Michigan Technological University has several computing and visualization clusters and one-off work-stations, for research and teaching. Researchers may refer to the computing resources section to find an optimum match for their needs.


Annual proactive maintenance


In order to ensure that Superior and Portage continue to operate consistently at a high level, IT will be performing proactive maintenance work starting at 8 am on Tuesday, 5th May 2014.

All nodes -- including the front end, login and storage nodes -- will need to be powered down. Any running simulation at that time will be automatically terminated without further notice.

As such, researchers are requested to complete (or at least checkpoint) their simulations, and transfer any necessary data to a local workstation so that analysis, visualization, and manuscript preparation work can still be continued.

Superior and Portage should be accessible as before starting at 8 am on Friday, 8th May 2014.



VASP Implicit Solvation Model


An additional variant of VASP to facilitate implicit solvation model computations has now been compiled (for version 5.3.3) and is now available for use on all clusters. Necessary job submission script can be generated via qgenscript.

Researchers specifically authorized to use VASP are strongly recommended to verify the results for simpler/inexpensive systems with available literature before proceeding ahead with subsequent, production quality attempts for bigger/complex systems.



VASP Spin Orbit Coupling


An additional variant of VASP to facilitate spin orbit coupling computations has now been compiled (for versions 5.3.3 and 5.3.5) and is now available for use. Necessary job submission script can be generated via qgenscript.

Researchers are strongly recommended to verify the results for simpler/inexpensive systems -- preferably with both versions when permissible -- with available literature before proceeding ahead with subsequent, production quality attempts for bigger/complex systems.



R 3.1.2 (Pumpkin Helmet)


R 3.1.2 (Pumpkin Helmet; Release Notes) has been compiled and is now available for use on all computing clusters. Necessary job submission script can be generated via qgenscript.

Researchers are strongly recommended to re-run a previously successful simulation with this newer version and make sure that the results are consistent before proceeding ahead with subsequent, production quality attempts.

Unless there is a request from researchers to do otherwise, 3.1.1 (Sock it to me) will be retired on Friday, 27th March 2015.



MATLAB R2014b


MATLAB R2014b (Release notes) has been installed and is now available for use on all computing clusters. Necessary job submission script can be generated via qgenscript.

Researchers are strongly recommended to re-run a previously successful simulation with this newer version and make sure that the results are consistent before proceeding ahead with subsequent, production quality attempts.

Unless there is a request from researchers to do otherwise, R2013b will be retired on Friday, 27th March 2015.



R 3.1.1 (Sock it to Me)


R 3.1.1 (Sock it to Me; Release Notes) has been compiled and is now available for use on all computing clusters. Necessary job submission script can be generated via qgenscript.

Researchers are strongly recommended to re-run a previously successful simulation with this newer version and make sure that the results are consistent before proceeding ahead with subsequent, production quality attempts.


Researcher spotlight

Dr. Gregory Odegard
Professor, ME-EM

Learn more about how computing is used in his research endeavors.

Dr. Maximilian Seel
Professor, Physics

Learn more about how computing is used in his research endeavors.


Royal jelly isn't what makes a queen bee a queen bee: http://t.co/SG6QUTuwYP



High-performance @MATLAB with (@NVIDIA) GPU acceleration: http://t.co/qDJtyqq2KZ



COSMOS team achieves 100x speedup: http://t.co/cKcEWPoi3d [via @HPCWire]



Interactive design powers visual intro to machine learning: http://t.co/Ow1BrexvQO [via @insideHPC]



Exploring large data for scientific discovery: http://t.co/FeqHFZcDQv [via @HPCWire]



Making progress by slowing down: http://t.co/6pPUFgT1i8



Arithmetic intensity of stencil operations: http://t.co/8L2Onlh4Kz [via @insideHPC]



Data Science 101 - scikit-learn: http://t.co/9sz0MZUwis [via @insideHPC]



Data Science 101 - k-means clustering: http://t.co/vExxAX8Qsp [via @insideHPC]



Data Science 101 - Machine Learning #5: http://t.co/4Kd03T4ksC [via @insideHPC]



Data Science 101 - Machine Learning #4: http://t.co/SdRrnrXyq8 [via @insideHPC]



Data Science 101 - Machine Learning #3: http://t.co/avQh0ePdni [via @insideHPC]