Below the cut, you will find the first screencast demo of our web-based simulation creator, manager, and analysis tool. In the video, three simulations are run. In the first, a plane wave is initiated moving across a sheet. In the second, a stimulus is applied behind the tail of the plane wave to start a re-entrant (spiral) wave. Finally, the spiral wave is allowed to stabilize for 200 ms.
The simulation manager (CSM) and backend are capable of much more sophisticated things, but this demo highlights how quick and easy it can be to get a simulation up and running.
For more information on the simulator, manager, or CardioSolv, please contact firstname.lastname@example.org or +1 (651) 967-7961.
We have just posted a summary of the various types of cardiac simulation and model development services that we offer, which you can find on the Services page. We have three different self-service simulation avenues, including one that integrates with your existing cluster(s), and one that uses the recently-announced Penguin On-Demand cloud HPC resource.
If you would like to try a demo account, please don’t hesitate to contact us at email@example.com or +1 (651) 967-7961.
We now have a demo of our web interface on which you can run some small models. If you’d like a demo account, please send an email to firstname.lastname@example.org.
The whole article is here.
The HPC service lets the small, five-employee company do the heavy lifting that would otherwise cost a fortune. “With what we could purchase out of pocket, we’d have to bootstrap very slowly, or look for VC [venture capital] funding,” said Dr. Brock Tice, the vice president of operations at Cardiosolv, a privately funded medical research firm. Instead, Tice uses a new HPC on-demand service from Penguin Computing called Penguin on Demand.
While Cardiosolv has its own small cluster on the premises for calculations, Tice estimates the resources he rents from Penguin would probably cost $500,000 to build, and other cloud options weren’t suitable.
“We can’t use [Amazon’s Elastic Compute Cloud] EC2, since there’s a lot of latency between the nodes,” he said.
The article is primarily about the Penguin On Demand service, which we are using, but there’s a little information about CardioSolv. You can find the story here.
We have been working with Penguin Computing to get our simulation software up and running on their new Penguin On Demand service. They have just announced it officially via a press release.
Cardiac electrophysiology simulation requires fast, low latency communication between processors, something that you won’t find in your garden-variety cluster solution, and something that’s very expensive up-front when building your own cluster. This service will allow us access to a high-performance system with a low-latency interconnect. It will integrate nicely as one option among three for making our services available. The other two options are the use of a shared, high-performance multiprocessor server, or a dedicated appliance that we can integrate with your existing cluster resources. These will be detailed on this web site shortly.
When modeling cardiac tissue, there are two dominant approaches: bidomain, in which both the intra- and extracellular spaces of the tissue are represented, and monodomain, in which only the intracellular space is represented. The bidomain approach is essential when the effects of stimuli, bath, or extracellular conductivities need to be considered. However, when the primary interest has to do only with wave propagation, monodomain is often sufficient.
Given bidomain’s greater realism, one might think that it should always be used. However, it comes with a much greater computational cost; bidomain simuluations are much slower than otherwise-identical monodomain simulations. Therefore, it’s important when setting up cardiac electrophysiology simulations to carefully consider which approach will work best.
Our simulator and web interface give you the choice. By simply checking a box, you can switch your simulation from monodomain to bidomain and back. You can also compromise by setting an option that will primarily use monodomain, but will use bidomain every few steps. This has minimal effects on the outcome of the simulation, but can reduce simulation time noticeably when compared with a full bidomain run. We’d be happy to discuss your simulation problem with you, and help you determine which type of cardiac simulation is appropriate for solving it. You can find our contact information here.