Install relax cloud.sagemath.com

The SageMathCloud is a cloud computing platform designed for computational mathematics that is a part of the Sage project.

Purpose

Why ?

• The first installation of relax can be an energy barrier
• It could be of interest to make relax available in "the cloud" for the first analysis, so time is not used on solving software dependencies

Problems with "the cloud"

• Very time consuming to "run an online" cloud.
• The benefits of cloud.sagemath.com are well explained here: http://www.sagemath.org/library-why.html

Other info

• The pricing is here: https://cloud.sagemath.com/policies/pricing.html
• There is also the Bio-NMR initiative: http://www.bio-nmr.net
• And a web service with all NMR software - WeNMR: http://www.wenmr.eu

Possibilities with cloud.sagemath.com

cloud.sagemath.com currently does not provide admin account or dedicated SMC servers.
A subscription can be purchased for $49 per month (or $499 per year), one can apply the quotas listed here to any projects:

* 16 upgrades Member hosting
* 40 upgrades Network access
* 8 days Idle timeout
* 24 GB Memory
* 40 GB Disk space
* 4 shares CPU shares
* 2 cores CPU cores

The procedure should be

• A new interested scientist would go to SMC, create a project.
• Add relax as a collaborator to their free project, and we would click a button to upgrade their project to member hosting.
• This add network access, increase the cpu shares, etc.
• "We" have a page which lists, for each project, what upgrades "we" have applied
• "We" can also see if the project is being used.
• "We" can remove upgrades at any time (in any granularity).

The value of upgrading to "member hosting" is:

• Computation is much faster. The free pool of computers are typically massively loaded during the week.
• The "free servers" work fine, but they can easily be 15% of the speed of the members-only machines.
• With the $49/month subscription, "we" get to move 16 projects to members-only machines.

Installation and analysis in the cloud

Installation of relax in cloud.sagemath.com

• Instead of making "relax" available for all users on cloud.sagemath.com, each user should install their local copy of relax
• When using svn to checkout relax, there will be a: install_relax_cloud_sagemath_com.sh bash script
• The user now have full control over any version of relax
• The user can potentially "develop" online with subversion

To do Analysis

• Use ssh to send data to the server
  • Supported right now, even for free projects. See the button "ssh to your project" in project settings.
• The uploaded data for the researcher is private:
  • Even for free projects, data in projects is only visible to collaborators on that project and nobody else.
  • SMC is not like github, where free = public by default.
  • With SMC everything is private by default, even for free users.
  • However, people can also explicitly share any file or directory tree, and that makes it publicly available
  • Could be a way to share data related to a paper or research project
• Use relax in the terminal in the server
  • Fine for free projects.

The test project at cloud.sagemath.com

This audit was performed: 2015/11/24

Project location

https://cloud.sagemath.com/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/files/test%20ipyth.ipynb

Project resources

Disk space: 3000 MB disk space available - 59 MB used    3000 MB given by free project
Memory: 1000 MB RAM memory available - 133 MB used    1000 MB given by free project
CPU cores:1 core    1 core given by free project
CPU shares: 1 share    1 share given by free project
Idle timeout: 1 hour of non-interactive use before project stops    1 hour given by free project
Network access: Yes    1 upgrade given by free project
Member hosting: No    0 upgrades given by free project

Software at cloud.sagemath.com

Is git available

~$ git --version
git version 2.7.6

Is scons available

~$ which scons
/usr/bin/scons

~$ scons --version
SCons by Steven Knight et al.:
        script: v2.3.1, 2014/03/02 14:18:15, by garyo on lubuntu
        engine: v2.3.1, 2014/03/02 14:18:15, by garyo on lubuntu
        engine path: ['/usr/lib/scons/SCons']

Is xmgrace available

~$ which xmgrace
/usr/bin/xmgrace

Is mpirun available

~$ which mpicc
/usr/bin/mpicc

~$ which mpiexec
/usr/bin/mpiexec

~$ which mpirun
/usr/bin/mpirun

~$ mpirun --version
mpirun (Open MPI) 1.6.5
 
Report bugs to http://www.open-mpi.org/community/help/

~$ mpirun --report-bindings -np 4 echo "hello world"
hello world
hello world
hello world
hello world

Server

~$ lscpu
Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                4
On-line CPU(s) list:   0-3
Thread(s) per core:    2
Core(s) per socket:    2
Socket(s):             1
NUMA node(s):          1
Vendor ID:             GenuineIntel
CPU family:            6
Model:                 63
Model name:            Intel(R) Xeon(R) CPU @ 2.30GHz
Stepping:              0
CPU MHz:               2299.972
BogoMIPS:              4599.94
Hypervisor vendor:     KVM
Virtualization type:   full
L1d cache:             32K
L1i cache:             32K
L2 cache:              256K
L3 cache:              46080K
NUMA node0 CPU(s):     0-3

Python packages at cloud.sagemath.com

Start Python:

python

The current Python version:

Python 2.7.9 (default, Apr  2 2015, 15:33:21)
[GCC 4.9.2] on linux2
Type "help", "copyright", "credits" or "license" for more information.

