Home | HPC Portal

Gurobi Hardware and Performance

Mon, 15 Jul 2024 22:10:10 +0200

How does Gurobi perform on different computer hardware?

It is difficult to predict how Gurobi performs on a certain machine. In general, the solver benefits from high CPU speeds and low-latency, high-bandwidth memory.
Having multiple cores at your disposal can improve performance, but this is highly problem-dependent. This is also true for the amount of memory (RAM).
More channels, e.g., DDR4, increase the data throughput and are preferred over single-channel RAM.
If you are solving a large MIP in parallel, it is best to use a system with the fastest possible clock rate, using the fastest available memory, with as many fully populated memory channels as are available. Current Intel Xeon systems support up to six channels per CPU, while current AMD EPYC systems support up to eight. Desktop and low-end server configurations typically have fewer channels.
References:

Why does Gurobi perform differently on different machines?

Solving the same model on different machines can result in different solution paths (Performance Variability). Hardware plays an important role and changing the order of variables or constraints or using random seeds can also drastically impact the performance even when the mathematical models are identical.
The reasons for performance variability are quite diverse and can originate from differences in operating systems, underlying libraries, computer hardware, etc. Gurobi makes many decisions in the search for an optimal solution.
Gurobi attempts to exploit performance variability when running in ConcurrentMIP mode. To get a sense of how susceptible your model is in regards to performance variability, you can compare the performance of several runs that only differ in the random seed parameter.
References:

Does using more threads make Gurobi faster?

Continuous models (LP, QP, SOCP):
- Yes for barrier algorithm.
- No for simplex method.
Mixed-Integer Programs:
- Yes, if a large number of nodes is required.
- No for MIPs that are solved at or near the root node.
The default value of threads (corresponding to the value 0) is an automatic setting which will use up to min(32, number of virtual cores) threads, i.e., up to 32 threads on HSUper.
In the following cases more threads are often not better (and can be even worse):
- When the first solution found by the MIP solver is almost always optimal, and that solution isn’t found at the root.
- When memory is tight.
References:
- Does using more threads make Gurobi faster?
- Documentation: Threads

What hardware should I select when running Gurobi?

Parallelization effects are model-dependent.
GPUs do not help.
The faster the CPU, the better.
RAM requirements are model-dependent.
It is crucial to utilize a wide range of instances to ensure that the obtained results robust and represent an accurate picture of the expected load on the system.
References:
- What hardware should I select when running Gurobi?

Does Gurobi support GPUs?

GPUs aren’t well suited to the needs of an LP/MIP/QP solver (sparse linear algebra not well suited for SIMD).
Gurobi Optimizer is designed to effectively exploit multiple cores in a CPU.
References:
- Does Gurobi support GPUs?

Are there performance differences amongst the different Gurobi APIs?

The Gurobi APIs (Java, Python, C++, C#, R, and Matlab) are thin layers built on top of the C API. These APIs collect the data and pass it on to the core library, which is written in C. From a solver and solving performance point of view, there is no difference.
However, the API can make a difference for the model building part of the code. Model building is in the chosen language.
References:
- Are there performance differences amongst the different Gurobi APIs?

What is the correlation between problem size and solution time?

In general, there is no correlation between problem size and solution time with Gurobi. The time it takes Gurobi to find and prove an optimal solution depends on how well the algorithms are able to tackle the particular problem at hand.
It is often possible to identify such correlations for a particular model family, i.e., a set of models that originates from the same algebraic description but with different data.
References:
- What is the correlation between problem size and solution time?

How do I get similar performance from Gurobi based on parameters tuned for another solver?

In terms of performance tuning, there are three types of parameters: termination criteria, tolerance values, and algorithm behavior.
For termination criteria such as MIPGap or tolerance values such as IntFeasTol, Gurobi parameters should correspond closely to those from other solvers.
In terms of algorithm behavior, it is generally best to start with default values, since default values in Gurobi have been tested to be fast and robust across a wide variety of models.
References:
- How do I get similar performance from Gurobi based on parameters tuned for another solver?

Is Gurobi deterministic?

Gurobi is designed to be deterministic. You always get the same results from the same inputs (model and parameters) on the same computer with the same Gurobi version. Gurobi is also deterministic for parallel optimization.
Not deterministic for concurrent optimization.
References:
- Is Gurobi deterministic?

IBM ILOG CPLEX Optimization Studio: Installation on HSUper

Mon, 15 Jul 2024 22:10:10 +0200

Obtain the IBM ILOG CPLEX Optimization Studio software

Download (latest) IBM ILOG CPLEX Optimization Studio software for linux, e.g., IBM ILOG CPLEX Optimization Studio V22.1.1 for Linux x86-64, i.e., cplex_studio2211.linux_x86_64.bin.

Install IBM ILOG CPLEX Optimization Studio on HSUper

Copy downloaded IBM ILOG CPLEX Optimization Studio software, e.g., cplex_studio2211.linux_x86_64.bin to HSUper, e.g., via MobaXterm (only on Windows) or via the scp command.
Enter pwd in the terminal to see the current folder (i.e., your home folder) and copy/remember this path since it is needed during the installation.
Enter bash cplex_studio2211.linux_x86_64.bin in the terminal to install the IBM ILOG CPLEX Optimization Studio software. You are not allowed to use the default install folder /opt/ibm/ILOG/CPLEX_Studio2211, i.e., use instead for example your home folder (which can be seen by entering pwd in the terminal) and expand this path for example by /CPLEX_Studio2211 to create a new folder CPLEX_Studio2211 (if this folder does not already exist). This folder will contain the installed IBM ILOG CPLEX Optimization Studio software.
Add the following content (i.e., assuming you used the folder CPLEX_Studio2211 in your home folder for installation) to the end of your .bashrc file (e.g., double left click in MobaXterm on the file to open MobaTextEditor or use the text editors nano or vim):

export PATH=$PATH:$HOME/CPLEX_Studio2211/cplex/bin/x86-64_linux
export PATH=$PATH:$HOME/CPLEX_Studio2211/cpoptimizer/bin/x86-64_linux
export PATH=$PATH:$HOME/CPLEX_Studio2211/opl/bin/x86-64_linux

Save the updated .bashrc file.

Gurobi Installation on HSUper

Mon, 15 Jul 2024 22:10:10 +0200

Step 1: Obtain the Gurobi Software

Register or sign in at the Gurobi User Portal
Download the Gurobi software for Linux from Gurobi Downloads, e.g., gurobi12.0.2_linux64.tar.gz.

Step 2: Install Gurobi Software on HSUper

Copy the downloaded Gurobi software to HSUper without unpacking it beforehand and log in to HSUper. This could be done via:
- The Web Browser access to HSUper
- MobaXterm (on Windows)
- The scp and ssh commands in your terminal
Unpack the Gurobi software by running tar -xvzf gurobi12.0.2_linux64.tar.gz or equivalent in the terminal, which creates a new folder in this case named gurobi1202.
Download and rename the license file from the HSU Intranet from gurobilic.sec to gurobi.lic and copy it to the gurobi1202 folder.
Add the following content to the end of your ~/.bashrc file

export GUROBI_HOME="$HOME/gurobi1202/linux64"
export PATH="${PATH}:${GUROBI_HOME}/bin"
export LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:${GUROBI_HOME}/lib"
export GRB_LICENSE_FILE="$HOME/gurobi1202/gurobi.lic"

to install the Gurobi software locally in your actual home folder instead of /opt.

