BlueM.Mines.Edu
BlueM.Mines.Edu
Mines’ Big Iron Supercomputer
Mines’ Big Iron Supercomputer
154 Tflops 17.4Tbytes 10,496 Cores 85KW
154 Tflops 17.4Tbytes 10,496 Cores 85KW


BlueM is Mines’ newest HPC platform (in addition to the “Mio”
BlueM is Mines’ newest HPC platform (in addition to the “Mio”
cluster). It is a unique machine, composed of two distinct compute
cluster). It is a unique machine, composed of two distinct compute
platforms or partitions that share a common file system. Both
platforms or partitions that share a common file system. Both
platforms, as well as the file system, were purchased from IBM as a
platforms, as well as the file system, were purchased from IBM as a
package. The common file system shared between the partitions
package. The common file system shared between the partitions
can hold 480 TB. It has efficient support for parallel operation
can hold 480 TB. It has efficient support for parallel operation
(that is, multiple cores accessing it at the same time). The two
(that is, multiple cores accessing it at the same time). The two
compute platforms are optimized for different purposes.
compute platforms are optimized for different purposes.




The smaller compute platform, in terms of capability, is AuN
The smaller compute platform, in terms of capability, is AuN
(“Golden”). It is a traditional HPC platform using
(“Golden”). It is a traditional HPC platform using
standard Intel processors. It contains 144
standard Intel processors. It contains 144
compute nodes connected by a high-speed
compute nodes connected by a high-speed
network. Each node contains 16 Intel
network. Each node contains 16 Intel
SandyBridge compute cores and 64 GB of
SandyBridge compute cores and 64 GB of
memory for a total 2,304 cores and 9,216 GB of
memory for a total 2,304 cores and 9,216 GB of
memory. AuN is rated at 50 Tflops. It is housed in
memory. AuN is rated at 50 Tflops. It is housed in
two double-wide racks with 72 nodes in each
two double-wide racks with 72 nodes in each
rack. AuN is designed to run jobs that require
rack. AuN is designed to run jobs that require
more memory per core.
more memory per core.




Mc2 (“Energy”) is an IBM BlueGene. Mc2 is housed in a single
Mc2 (“Energy”) is an IBM BlueGene. Mc2 is housed in a single
large 4’ x 4’ rack, currently half full with room for expansion. The
large 4’ x 4’ rack, currently half full with room for expansion. The
BlueGene computer is designed from the ground up as an HPC
BlueGene computer is designed from the ground up as an HPC
platform. It has a very-high-speed network connecting the nodes
platform. It has a very-high-speed network connecting the nodes
so applications can scale well. Each node has a
so applications can scale well. Each node has a
processor dedicated to systems operations in
processor dedicated to systems operations in
addition to the 16 cores that are available for users.
addition to the 16 cores that are available for users.
The processors on Mc2 are IBM “Power” processors.
The processors on Mc2 are IBM “Power” processors.
Mc2 has 512 compute nodes, each with 16 GB of
Mc2 has 512 compute nodes, each with 16 GB of
memory for a total core count of 8,912 user cores
memory for a total core count of 8,912 user cores
and 8,912 GB of memory. Mc2 is rated at 104 Tflops.
and 8,912 GB of memory. Mc2 is rated at 104 Tflops.
Mc2 is designed for jobs that can make use of a
Mc2 is designed for jobs that can make use of a
large number of cores.
large number of cores.


The total power consumption of the system is about
The total power consumption of the system is
85 kW with only 35 kW used by Mc2.
about 85 kW with only 35 kW used by Mc2.


Mc2 is water cooled. AuN is currently running with rear door heat
Mc2 is water cooled. AuN is currently running with rear door heat
exchangers but could be run using air cooling only. BlueM is
exchangers but could be run using air cooling only. BlueM is
housed at the National Renewable Energy Laboratory in
housed at the National Renewable Energy Laboratory in
Golden, CO.
Golden, CO.





Sample Projects
Sample Projects
PI
Project Title
PI
Project Title
K. Smits Sustainable Energy Pathways Collaborative: Pathways to Scalable,
K. Smits Sustainable Energy Pathways Collaborative: Pathways to Scalable,
Efficient and Sustainable Soil Borehole Thermal Energy Storage
Efficient and Sustainable Soil Borehole Thermal Energy Storage
Systems
Systems
Ganesh
• HPC algorithms for deterministic and stochastic scattering and
Ganesh
• HPC algorithms for deterministic and stochastic scattering and
absorption models
absorption models

