,
Gerton Lunter, Sidney Markowitz, Vladimir Minin, Michael Defoin Platel,
Oliver Pybus, Chieh-Hsi Wu, Walter Xie
Thanks to:
Roald Forsberg, Beth Shapiro and Korbinian Strimmer
Using BEAGLE library v4.0.0 (PRE-RELEASE) for accelerated, parallel likelihood evaluation
2009-, BEAGLE Working Group - https://beagle-dev.github.io/
Citation: Ayres et al (2019) Systematic Biology 68: 1052-1061 | doi:10.1093/sysbio/syz020
BEAGLE resources available:
0 : CPU (x86_64)
Flags: PRECISION_SINGLE PRECISION_DOUBLE COMPUTATION_SYNCH EIGEN_REAL EIGEN_COMPLEX SCALING_MANUAL SCALING_AUTO SCALING_ALWAYS SCALERS_RAW SCALERS_LOG VECTOR_SSE VECTOR_NONE THREADING_CPP THREADING_NONE PROCESSOR_CPU FRAMEWORK_CPU
1 : NVIDIA GeForce GTX 1080 Ti
Global memory (MB): 11179
Clock speed (Ghz): 1.58
Number of cores: 3584
Flags: PRECISION_SINGLE PRECISION_DOUBLE COMPUTATION_SYNCH COMPUTATION_ASYNCH EIGEN_REAL EIGEN_COMPLEX SCALING_MANUAL SCALING_AUTO SCALING_ALWAYS SCALERS_RAW SCALERS_LOG VECTOR_NONE THREADING_CPP THREADING_NONE PROCESSOR_GPU FRAMEWORK_CUDA PARALLELOPS_STREAMS PARALLELOPS_GRID
2 : NVIDIA GeForce GTX 1080 Ti
Global memory (MB): 11179
Clock speed (Ghz): 1.58
Number of cores: 3584
Flags: PRECISION_SINGLE PRECISION_DOUBLE COMPUTATION_SYNCH COMPUTATION_ASYNCH EIGEN_REAL EIGEN_COMPLEX SCALING_MANUAL SCALING_AUTO SCALING_ALWAYS SCALERS_RAW SCALERS_LOG VECTOR_NONE THREADING_CPP THREADING_NONE PROCESSOR_GPU FRAMEWORK_CUDA PARALLELOPS_STREAMS PARALLELOPS_GRID
3 : NVIDIA GeForce GTX 1080 Ti
Global memory (MB): 11179
Clock speed (Ghz): 1.58
Number of cores: 3584
Flags: PRECISION_SINGLE PRECISION_DOUBLE COMPUTATION_SYNCH COMPUTATION_ASYNCH EIGEN_REAL EIGEN_COMPLEX SCALING_MANUAL SCALING_AUTO SCALING_ALWAYS SCALERS_RAW SCALERS_LOG VECTOR_NONE THREADING_CPP THREADING_NONE PROCESSOR_GPU FRAMEWORK_CUDA PARALLELOPS_STREAMS PARALLELOPS_GRID
4 : NVIDIA GeForce GTX 1080 Ti
Global memory (MB): 11179
Clock speed (Ghz): 1.58
Number of cores: 3584
Flags: PRECISION_SINGLE PRECISION_DOUBLE COMPUTATION_SYNCH COMPUTATION_ASYNCH EIGEN_REAL EIGEN_COMPLEX SCALING_MANUAL SCALING_AUTO SCALING_ALWAYS SCALERS_RAW SCALERS_LOG VECTOR_NONE THREADING_CPP THREADING_NONE PROCESSOR_GPU FRAMEWORK_CUDA PARALLELOPS_STREAMS PARALLELOPS_GRID
$ ~/zl/BEASTv1.10.4/bin/beast -beagle_gpu -beagle_order 4 -overwrite FA1090_gubbins.filtered_polymorphic_sites_3.xml
Creating the tree model, 'treeModel'
taxon count = 1075
tree height = 60.29346452278587
Using discretized relaxed clock model.
over sampling = 1
parametric model = exponentialDistributionModel
rate categories = 1
Using discretized relaxed clock model.
over sampling = 1
parametric model = exponentialDistributionModel
rate categories = 1
Creating state frequencies model 'frequencies': Initial frequencies = {0.25, 0.25, 0.25, 0.25}
Creating site rate model:
4 category discrete gamma with initial shape = 0.5
initial proportion of invariant sites = 0.5
Using BEAGLE DataLikelihood Delegate
CUDA error: "Out of memory" (2) from file <zl/beagle-lib/libhmsbeagle/GPU/GPUInterfaceCUDA.cpp>, line 596.@mlzhangli From the cuda error, it seems that your card does not have enough memory to run your analysis.
This is my GPU information
And beagle_info in beast
when I run beast it says
However, the seq length in FA1090_gubbins.filtered_polymorphic_sites_3.xml is 30408, I use a different xml whose seq length is 2232367 and Sequences is 603, and this command works.