IBM ILOG CPLEX Optimization Studio Solver: Running, Configuration, Tuning

Mon, 15 Jul 2024 22:10:10 +0200

Running IBM ILOG CPLEX Optimization Studio on HSUper

Login on HSUper, e.g., via MobaXterm (only on Windows), PuTTY or via the ssh command.
Use a SLURM script to submit jobs on HSUper via entering in the terminal sbatch (followed by the name of the SLURM script).
Enter squeue -u $USER to see all your jobs in the queue (or replace $USER by an username); R at ST means that your job is currently running.
Enter squeue --format="%.18i %.9P %.30j %.8u %.8T %.10M %.9l %.6D %R" --me to extend the length of the output of squeue.
Enter scancel -u $USER to scancel all your jobs or scancel followed by a jobid to cancel the corresponding job.
Enter sacct -S now-7days -X --user=$USER --format="jobid,jobname,user,account,partition,AllocCPUS,AllocNodes,State,ConsumedEnergy,Elapsed,TimeLimit,ExitCode" to gain information about all your jobs during the last 7 days.
The oplrun command enables you to execute an OPL model or project from the Windows command line or in a UNIX environment.
A basic SLURM script could look like this:

#!/bin/bash

#SBATCH --job-name=name ### the name of your job
#SBATCH --output=%x_%j.out ### the output file for errors etc.

#SBATCH --nodes=1 ### number of nodes
#SBATCH --ntasks=1 ### number of (MPI) tasks needed
#SBATCH --partition=small ### partition (small, small_shared, small_fat, dev)
#SBATCH --cpus-per-task=72 ### number of threads per task (OMP threads)
#SBATCH --time=24:00:00 ### maximum wall clock limit for job execution

oplrun myModel.mod myData.dat > log.txt

If you want to solve several problems, e.g., with different .mod or .dat files or with varying numbers of threads (see How to change parameters in oplrun), you can add these runs in the same SLURM script in a new line after oplrun myModel.mod myData.dat > log.txt such that these problems are solved one after the other.
If you want to start all your runs as soon as possible, then you can use individual SLURM scripts and submit each of them via sbatch (followed by the name of the corresponding SLURM script). Or you could use a bash script (chmod +x myBash makes the bash script myBash executable, ./myBash executes it) to start the different SLURM scripts. The following bash script starts runs for 30 different problems (each problem is indentical except that a different data file is used with j = 1, …, 30 with start p=1 and end P=30) and different numbers of threads (i.e., i = 1, 2, 4, 8, 16, 32, 64 with start t=1 and end T=72).

#!/bin/bash

t=1
T=72
p=1
P=30
for (( j=$p ; j<=$P ; j++ ));
do
 for (( i=$t ; i<=$T ; i*=2 ));
 do
 sbatch --job-name="threads$i.problem$j" --export=ALL,i="$i",j="$j" var_param_oplrun.slurm
 done
done

The SLURM script var_param_oplrun.slurm is identical to the SLURM script from above, except that the line
oplrun myModel.mod myData.dat > log.txt
is changed to
oplrun -D myThreads=$i myModel.mod myData$j.dat > log$i.$j.txt.

Acknowledgement

Mon, 15 Jul 2024 22:10:10 +0200

Created by Dr. rer. nat. Willi Leinen, Helmut-Schmidt-Universität, Chair for High Performance Computing.
Last update: 12/08/2024

Gurobi Solver: Running, Configuration, Tuning

Mon, 15 Jul 2024 22:10:10 +0200

Solving a Model

Command-line commands:
- Run: gurobi_cl model.mps
- Write solution: gurobi_cl ResultFile=model.sol model.mps
References:

Running Gurobi on HSUper

Login on HSUper, e.g., via MobaXterm (only on Windows), PuTTY or via the ssh command.
Use a SLURM script to submit jobs on HSUper via entering in the terminal sbatch (followed by the name of the SLURM script).
Enter squeue -u $USER to see all your jobs in the queue (or replace $USER by an username); R at ST means that your job is currently running.
Enter squeue --format="%.18i %.9P %.30j %.8u %.8T %.10M %.9l %.6D %R" --me to extend the length of the output of squeue.
Enter scancel -u $USER to scancel all your jobs or scancel followed by a jobid to cancel the corresponding job.
Enter sacct -S now-7days -X --user=$USER --format="jobid,jobname,user,account,partition,AllocCPUS,AllocNodes,State,ConsumedEnergy,Elapsed,TimeLimit,ExitCode" to gain information about all your jobs during the last 7 days.
A basic SLURM script could look like this:

#!/bin/bash

#SBATCH --job-name=name ### the name of your job
#SBATCH --output=%x_%j.out ### the output file for errors etc.

#SBATCH --nodes=1 ### number of nodes
#SBATCH --ntasks=1 ### number of (MPI) tasks needed
#SBATCH --partition=small ### partition (small, small_shared, small_fat, dev)
#SBATCH --cpus-per-task=72 ### number of threads per task (OMP threads)
#SBATCH --time=24:00:00 ### maximum wall clock limit for job execution

gurobi_cl model.mps

If you want to solve several problems, e.g., with different .mps files or with varying numbers of threads (see How to change parameters in Gurobi command line), you can add these runs in the same SLURM script in a new line after gurobi_cl model.mps such that these problems are solved one after the other.
If you want to start all your runs as soon as possible, then you can use individual SLURM scripts and submit each of them via sbatch (followed by the name of the corresponding SLURM script). Or you could use a bash script (chmod +x myBash makes the bash script myBash executable, ./myBash executes it) to start the different SLURM scripts. The following bash script starts runs for 30 different problems (each problem is indentical except that a different mps file is used with j = 1, …, 30 with start p=1 and end P=30) and different numbers of threads (i.e., i = 1, 2, 4, 8, 16, 32, 64 with start t=1 and end T=72) and writes the log to individual log files.

#!/bin/bash

t=1
T=72
p=1
P=30
for (( j=$p ; j<=$P ; j++ ));
do
 for (( i=$t ; i<=$T ; i*=2 ));
 do
 sbatch --job-name="threads$i.problem$j" --export=ALL,i="$i",j="$j" var_param_gurobi.slurm
 done
done

The SLURM script var_param_gurobi.slurm is identical to the SLURM script from above, except that the line
gurobi_cl model.mps
is changed to
gurobi_cl Threads=$i LogFile="$HOME/gurobiLog/model$j.threads$i.log" model$j.mps.

Acknowledgement

Mon, 15 Jul 2024 22:10:10 +0200

Created by Dr. rer. nat. Willi Leinen, Helmut-Schmidt-Universität, Chair for High Performance Computing.
Last update: 02/06/2025

Technical Specifications

Mon, 15 Jul 2024 22:10:10 +0200

Compute Nodes

Compute nodes rely on remote file systems (like the parallel file system BeeGFS), as they do not have local storage capabilities. The local file system resides in memory (RAM).