Paste in:

try:
    import wx
    print wx.__file__
    print wx.version()
    print wx.__version__
except ImportError:
    print "no wx module"


try:
    import bmrblib
    print bmrblib.__file__
    print bmrblib.__version__
except ImportError:
    print "no bmrblib module"


try:
    import minfx
    print minfx.__file__
    print minfx.__version__
except ImportError:
    print "no minfx module"


try:
    import mpi4py
    print mpi4py.__file__
    print mpi4py.__version__
except ImportError:
    print "no mpi4py module"

Output:

>>> try:
...     import wx
...     print wx.__file__
...     print wx.version()
...     print wx.__version__
... except ImportError:
...     print "no wx module"
...
no wx module
>>> try:
...     import bmrblib
...     print bmrblib.__file__
...     print bmrblib.__version__
... except ImportError:
...     print "no bmrblib module"
...
no bmrblib module
>>> try:
...     import minfx
...     print minfx.__file__
...     print minfx.__version__
... except ImportError:
...     print "no minfx module"
...
no minfx module
>>> try:
...     import mpi4py
...     print mpi4py.__file__
...     print mpi4py.__version__
... except ImportError:
...     print "no mpi4py module"
...
no mpi4py module

Install Python modules

Bmrblib

See Bmrblib

v=1.0.4

mkdir -p $HOME/Downloads
cd $HOME/Downloads
curl http://download.gna.org/bmrblib/bmrblib-$v.tar.gz -o bmrblib-$v.tar.gz
tar -xzf bmrblib-$v.tar.gz
cd bmrblib-$v
pip install .
cd ..

Minfx

See Minfx

v=1.0.12

mkdir -p $HOME/Downloads
cd $HOME/Downloads
curl http://download.gna.org/minfx/minfx-$v.tar.gz -o minfx-$v.tar.gz
tar -xzf minfx-$v.tar.gz
cd minfx-$v
pip install .
cd ..

Mpi4py

See OpenMPI#Install_mpi4py

v=2.0.0 

mkdir -p $HOME/Downloads
cd $HOME/Downloads
wget https://bitbucket.org/mpi4py/mpi4py/downloads/mpi4py-$v.tar.gz
tar -xzf mpi4py-$v.tar.gz
cd mpi4py-$v
pip install .
cd ..

Test Python packages

Paste in:

try:
    import bmrblib
    print bmrblib.__file__
    print bmrblib.__version__
except ImportError:
    print "no bmrblib module"


try:
    import minfx
    print minfx.__file__
    print minfx.__version__
except ImportError:
    print "no minfx module"


try:
    import mpi4py
    print mpi4py.__file__
    print mpi4py.__version__
except ImportError:
    print "no mpi4py module"

Output:

>>> try:
...     import bmrblib
...     print bmrblib.__file__
...     print bmrblib.__version__
... except ImportError:
...     print "no bmrblib module"
...
/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/.local/lib/python2.7/site-packages/bmrblib/__init__.pyc
1.0.4
>>> try:
...     import minfx
...     print minfx.__file__
...     print minfx.__version__
... except ImportError:
...     print "no minfx module"
...
/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/.local/lib/python2.7/site-packages/minfx/__init__.pyc
1.0.12
>>> try:
...     import mpi4py
...     print mpi4py.__file__
...     print mpi4py.__version__
... except ImportError:
...     print "no mpi4py module"
...
/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/.local/lib/python2.7/site-packages/mpi4py/__init__.pyc
2.0.0

Checkout relax and build

cd $HOME
git clone git://git.code.sf.net/p/nmr-relax/code/ relax

# Build
cd relax
scons

# Link to executable
cd $HOME
mkdir -p $HOME/bin
ln -s $HOME/relax/relax $HOME/bin

# Test all of relax with unittests and systemtests
relax --time -u
relax --time -s

relax works!

Test of OpenMPI

See OpenMPI#Linux_and_Mac

mpirun --report-bindings -np 4 echo "hello world"

mpirun --report-bindings -np 2 relax --multi='mpi4py'

Traceback (most recent call last):
  File "/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/bin/relax", line 7, in <module>
    relax.start(profile_flag=False)
  File "/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/relax_trunk/relax.py", line 114, in start
    processor = load_multiprocessor(relax.multiprocessor_type, callbacks, processor_size=relax.n_processors, verbosity=verbosity)
  File "/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/relax_trunk/multi/__init__.py", line 189, in load_multiprocessor
    object = clazz(callback=callback, processor_size=processor_size)
  File "/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/relax_trunk/multi/mpi4py_processor.py", line 56, in __init__
    raise Too_few_slaves_exception()
multi.multi_processor_base.Too_few_slaves_exception: master slave processing requires at least 2 processors to run you only provided 1, exiting....
Traceback (most recent call last):
  File "/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/bin/relax", line 7, in <module>
    relax.start(profile_flag=False)
  File "/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/relax_trunk/relax.py", line 114, in start
    processor = load_multiprocessor(relax.multiprocessor_type, callbacks, processor_size=relax.n_processors, verbosity=verbosity)
  File "/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/relax_trunk/multi/__init__.py", line 189, in load_multiprocessor
    object = clazz(callback=callback, processor_size=processor_size)
  File "/projects/9cd9d431-3eb2-4cde-a03a-4b9c6b49a1c2/relax_trunk/multi/mpi4py_processor.py", line 56, in __init__
    raise Too_few_slaves_exception()
multi.multi_processor_base.Too_few_slaves_exception: master slave processing requires at least 2 processors to run you only provided 1, exiting....
--------------------------------------------------------------------------
mpirun noticed that the job aborted, but has no info as to the process
that caused that situation.
--------------------------------------------------------------------------