• Simulation of Dynamics in Multiscale and Evolution Models

• Simulation of Dynamics in Multiscale and Evolution Models
Dean
Diesel Reforming Technology with Hydrogen Peroxide as an
Dean
Diesel Reforming Technology with Hydrogen Peroxide as an
Oxidant for Submarine Applications.
Oxidant for Submarine Applications.
Hogue
• Advancing Hydrometeorological Predictions in Disturbed
Hogue
• Advancing Hydrometeorological Predictions in Disturbed
Mountain Watersheds: Integrating MODIS and Landsat products
Mountain Watersheds: Integrating MODIS and Landsat products
into the WRF-Hydro Framework
into the WRF-Hydro Framework
• Fate of imported water in a semi-arid metropolis using a fully
• Fate of imported water in a semi-arid metropolis using a fully
coupled hydrologic model4
coupled hydrologic model4
D. Wu
Understanding Extractant Aggregation Through Molecular
D. Wu
Understanding Extractant Aggregation Through Molecular
Dynamics Simulations
Dynamics Simulations
Sum
Investigations of Gas Hydrate Nucleation and Growth Using
Sum
Investigations of Gas Hydrate Nucleation and Growth Using
Molecular Simulations
Molecular Simulations
Koh
Computational modeling of clathrate hydrates to study guest-
Koh
Computational modeling of clathrate hydrates to study guest-
guest and guest-host interactions
guest and guest-host interactions
Maxwell
• Investigating Impacts of Landscape Scale Change in California:
Maxwell
• Investigating Impacts of Landscape Scale Change in California:
Response of Important Hydrologic Systems to Climate Change
Response of Important Hydrologic Systems to Climate Change

and Irrigation

and Irrigation
• Subsurface residence times of dissolved organic carbon in
• Subsurface residence times of dissolved organic carbon in
Mountain Pine Beetle infested mountain catchments
Mountain Pine Beetle infested mountain catchments
• East River High Resolution Modeling
• East River High Resolution Modeling
Eberhart DFT-based materials, enzyme and explosives research
Eberhart DFT-based materials, enzyme and explosives research
Z. Wu
Quantum Mechanical Simulations of Complex Nanostructures for
Z. Wu
Quantum Mechanical Simulations of Complex Nanostructures for
Photovoltaic Applications
Photovoltaic Applications
Lusk
• Carrier Collection and Transport in Thin Film Silicon with
Lusk
• Carrier Collection and Transport in Thin Film Silicon with
Tailored Nanocrystalline/Amorphous Structure
Tailored Nanocrystalline/Amorphous Structure

• Non-Linear Up-Conversion Processes in Organic Molecular

• Non-Linear Up-Conversion Processes in Organic Molecular
Systems
Systems
• Nano-Photonic Control for Enhanced Upconversion
• Nano-Photonic Control for Enhanced Upconversion
• Phonon assisted transport and carrier relaxation in amorphous
• Phonon assisted transport and carrier relaxation in amorphous
and nanocrystalline silicon materials
and nanocrystalline silicon materials
• Electronic structure and reactivity study of substituted
• Electronic structure and reactivity study of substituted
silsesquioxanes
silsesquioxanes
• Charge and Exciton Transport in Quantum Dot Solids
• Charge and Exciton Transport in Quantum Dot Solids
Maupin
• Bio Fuel Degradation
Maupin
• Bio Fuel Degradation
• Ionic Liquid-Water-Enzyme Mixtures
• Ionic Liquid-Water-Enzyme Mixtures
• Ionic Liquids and Metal Chlorides
• Ionic Liquids and Metal Chlorides
• Computational Investigations into Novel Polymer Membranes
• Computational Investigations into Novel Polymer Membranes
Kappes Combinatorial computational screening of organic flow battery
Kappes Combinatorial computational screening of organic flow battery
electrolytes
electrolytes
Ciobanu Structural polymorphs of lithium alumino-silicates
Ciobanu Structural polymorphs of lithium alumino-silicates
L. Carr
Entangled Many-Body Quantum Dynamics
L. Carr
Entangled Many-Body Quantum Dynamics
Vyas
Rational Design of Superoxide Precursor
Vyas
Rational Design of Superoxide Precursor
B. Wu
Large-Scale Graph Analytics
B. Wu
Large-Scale Graph Analytics




More Information
More Information
http://inside.mines.edu/HPC-BlueM
http://inside.mines.edu/HPC-BlueM
http://hpc.mines.edu
http://hpc.mines.edu
http://inside.mines.edu/HPC-MachineStat
http://inside.mines.edu/HPC-MachineStat


Timothy H. Kaiser, Ph.D. (tkaiser@mines.edu)
Timothy H. Kaiser, Ph.D. (tkaiser@mines.edu)
Director of High Performance Computing
Director of High Performance Computing