Regular

571 compute nodes each equipped with 256 GB RAM and 2 Intel Icelake sockets; each socket features a Intel(R) Xeon (R) Platinum 8360Y processor with (up to) 36 cores, yielding a total of 72 cores per node

Limitations

Tue, 15 Oct 2024 22:10:10 +0200

If the limitations on this page are not suitable for your use case, we invite you to contact our support team at cbrz-support@hsu-hh.de. If you believe another limitation should be documented here, please do not hesitate to reach out to us.

Job Limitations

Maximum concurrent jobs per user: Up to 1,000 jobs can be submitted by a user at any given time; completed or canceled jobs do not affect this limit.
Maximum running jobs per user: 30 concurrent jobs are allowed, but partition-specific limitations can override this limit.
Slurm job capacity: Slurm has a maximum capacity to handle up to 10,000 active jobs concurrently, including array tasks that contribute to this total.

Partitions

Mon, 15 Jul 2024 22:10:10 +0200

The table below presents various partition configurations along with their respective limitations. Note that additional constraints (*), not listed here, also apply. If the limitations presented here pose a problem for your use case, we invite you to contact our support team at cbrz-support@hsu-hh.de.

Partition Name	Nodes per Job	Wall-clock Limit	Concurent Jobs Limit	Nodes	Remarks
`dev`	1-2	1h	1	1-571	For testing purposes only, max. two queued jobs
`small`	1-5	72h	- (*)	3-571	Regular nodes, exclusive node reservation
`small_shared`	1-5	72h	- (*)	3-571	Same settings as `small` but node resources are by default shared
`small_fat`	1-5	24h	- (*)	572-576	Fat memory nodes, exclusive node reservation
`small_gpu`	1-5	72h	- (*)	gpu 1-5	Up to two A100 GPUs can be allocated per node
`small_gpu8`	1-4	24h	- (*)	gpu 6-9	Up to eight L40S GPUs can be allocated per node
`medium-short`	6-16	12h	20	3-571	Regular nodes, exclusive node reservation
`medium-s`	6-32	24h	5	3-571	Regular nodes, exclusive node reservation
`medium-m`	33-64	24h	3	3-571	Regular nodes, exclusive node reservation
`medium-l`	65-256	24h	1	3-571	Regular nodes, exclusive node reservation
`large`	>256	24h	- (*)	3-571	Regular nodes, exclusive node reservation ! available to selected users only !

(*): HSUper has a general maximum of concurrent jobs per user, which can be found here. Other limitations are specified there as well.

Storage & Quota

Mon, 15 Jul 2024 22:10:10 +0200

Contact support if you have any questions or concerns not addressed here.

BeeGFS

BeeGFS is a parallel file system available on all nodes and mostly used (e.g. for the home directory). Each user has a personal quota, which limits additional files that can be written once reached.

Check your quotas: Your personal quota and group allocations (chair, research, project) are displayed at login.

Size: Total amount of disk space allocation.
Chunk Files: Number of fixed-size chunks (512KB for home directory, else 1MB each) allocated for storing data. In BeeGFS, large files are broken down into these fixed-size chunks to optimize storage performance by enabling parallel transfers to and from servers.

Temporary Files

When using temporary files on HSUper, be mindful of their impact:

Useful Environment Variables

Mon, 15 Jul 2024 22:10:10 +0200

HSUper Specific Variables

$HOME- Path to ones home directory
$PROJECT- Path to the BeeGFS project directory
$SLURM_TMPDIR- Path to the temporary folder for the current slurm job. On job completion all created files will be automatically deleted. Hence, the user’s BeeGFS quota will not be affected by the temporary files that are created in this path, once the slurm job completes. To keep files, essential files should be copied during the jobs runtime (copy commands in the slurm job script).

Access

Mon, 15 Jul 2024 22:10:10 +0200

Please apply for HSUper access using the following form (available on campus/VPN only, login with RZ credentials), if needed.

CBRZ: HSUper Access Registration

Access to HSUper via Web Browser

Navigate to hsuper.hsu-hh.de using your preferred web browser. Log in with your RZ credentials to proceed.

Only users that were authorized to utilize HSUper computing resources are able to login.

Access to HSUper using Linux / Windows 10+ / MacOS 10+

First Steps

Mon, 15 Jul 2024 22:10:10 +0200

To help new users get started with HSUper, we provide a range of educational materials designed to support independent learning and foster essential skills for high performance computing.

We strongly recommend completing our training units as a starting point. These structured courses are tailored to introduce key concepts and guide you step-by-step in effectively using the HSUper system.

Basic Concepts

More than eight digital trainings are available. These trainings are the result of a reimagined HSUper Workshops series, originally offered as a six-part on-site program. Now presented as flexible, self-paced digital trainings, these workshops are designed to empower HSUper users to independently develop the competencies needed to make the most of HPC resources.

Available Software

Mon, 15 Jul 2024 22:10:10 +0200

HSUper has its modules hierarchically organized by compiler and MPI implementation. The command module avail shows all modules that are available in the current environment. If you are missing a software package, it might not be available for the selected compilers or MPI implementation. module spider <modulename> is useful to find out how a module could be loaded. More information on the module system is provided by module --help and the Lmod documentation.

Installing Python Modules

Mon, 15 Jul 2024 22:10:10 +0200

Contemporary research pipelines often employ Python - either for data analysis or even for data generation (e.g., simulation) itself. On HSUper, Python interpreters and corresponding packages can be installed via Spack or via miniforge3, where the latter will be most suitable for typical users. This article briefly describes how to configure miniforge3 on HSUper and provides an example installation of typical deep learning libraries to a user-defined environment.

Please note that Python and pip are already installed on each node, but they’re only available via an alias for the respective Python version, e.g., python3.6 and pip3.6. Be aware of this distinction to avoid confusion with system-wide installations.

USER-SPACK

Mon, 15 Jul 2024 22:10:10 +0200

The module integrates your locally installed spack packages directly into the environment module system Lmod — if compiled with a different compiler than the default system compiler (currently gcc@8.5.0). For example spack install <package> %gcc@12.1.0 installs the < package > using the gcc compiler in version 12.1.0 and creates a module file for you. Not mentioning any compiler (with %) result in using the system compiler and no module file creation. In these cases you rely on ’ spack load. Furthermore, all globally available software packages are made known to your local user spack installation on the first load of the module – no need to recompile these nor make them known to your spack installation. Hence, you are free to choose between spack and module for loading and unloading installed packages.

Preparing Jobs & Testing

Mon, 15 Jul 2024 22:10:10 +0200

Software should be compiled/installed on the login nodes – manually or using USER-SPACK.

Container (for Apptainer/Singularity) may be prepared on your local computer (or the login node) and later uploaded to the cluster.

Jobs running on compute nodes should only contain the execution of the already set up environment. All environment preparations (download of additional software packages and data from the Internet) should be done beforehand.

There is a special dev partition (see Partitions) meant to be used for testing purposes. Feel free to use that partition to test different job settings etc.

