Posted: February 25, 2013

# Introducing FMQ

## What is FMQ?

FMQ is Xogeny's first product. It is a cloud-based system for simulating FMI compliant models.

Using FMQ, users can dispatch large numbers of simulations to the cloud. Because we use a scalable cloud computing platform behind the scenes, we can support running large numbers of concurrent simulations which means much faster turn around times for large batches than conventional desktop solutions.

The complexity of queuing, scaling and load-balancing simulation jobs is completely handled behind the scenes. This makes the entire process very simple for the user.

FMQ is just one component in our overall vision for enhancing engineering through web and cloud-based capabilities.

## Architecture

FMQ is available in both public cloud and private cloud versions. Today we are launching a private beta for the public cloud version. But FMQ will also be available in a private cloud version that can be hosted behind a firewall using private computing resources.

There are currently two ways to access FMQ. The first is through a web services API. This API can be used by nearly any modern language to dispatch simulation jobs to the cloud. We have also developed a native Python API, called pyfmq, that provides an easy way to use FMQ in scripts to support large scale analyses.

Here is an example of using pyfmq to run a batch of simulations with different parameter values:

import pyfmq

pyfmq.setKeys("<your public key>", "<your private key>")

# This registers the FMU file with the FMQ service
fmu = pyfmq.register("./bouncingBall.fmu")

results = []
for i in range(1,10):
promise = pyfmq.sim(fmu, sim_args={"final_time": 10.0},
params={"e": 0.5+0.02*i},
signals=["h", "v"])
results.append(promise)

for r in results:
print r.get()


## Pricing

During the beta period, use of the service will be free. If you are interested in participating in the free private beta, just drop us a note.

## Conclusion

As I mentioned at the outset, FMQ is just the first piece in the puzzle. We hope to start unveiling other capabilities in the near future as well.

### Share your thoughts

comments powered by Disqus