Run Your First Simulation

This tutorial will show you how to run OpenSees simulations using the Inductiva API.

This tutorial will cover the SmallMP use case of the OpenSees Examples, available in the OpenSees GitHub repository, to help you get started with simulations.

We will also demonstrate Inductiva’s ability to efficiently scale this use case on a more powerful machine.

Prerequisites

Download the required files here and place them in a folder called SmallMP. Then, you’ll be ready to send your simulation to the Cloud.

Running an OpenSees Simulation

Here is the code required to run OpenSees simulation using the Inductiva API:

"""OpenSees Simulation."""
import inductiva

# Allocate a machine on Google Cloud Platform
cloud_machine = inductiva.resources.MachineGroup( \
    provider="GCP",
    machine_type="c2d-highcpu-4",
	spot=True)

# Initialize the Simulator
opensees = inductiva.simulators.OpenSees( \
    interface="tcl",
    version="3.7.1")

# Run simulation
task = opensees.run( \
    input_dir="/Path/to/SmallMP",
    sim_config_filename="Example.tcl",
    on=cloud_machine)

# Wait for the simulation to finish and download the results
task.wait()
cloud_machine.terminate()

task.download_outputs()

task.print_summary()

In this basic example, we’re using a cloud machine (c2d-highcpu-4) equipped with 16 virtual CPUs. For larger or more compute-intensive simulations, consider adjusting the machine_type parameter to select a machine with more virtual CPUs and increased memory capacity. You can explore the full range of available machines here.

Note: Setting spot=True enables the use of spot machines, which are available at substantial discounts. However, your simulation may be interrupted if the cloud provider reclaims the machine.

To adapt this script for other OpenSees simulations, replace input_dir with the path to your OpenSees input files and set the sim_config_filename accordingly.

When the simulation is complete, we terminate the machine, download the results and print a summary of the simulation as shown below.

Task status: Success

Timeline:
	Waiting for Input         at 21/04, 16:02:53      1.589 s
	In Queue                  at 21/04, 16:02:55      34.667 s
	Preparing to Compute      at 21/04, 16:03:29      4.888 s
	In Progress               at 21/04, 16:03:34      27.232 s
		└> 27.087 s        /opt/openmpi/4.1.6/bin/mpirun --use-hwthread-cpus OpenSeesMP Example.tcl
	Finalizing                at 21/04, 16:04:01      1.139 s
	Success                   at 21/04, 16:04:03      

Data:
	Size of zipped output:    19.60 MB
	Size of unzipped output:  49.79 MB
	Number of output files:   120

Estimated computation cost (US$): 0.00031 US$

As you can see in the “In Progress” line, the part of the timeline that represents the actual execution of the simulation, the core computation time of this simulation was approximately 27.2 seconds.

Although it’s short, there’s still room for improvement to reduce the processing time.

Scaling Up Your Simulation

Scaling up your simulation is as simple as updating the machine_type parameter to a 16 vCPU machine (c2d-highcpu-16).

By increasing the number of vCPUs, we’ve reduced the processing time from 27.2 to 9.4 seconds.

Here are the results of running the same simulation on a few machines:

Machine Type	Virtual CPUs	Execution Time	Estimated Cost
c2d-highcpu-4	4	27.2 seconds	0.00031 US$
c2d-highcpu-8	8	14.2 seconds	0.00034 US$
c2d-highcpu-16	16	9.4 seconds	0.00046 US$

Still in the testing phase? No problem! Just skip this step for now and start with a machine with fewer vCPUs. Once you’re satisfied with your results, you can seamlessly scale your OpenSees simulation.

Run OpeenSeesPy

To run OpenSees scripts written in Python, all you need to do is change the interface parameter to python to match the file type of your OpenSeesPy use case.

It’s that simple!