How to Submit Jobs

Mon, 15 Jul 2024 22:10:10 +0200

Exemplary Slurm job scripts are provided further below. They can be scheduled using sbatch, for example, with sbatch helloworld-omp.job. To get an overview of your queued and running jobs, you may use squeue --me. For more information on Slurm, please refer to its comprehensive documentation. You can also access the manpages directly on HSUper using commands like man sbatch or man squeue.

If you are new to HPC systems or Linux, consider visiting our First Steps page for a little guidance.

Accounting

Mon, 15 Jul 2024 22:10:10 +0200

Job Output

At the end of each slurm job, some accounting information is appended to the output file:

########################################
# Job Accounting #
########################################
Name : my-job-name
User : myuser
Account : hsuper
Partition : small
QOS : normal
NNodes : 1
Nodes : node0267
Cores : 144 (72 physical)
GPUs : 0
State : COMPLETED
ExitCode : 0:0
Submit : 2024-02-29T14:28:30
Start : 2024-02-29T14:28:31
End : 2024-02-29T17:55:27
Waited : 00:00:01
Reserved walltime : 1-00:00:00
Used walltime : 03:26:56
Used CPU time : 10-07:02:27 (Efficiency: 99.48%)
% User (Computation): 99.91%
% System (I/O) : 0.09%
Mem reserved : 245000M
Max Mem used : 205.48M (node0267)
Max Disk Write : 16.36M (node0267)
Max Disk Read : 1.14M (node0267)
Energy (CPU+Mem) : 2.26kWh (0.95kg CO2, 1.17€)

This output is generated by the jobinfo script.
If a GPU node was used, the statistics from the NVIDIA Data Center GPU Manager are also appended.

Energy Measurement

Mon, 15 Jul 2024 22:10:10 +0200

RAPL (CPU and DRAM Energy)

The Slurm also includes energy measurements using the RAPL interface.¹ The corresponding energy consumption is reported in the job output as shown on the previous page.

Manual Usage

To obtain the raw RAPL measurements, one can utilize the linux Power Capping Framework through the file system. All relevant attributes are located under /sys/devices/virtual/powercap/intel-rapl/ with the following subdirectories mapping to individual RAPL domains:

Support

Mon, 15 Jul 2024 22:10:10 +0200

The administration / technical support can be reached via the ticket system (login with RZ credentials):

Alternatively, send an email to: cbrz-support@hsu-hh.de

Acknowledgement

Mon, 15 Jul 2024 22:10:10 +0200

The HPC cluster HSUper has been provided by the project hpc.bw, funded by dtec.bw — Digitalization and Technology Research Center of the Bundeswehr. dtec.bw is funded by the European Union – NextGenerationEU.

See below for a suggestion of an acknowledgement statement for publications:

Computational resources (HPC cluster HSUper) have been provided by the project hpc.bw, funded by dtec.bw – Digitalization and Technology Research Center of the Bundeswehr. dtec.bw is funded by the European Union – NextGenerationEU.

Technical Specifications

Mon, 15 Jul 2024 22:10:10 +0200

The ISCC cluster consists of 13 hosts in total, with the same hardware specifications as HSUper, except for the interconnect, which is capable of 50Gb/s over Ethernet instead of InfiniBand HDR100.

Regular hosts: 10 hosts each equipped with 256 GB RAM and 2 Intel Icelake sockets; each socket features a Intel(R) Xeon (R) Platinum 8360Y processor with 32 cores, yielding a total of 64 cores per host.
GPU hosts:
- 2 hosts each equipped with 1 TB RAM, 2 Intel Icelake sockets, 8 NVIDIA A30 (24GB) GPUs and 2TB local scratch storage; each socket features a Intel(R) Xeon (R) Platinum 8360Y processor with 32 cores, yielding a total of 64 cores per host

Virtual Machines (VM)

Mon, 15 Jul 2024 22:10:10 +0200

The maximum number of resources a virtual machine can acquire is the same as a single host can provide. Resources can be divided as small as a single core, 4MB of RAM, and a quarter of a single NVidia A30 GPU.

In theory, this allows a GPU host to be sliced into 32 equal virtual machines using a quarter of an NVidia A30, 2 CPU cores and 32GB of RAM.

Storage

Mon, 15 Jul 2024 22:10:10 +0200

Virtual Machines (VMs) generally do not have a local storage, but use storage provided by CEPH over a 50GbE network.

Increasing or decreasing the storage of a VM is generally possible, but sometimes the time and effort does not justify the result. Therefore, it is important to make a good guess when creating the VM or requesting the resources for a VM.

Besides, HSUper’s 1PB BeeGFS file system may be mounted using SSHFS utilizing the 50GbE network connection of the ISCC.

Access to ISCC Resources

Mon, 15 Jul 2024 22:10:10 +0200

The ISCC can be accessed from within the HSU network only. If you are not located at the HSU, you need to connect using the HSU VPN (use an eduVPN client).

In order to use ISCC resources, someone needs to provide you with access information for a VM running in the ISCC.

Please apply for ISCC resources using the following form (available on campus/VPN only, login with RZ credentials, Microsoft’s multi-factor authorisation may intercept your request).

Support

Thu, 22 May 2025 11:17:10 +0200

The administration / technical support can be reached via the ticket system (login with RZ credentials, Microsoft’s multi-factor authorisation may intercept your request):

Alternatively, send an email to: cbrz-support@hsu-hh.de

Acknowledgement

Mon, 15 Jul 2024 22:10:10 +0200

The cloud solution ISCC has been provided by the project hpc.bw, funded by dtec.bw — Digitalization and Technology Research Center of the Bundeswehr. dtec.bw is funded by the European Union – NextGenerationEU.

See below for a suggestion of an acknowledgement statement for publications:

Computational resources (Interactive Scientific Computing Cloud) have been provided by the project hpc.bw, funded by dtec.bw – Digitalization and Technology Research Center of the Bundeswehr. dtec.bw is funded by the European Union – NextGenerationEU.

General Information

Tue, 19 May 2026 22:10:10 +0200

The HPC Portal is designed as an open and collaborative learning environment. Members of the HPC community are encouraged to actively contribute their own learning materials, documentation, tutorials, training concepts, and other educational resources. This documentation explains the technical structure of the platform, the workflow for contributing content, and the available contribution formats.

Technical Foundation of the HPC Portal

The HPC Portal is built using the static website generator Hugo. All website content, configuration files, and documentation are maintained in a central Git repository (Link only works on HSU campus or in HSU VPN). Access to the repository is provided via self-registration in GitLab.

Available Contributions Formats

Tue, 19 May 2026 22:10:10 +0200

The HPC Portal supports several different contribution formats.

1. Blog Entries

Blog posts are suitable for:

project reports,
event summaries,
methodological reflections,
announcements or
short educational articles.

Example: Green HPC (Path in Git-Repository: content/english/blog/green-hpc/index.md)

2. Documentation

Documentation pages provide structured technical or educational information, for example:

software installation guides,
HPC usage tutorials,
workflow explanations or
introductions to computational methods.

