Cardiac Arrhythmia Research Package (CARP)
The CARP simulator comprises a validated ionic model library with a single-cell simulation tool (LIMPET and ‘bench’, respectively), and a tissue simulator. In particular, the Luo-Rudy, Shiferaw-Mahajan, and ten Tusscher I and II ionic models have been used extensively and validated quite thoroughly. The single-cell simulation tool is a Linux/UNIX command-line program and can be used to quickly run a number of simulation types. All of the models in the LIMPET library may be loaded into the CARP simulator for full 3D tissue simulation.
The CARP simulator will run on anywhere from 1 to at least 256 CPUs, using the MPI parallelization library. It can handle models from the size of one cell through millions of cells, and has been used to simulate the human heart. Like LIMPET, it is designed to run on Linux/UNIX systems, and scales very well on low-latency interconnects such as Infiniband and Myrinet. CARP is currently used by several of the leading academic labs in the field of cardiac simulation, and has also been used for commercial cardiac simulation.
Tarantula (3D meshing tool)
CaridoSolv resells and supports the Tarantula meshing tool, developed by CAE Software Solutions. It is a sibling to CAE’s Spider simulator, which has been used extensively for industrial modeling applications. Tarantula has been developed under extensive use and feedback from the founders of CardioSolv, and is ideally suited for producing tissues meshes to be used with the CARP simulator.
Tarantula is currently available for commercial, government, and academic licensing. Please contact us for further information.
We have two types of services available — bespoke model development and consulting (including the running of simulations), and self-serve simulation. The first type will require a thorough discussion of your needs. If you’re looking for custom model development, consulting, or simulation services, please contact us.
Our self-serve simulation services are available via three avenues:
- Shared Dedicated Server – We have dedicated servers running our web interface and simulation software. An account will be created for you on one of these servers. Other users may have accounts on the server. Your jobs are run in a single queue, so you’re limited to one job running at a time (though you can queue them up), and if other users have jobs in the queue you may have to wait a little while. This is our most economical option and is suitable for infrequent simulations and/or small models.
- On-Demand Supercomputing – Via our partnership with Penguin On Demand, you will be able to submit jobs to a large supercomputing cluster. Multiple simulations and analyses can be run simultaneously on this resource. It is unlikely at this point that you will have to wait for your job to run unless you create quite a few simulations and queue them up. This option is somewhat more expensive than the shared dedicated server, because we must assess a per-job fee in addition to the basic service package.
- Remora Cluster Add-On – If you have your own supercomputing resource, and wish to use our services on that resource, we provide a pre-configured 1U rackmount server that we will integrate with your cluster. The Remora add-on contains everything you need to run simulations on your own resource. You will need to have appropriate facilities for the integration of the server.
If you have any questions about these services, please contact us. Demo accounts are available if you’d like to try out the interface and simulator.