Example: Energy Measurement (Path in Git-Repository: content/english/documentation/hsuper/energy.md)

3. Wiki / Docs-Style Trainings

This format is intended for modular self-learning content and step-by-step educational material.

Contribution Workflow

Tue, 19 May 2026 22:10:10 +0200

After editing or creating files locally, contributions must be submitted through the Git workflow.

The general process is:

Clone or download the repository
Edit or create content locally
Create a new branch
Commit changes
Push the branch to the repository
Open a Merge Request
Wait for review and approval

All contributions follow the four-eyes principle and changes are reviewed by authorized maintainers before being merged into the main branch.

Creating a Branch in VSC

The following workflow is recommended when using Visual Studio Code with the integrated Git functionality. The general steps might differ depending on your software, but the principle of order stays the same.

HPC Café

Mon, 01 Jun 2026 16:00:00 +0000

About Event

At the HPC Café, we work together on your HPC-related projects. We offer to solve your problems and issues interactively together, as a hackathon-style spontaneous mini-workshop! Whether it’s about scientific software, code optimization, or parallelization - you name it. Please bring your project or a specific question about Slurm, Spack, R, ANSYS, C++, Matlab, PyTorch or anything else. We will either help you directly or try to connect you with other researchers facing similar challenges.

Seminar Series: Computation & Data

Mon, 01 Jun 2026 16:00:00 +0000

About Event

The goal of the interdisciplinary seminar series Computation & Data at HSU is to bring together researchers and foster exchange on the development of algorithms, methods and software. The seminar series is typically scheduled for the last Wednesday every month, 16:00-17:00, with 1 presentation per hybrid session (digital and at HSU). Immediately after the seminar series, the HPC Café take place.

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

HPC Café

Wed, 01 Apr 2026 16:00:00 +0000

About Event

Seminar Series: Computation & Data

Wed, 01 Apr 2026 16:00:00 +0000

About Event

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

HPC Café

Sun, 01 Mar 2026 16:00:00 +0000

About Event

Seminar Series: Computation & Data

Sun, 01 Mar 2026 16:00:00 +0000

About Event

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

HPC Café

Sun, 01 Feb 2026 16:00:00 +0000

About Event

Seminar Series: Computation & Data

Sun, 01 Feb 2026 16:00:00 +0000

About Event

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

HPC Café

Thu, 01 Jan 2026 16:00:00 +0000

About Event

Seminar Series: Computation & Data

Thu, 01 Jan 2026 16:00:00 +0000

About Event

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

HPC Café

Tue, 14 Oct 2025 16:00:00 +0000

About Event

What is the HPC Café? The goal of this new format is peer-to-peer learning within the community. At each session, an expert from the discipline will be present and available to answer questions. All target groups are warmly invited – both those with prior knowledge in the field of HPC or related disciplines and those who are just starting out.

This collaborative exchange aims to discuss direct questions and application issues in the field of HPC and related disciplines, and to jointly search for solutions. Feel free to bring your questions and concerns as well as a laptop and a cup of coffee with you.

Seminar Series: Computation & Data

Wed, 03 Dec 2025 16:00:00 +0000

About Event

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

HPC@HSU – A short review of the latest HPC Workshop

Thu, 27 Nov 2025 00:00:00 +0100

From October 16 to October 17, 2025, the HPC@HSU Workshop took place at the Helmut-Schmidt-University / University of the Federal Armed Forces Hamburg (HSU/UniBw H), organized by the hpc.bw project. The event represents the third part of HSUper workshop series for knowledge and competence transfer in the field of HPC. While the first two workshops strongly focused on the transfer of knowledge between HPC experts within the hpc.bw project and researchers in the field of HPC, this year’s workshop centered on networking and exchange about similar research challenges resp. questions through peer groups. Moreover, the workshop HPC@HSU conceptually combined offerings for self-organized learning with direct support from experts for joint problem-solving within the research projects. A total of 19 participants and 9 experts took part in the workshop.

HPC Café

Tue, 14 Oct 2025 16:00:00 +0000

About Event

Seminar Series: Computation & Data

Tue, 14 Oct 2025 16:00:00 +0000

About Event

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

HPC Workshop 4 - Advanced Slurm

Mon, 17 Nov 2025 15:27:17 +0600

Welcome to HPC Workshop 4 - Advanced Slurm — an essential part of your journey toward mastering High Performance Computing (HPC) with HSUper!

This training is part of the reimagined HSUper Workshops series, originally offered as a six-part on-site program and now presented as flexible, self-paced digital trainings.

Learning Goals

Job Priorities and Updates: You will learn about what various Slurm messages on HSUper mean, what might be hindering your job from running, and how to update your job to work around the limitations set by HSUper. You will also learn about job priorities, how to check the priorities of your job, and tips on getting your jobs running faster.

HPC Café

Tue, 14 Oct 2025 16:00:00 +0000

About Event

Seminar Series: Computation & Data

Tue, 14 Oct 2025 16:00:00 +0000

About Event

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

HPC Workshop @ HSU

Wed, 01 Oct 2025 09:00:00 +0000

About Event

The dtec.bw-funded project “hpc.bw – Competence Platform for Software Efficiency and Supercomputing” focuses on providing High Performance Computing (HPC) resources such as HSUper or the ISCC at HSU and developing competences for using them. In doing so, we conceptualize and organize various learning opportunities such as e-learning via the HPC Portal or on-site workshops.

This year, we will once again be offering a workshop to which we would like to invite you. The HPC Workshop will take place at HSU on October 16-17, 2025. October 16 will take place in a self-learning format with support from our experts if needed; October 17 will take place in small groups at HSU each with an expert on site. The program will take place on both days between 9 a.m. and 5 p.m.

HPC Workshop 5 - Loading and Installing Software with Spack

Tue, 14 Oct 2025 15:27:17 +0600

Welcome to HPC Workshop 5 - Loading and Installing Software with Spack — an essential part of your journey toward mastering High Performance Computing (HPC) with HSUper!

This training is part of the reimagined HSUper Workshops series, originally offered as a six-part on-site program and now presented as flexible, self-paced digital trainings.

Learning Goals

Spack and Modules: You will learn the difference between Spack and module environments.
Software Installation: You will learn how to install software packages not available yet as a module at HSUper via Spack.

HPC Workshop 6 - Running Graphical Applications with X-Forwarding

Tue, 14 Oct 2025 15:27:17 +0600

Welcome to HPC Workshop 6 - Running Graphical Applications with X-Forwarding — an essential part of your journey toward mastering High Performance Computing (HPC) with HSUper!

This training is part of the reimagined HSUper Workshops series, originally offered as a six-part on-site program and now presented as flexible, self-paced digital trainings.

Learning Goals

Run an application with a graphical user interface (GUI) on HSUper: You will be able to launch applications that require a GUI from the terminal, when connected to HSUper.

HPC Workshop 7 - Best Practices for HPC

Tue, 14 Oct 2025 15:27:17 +0600

Welcome to HPC Workshop 7 - Best Practices for HPC — an essential part of your journey toward mastering High Performance Computing (HPC) with HSUper!

This training is part of the reimagined HSUper Workshops series, originally offered as a six-part on-site program and now presented as flexible, self-paced digital trainings.

Learning Goals

General HPC Best Practices: You will learn about selected topics of HPC practicalities.
Build Systems: Use of build systems as a best practice for project building.

HPC Workshop 8 - Performance Engineering for Python and Deep Learning

Tue, 14 Oct 2025 15:27:17 +0600

Welcome to HPC Workshop 8 - Performance Engineering for Python and Deep Learning — an essential part of your journey toward mastering High Performance Computing (HPC) with HSUper!

This training is part of the reimagined HSUper Workshops series, originally offered as a six-part on-site program and now presented as flexible, self-paced digital trainings.

Learning Goals

Performance Engineering - Optimization: You will learn about Python libraries that you can use to replace traditional data structures and libraries for better performance, as well as about Just-In-Time compilation.

HPC Café

Sun, 01 Jun 2025 16:00:00 +0000

About Event

Seminar Series: Computation & Data

Sun, 01 Jun 2025 16:00:00 +0000

About Event

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

GreenHPC

Thu, 22 May 2025 00:00:00 +0200

Despite the global challenges posed by climate change and emissions reductions in various industries in recent years, the ever growing demand for computation has ensured the opposite effect in this field as the computational capacity and therefore the associated energy requirement keeps growing¹. Meanwhile, recent legislation introduced explicit responsibilities to avoid or re-use waste heat². This amplifies the need for improving sustainability of (scientific) computing. GreenHPC is only one part of GreenIT which comprises efforts to increase the energy efficiency across all domains of computation, from smartphones to desktops, industrial applications and the cloud. With the Software Carbon Intensity (SCI) score, the aims to provide a standardized³ metric to reduce carbon emissions. The actual formula $SCI=(O+M)/R$ accounts for both, the operational emissions $O$ as well as the embodied emissions $M$, per functional unit $R$. The operational emissions are just the product of energy consumed and region-specific carbon intensity while the embodied emissions capture the share of total emissions for the hardware components themselves.

HPC Café

Thu, 01 May 2025 16:00:00 +0000

About Event

Seminar Series: Computation & Data

Thu, 01 May 2025 16:00:00 +0000

About Event

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

'Data Toxicality': The hidden harms of information

Thu, 01 May 2025 00:00:00 +0200

On February 27th, the Computation & Data seminar series welcomed Prof. Gerhard Schreiber for a thought-provoking talk on ‘Data Toxicality: Observations and Reflections from a Techno-Ethical Perspective.’¹ In his presentation, Prof. Schreiber introduced a compelling and originative concept - data toxicality - which extends the notion of toxicity beyond its traditional biochemical roots into the digital realm.

But what exactly does the concept of the “data toxicality”, coined by Prof. Schreiber mean, and why should we be concerned about it?

HPC Café

Tue, 01 Apr 2025 16:00:00 +0000

About Event

Seminar Series: Computation & Data

Tue, 01 Apr 2025 16:00:00 +0000

About Event

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

Seminar Series: Computation & Data

Sat, 01 Mar 2025 16:00:00 +0000

About Event

The goal of the interdisciplinary seminar series Computation & Data at HSU is to bring together researchers and foster exchange on the development of algorithms, methods and software. The seminar series is typically scheduled for the last Wednesday every month, 16:00-18:00, with 1-2 presentations per hybrid session (digital and at HSU).

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

HPC@dtec.bw - Perspectives & Next Horizons

Sat, 01 Feb 2025 16:00:00 +0000

About Event

with the new year, we will invite once again to the wrap-up and kick-off meeting, ‘HPC@dtec.bw: Perspectives & Next Horizon’, on March 3-4, 2025 at HSU. The goal is to present and reflect on the results and activities of the first funding phase on one hand, and to present plans and research horizons for the second funding phase on the other.

The first day (the kick-off) is open to the public. You are welcome to the event. If this includes individuals who are not affiliated with the universities of the German Armed Forces, we kindly request that they also register by February 7, 2025. Moreover, on the first day, there will be a poster session, which we cordially invite you and HSUper and ISCC users to attend. You can find the relevant information on our (website)[https://www.hsu-hh.de/hpccp/hpcdtec-bw-perspectives-and-next-horizon/]

Seminar Series: Computation & Data

Sat, 01 Feb 2025 16:00:00 +0000

About Event

The goal of the interdisciplinary seminar series Computation & Data at HSU is to bring together researchers and foster exchange on the development of algorithms, methods and software. The seminar series is typically scheduled for the last Wednesday every month, 16:00-18:00, with 1-2 presentations per hybrid session (digital and at HSU).

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

Seminar Series: Computation & Data

Wed, 01 Jan 2025 16:00:00 +0000

About Event

The goal of the interdisciplinary seminar series Computation & Data at HSU is to bring together researchers and foster exchange on the development of algorithms, methods and software. The seminar series is typically scheduled for the last Wednesday every month, 16:00-18:00, with 1-2 presentations per hybrid session (digital and at HSU).

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

Peer-based Learning in the HPC Café

Wed, 01 Jan 2025 00:00:00 +0100

For the last two year, the Computation & Data Seminar Series has served as a space for knowledge exchange on topics related to High Performance Computing (HPC). However, participation patterns and logistical challenges have highlighted the need for a fresh approach. Starting in the second quarter of 2025, the format will undergo a transformation to create a more interactive and engaging learning environment.

The Original Concept: Challenges and Limitations

The seminar series, organized as part of the HPC initiative, has traditionally followed a structured format: two invited speakers present their research on HPC-related topics, followed by a brief Q&A session. Meetings take place once a month and typically attract a small but engaged audience.

Seminar Series: Computation & Data

Sun, 01 Dec 2024 16:00:00 +0000

About Event

The goal of the interdisciplinary seminar series Computation & Data at HSU is to bring together researchers and foster exchange on the development of algorithms, methods and software. The seminar series is typically scheduled for the last Wednesday every month, 16:00-18:00, with 1-2 presentations per hybrid session (digital and at HSU).

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

HSUper Workshops - a Peer-to-Peer approach

Sun, 01 Dec 2024 00:00:00 +0100

The HSUper Workshop is undergoing a transformation. Designed to equip researchers with the skills needed to effectively utilize HPC resources, the updated format reflects a shift toward flexibility, personalization, and efficiency. Instead of a traditional lecture-based approach, the workshop will now emphasize self-directed learning, peer-to-peer support, and expert consultation, making it more relevant and impactful for its diverse audience.

Tailored to Different Experience Levels

High-Performance Computing (HPC) is used across disciplines, but not everyone approaches it with the same background. The reimagined HSUper Workshop recognizes this by catering to three key groups:

Seminar Series: Computation & Data

Fri, 01 Nov 2024 16:00:00 +0000

About Event

The goal of the interdisciplinary seminar series Computation & Data at HSU is to bring together researchers and foster exchange on the development of algorithms, methods and software. The seminar series is typically scheduled for the last Wednesday every month, 16:00-18:00, with 1-2 presentations per hybrid session (digital and at HSU).

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

Seminar Series: Computation & Data

Tue, 01 Oct 2024 16:00:00 +0000

About Event

The goal of the interdisciplinary seminar series Computation & Data at HSU is to bring together researchers and foster exchange on the development of algorithms, methods and software. The seminar series is typically scheduled for the last Wednesday every month, 16:00-18:00, with 1-2 presentations per hybrid session (digital and at HSU).

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

Seminar Series: Computation & Data

Sat, 01 Jun 2024 16:00:00 +0000

About Event

The goal of the interdisciplinary seminar series Computation & Data at HSU is to bring together researchers and foster exchange on the development of algorithms, methods and software. The seminar series is typically scheduled for the last Wednesday every month, 16:00-18:00, with 1-2 presentations per hybrid session (digital and at HSU).

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

Seminar Series: Computation & Data

Wed, 01 May 2024 16:00:00 +0000

About Event

The goal of the interdisciplinary seminar series Computation & Data at HSU is to bring together researchers and foster exchange on the development of algorithms, methods and software. The seminar series is typically scheduled for the last Wednesday every month, 16:00-18:00, with 1-2 presentations per hybrid session (digital and at HSU).

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

Seminar Series: Computation & Data

Mon, 01 Apr 2024 16:00:00 +0000

About Event

The goal of the interdisciplinary seminar series Computation & Data at HSU is to bring together researchers and foster exchange on the development of algorithms, methods and software. The seminar series is typically scheduled for the last Wednesday every month, 16:00-18:00, with 1-2 presentations per hybrid session (digital and at HSU).

Feel free to subscribe the seminar newsletter by sending an e-mail to info-hpc-bw@hsu-hh.de with the subject line „Subscription Seminar Computation & Data”.

HSUper Workshop for Advanced User

Fri, 19 Apr 2024 09:00:00 +0000

About Event

We cordially invite you to the HSUper workshop for Beginners on 18.04.2024 and to the HSUper workshop for advanced users on 19.04.2024 to gain a comprehensive insight into HPC and an introduction to using the supercomputer HSUper!

You are welcome to attend the workshops and to learn along theoretical insights as well as practical examples more about the broad world of HPC. The workshop will take place on-site at the HSU.

HSUper Workshop for Beginners

Thu, 18 Apr 2024 09:00:00 +0000

About Event

You are welcome to attend the workshops and to learn along theoretical insights as well as practical examples more about the broad world of HPC. The workshop will take place on-site at the HSU.

Seminar Series: Computation & Data

Fri, 01 Mar 2024 16:00:00 +0000

About Event

The project hpc.bw establishes the seminar series “Computation & Data” at HSU. The goal of this interdisciplinary seminar is to bring together researchers and foster exchange on the development of algorithms, methods and software. The seminar series is scheduled for the last Wednesday every month, with two presentations per hybrid session.

Sebastian Brandstäter (UniBw M): Sensitivity Analysis for Biomechanical Models

Uncertainty plays a crucial role in active biomechanical systems, primarily due to their inherent high natural variability. This variability should be considered in any biomechanical modelling endeavour, but particularly when the models are used in clinical decision-making. Most notably, biomechanical models incorporate numerous parameters with imprecise values, leading to uncertainty propagation through the models. Ultimately, this results in uncertain model predictions. Global sensitivity analysis aims to analyse and quantify how much of the prediction uncertainty can be attributed to the individual model input parameters and their interactions. The talk will introduce global sensitivity analysis methods applicable to studying complex, large-scale computational models such as finite-element models of active biomechanical systems. A focus will be placed on computational aspects, such as making the computational burden tractable with data-driven surrogate models. The practical utility of these approaches in studying the biomechanical systems will be demonstrated through examples, including a model of growth and remodelling during the formation of an abdominal aortic aneurysm, drug delivery to a tumour, and the electromechanics of gastric peristalsis.

Some simple Rules to save Computation Time in Python

Fri, 01 Mar 2024 00:00:00 +0100

Let’s discuss the following python code. It is a simple piece of code one could come up to solve a minor daily task. This script has been written to conduct a simple grid search over a 4D parameter space, it compute the values and stores parameter combinations of interest. It is not wrong to write it that way and it works, a classic throw away code. The story changes when you want to scan big parameter spaces. You will realize that your program will need a increasing time to run. What to do now? A simple solution would be to parallelize it. But! The first step before parallelization should always be a optimization of your program!

Seminar Series: Computation & Data

Thu, 01 Feb 2024 16:00:00 +0000

About Event

Ali Khalifa: Neural Network-Based Multiscale Modeling of Deagglomeration due to Wall Impact and Collisions

Predicting the evolution of micron-sized particle system in fluid flows is crucial for natural processes and industrial applications like pharmaceuticals. Fully-resolved simulations are costly due to scale variations. Hence, data-driven approaches offer an attractive alternative. This study enhances an Euler-Lagrange method with neural-network models for deagglomeration from collisions and wall impacts, integrated into LES-based simulations. Tested in various turbulent flows, such as funnel-duct dispersers and bend pipes, this approach provides cost-effective predictions.

Code Optimization

Mon, 12 Feb 2024 15:27:17 +0100

Code optimization is an important aspect of software development, focusing on improving the efficiency, speed, and resource utilization of computer programs. Through careful analysis and refinement, developers aim to enhance the performance of their software, resulting in faster execution times, reduced memory footprint, and improved energy efficiency.

The Importance of Optimization

Optimization is crucial in various domains, including scientific computing, web development, and embedded systems. In scientific computing, optimised algorithms enable researchers to analyse vast datasets, simulate complex phenomena, and solve mathematical problems more efficiently. In environments where resources are often limited, optimization is essential to maximise performance while minimizing resource usage. In web development, optimized code leads to faster loading web pages and therefore better user experience while in scientific research code optimiziation can reduce the run time of expensive simulations from days to minutes.

Fluid Mechanics

Mon, 12 Feb 2024 15:27:17 +0100

Fluid mechanics is a branch of physics and engineering that studies the behavior of fluids, including liquids, gases, and plasmas. It seeks to understand how fluids move, interact with each other and with solid bodies, and respond to external forces, providing insights into a wide range of natural and engineered systems.

Fundamental Principles

At its core, fluid mechanics is regulated by fundamental principles, including conservation of mass, conservation of momentum, and conservation of energy. These principles are used in many mathematical models such as the Navier-Stokes equations. They provide a tool for analyzing fluid flow phenomena and predicting fluid behavior under various conditions.

Machine Learning

Mon, 12 Feb 2024 15:27:17 +0100

Machine learning (ML) focuses on the development of algorithms and statistical models that enable computers to learn from and make predictions or decisions based on data. It encompasses a diverse range of techniques and methodologies aimed at extracting meaningful patterns and insights from complex datasets.

Fundamental Principles

At the heart of machine learning are fundamental principles drawn from statistics, probability theory, optimization, and computer science. These principles provide the theoretical foundation for various machine learning algorithms and approaches, including supervised learning, unsupervised learning, and reinforcement learning. By leveraging these principles, machine learning algorithms can iteratively improve their performance and adapt to new data over time.

Materials Science

Mon, 12 Feb 2024 15:27:17 +0100

Materials science is a field of scientific study that explores the properties and behavior of various materials, such as metals, ceramics, polymers, and composites. It investigates how these materials are structured at the atomic and molecular levels and how their structure affects their mechanical, electrical, and thermal properties.

In materials science, researchers aim to develop new materials with improved properties for specific applications, such as stronger and lighter alloys for aerospace components, more efficient semiconductors for electronics, or biocompatible materials for medical implants.

Mechanical Engineering

Mon, 12 Feb 2024 15:27:17 +0100

Mechanical engineering is a multidisciplinary field that focusses on the design, analysis, and manufacturing of mechanical systems and devices, ranging from automobiles and aircraft to robots and renewable energy technologies. With its roots deeply entrenched in science, mathematics, and engineering principles, mechanical engineering drives innovation and technological advancement across various disciplines, shaping the world we live in and paving the way for a sustainable future.

Designing Innovative Solutions

At the heart of mechanical engineering lies the creative process of designing innovative solutions to complex engineering challenges. Mechanical engineers apply principles of physics, materials science, and mathematics to conceptualize, develop, and optimize mechanical systems and components that meet specific performance criteria and requirements. Whether it is designing more fuel-efficient engines, developing advanced materials for aerospace applications, or creating next-generation medical devices, mechanical engineers play a pivotal role in transforming ideas into reality through the application of scientific principles and engineering expertise.

Adult Education

Mon, 12 Feb 2024 15:27:00 +0100

Adult education is a diverse and dynamic field that encompasses a wide range of learning opportunities and programs designed to meet the needs and interests of adult learners. From formal classroom-based courses to informal learning experiences, adult education plays a crucial role in supporting lifelong learning and personal development for individuals of all ages.

Empowering Adult Learners

Adult education aims to empower learners by providing them with the knowledge, skills, and competencies needed to succeed in various aspects of life, including their careers, personal relationships, and civic engagement. It recognizes that adults have unique learning needs and preferences, often shaped by their life experiences, responsibilities, and goals. As such, adult education programs strive to be flexible, accessible, and relevant to the diverse backgrounds and circumstances of adult learners.

Insights into the first HSUper workshop

Thu, 01 Feb 2024 00:00:00 +0100

With great pleasure we provide insights into the results of the questionnaire following our recent HSUper workshop on 28.09.2023. We received a total of 29 completed forms and we will use the results to expand our offerings in the HPC Portal and to improve future workshops.

Scientific Disciplines

In Figure 1, you can see the distribution of scientific disciplines among the 29 participants. This reveals a broad range of disciplines, each with a different focus. These include for example engineering with focus on structural optimization or geodesy, data science such as statistics as well as computer science, especially cloud computing and neural networks. Moreover, there were also participants from the social sciences such as political science, economics in relation to operational research as well as cognitive neuroscience.

Welcome to the HPC Portal

Mon, 01 Jan 2024 00:00:00 +0100

We are delighted to announce the official launch of the High Performance Computing (HPC) Portal. It is an e-learning platform designed to target to the various needs of various target groups in the field of HPC. This platform is designed to offer learning materials and training resources.

HPC Training:

The HPC Portal will be designed to support users to tailor their learning experience according to their individual needs. Whether one is a novice seeking fundamental insights or a professional delving into advanced software engineering, the portal will accommodate multiple levels of expertise. The content encompasses various formats, including cheat-sheets, selflearning assessments, Open Educational Resources (OER), videos, instruction papers, technical documentation and more.

HPC Workshop 1 - A short introduction to HSUper and the Linux Terminal

Sun, 01 Jan 2023 15:27:17 +0600

Welcome to HPC Workshop 1 - A short introduction to HSUper and the Linux Terminal - the first step in your journey to mastering High Performance Computing (HPC) with HSUper! This training is part of a series of self-paced digital trainings designed to help you build the skills needed to effectively utilize HPC resources on HSUper at the HSU.

This training is part of the reimagined HSUper Workshops series, originally offered as a six-part on-site program and now presented as flexible, self-paced digital trainings.

HPC Workshop 2 - Data Transfer and System Resources

Sun, 01 Jan 2023 15:27:17 +0600

Welcome to HPC Workshop 2 - Data Transfer and System Resources — your next step in mastering High Performance Computing (HPC) with HSUper! This training builds on the foundations established in Workshop 1 - A short Introduction to HSUper and the Linux Terminal and focuses on essential skills for effectively managing data and understanding system resources.

This training is part of the reimagined HSUper Workshops series, originally offered as a six-part on-site program and now presented as flexible, self-paced digital trainings.

HPC Workshop 3 - Module Systems and Slurm Jobs

Sun, 01 Jan 2023 15:27:17 +0600

Welcome to HPC Workshop 3 - Module Systems and Slurm Jobs — an essential part of your journey toward mastering High Performance Computing (HPC) with HSUper! This workshop introduces you to two critical aspects of HPC: managing software environments with module systems and efficiently submitting and managing tasks using the Slurm workload manager.

This training is part of the reimagined HSUper Workshops series, originally offered as a six-part on-site program and now presented as flexible, self-paced digital trainings.

HSUper and Gurobi

Sun, 01 Jan 2023 15:27:17 +0600

This training provides comprehensive guidance on obtaining, installing, and running the Gurobi software on HSUper. It covers essential steps for software installation, job submission using SLURM scripts, running the software, and parameter configuration and tuning. Hence, this course is an invaluable resource for users aiming to effectively utilize the Gurobi solver on HSUper.

Requirements

Access to HSUper

Content

Gurobi Hardware and Performance

This guide discusses how the performance of the Gurobi solver and the influence of the hardware.

Gurobi Installation on HSUper

This guide provides detailed instructions on installing the Gurobi software on HSUper, including starting a Gurobi Interactive Shell.

Gurobi Solver: Running, Configuration, Tuning

This guide provides detailed instructions on running, configuring and tuning the Gurobi solver on HSUper.

IBM ILOG CPLEX Optimization Studio on HSUper

Sat, 06 Jul 2019 15:27:17 +0600

This training provides comprehensive guidance on obtaining, installing, and running the IBM ILOG CPLEX Optimization Studio software on HSUper. It covers essential steps for software installation, job submission using SLURM scripts, running the software via oplrun, cplex, or cpoptimizer, and parameter configuration. Additionally, the training includes detailed instructions for exporting MPS files and fine-tuning optimization settings, making it an invaluable resource for users aiming to effectively utilize IBM ILOG CPLEX Optimization Studio on HSUper.

Alexander Kolling

Mon, 01 Jan 0001 00:00:00 +0000

Alexander Kolling has been employed as a research assistant at the Chair of Continuing Education and Lifelong Learning, as well as at the Chair of High-Performance Computing (HPC) at the HSU, since May 2023. Currently, he is responsible for establishing the digital HPC Portal of the dtec.bw-funded project hpc.bw, as well as for creating the platform’s content.

Furthermore, he is currently researching as part of his PhD in the field of Transdisciplinarity and Transversal Research, employing a scientific-theoretical approach.