Simulating cardiac dynamics using Maxeler dataflow super-computing

Abstract

Nowadays, computer-based simulation is an important tool to study various phenomena in cardiac biology ranging from investigating the properties of single ion channels to predict the onset of cardiac arrhythmia. Due to the complexity of calculations, these simulations are generally computationally expensive and resource intensive. In the last decade, there has been a great effort to accelerate computer-based cardiac simulation by implementing efficient parallel algorithms that leverage multi-cores CPU and many-cores GPUs currently available also in common personal computers. These algorithms are generally developed using imperative languages dictating the control flow of the program. In this work I am investigating the use of hardware-based accelerators provided by Maxeler Dataflow Technology to improve the simulation time of the electrical activity of a homogenous 1D cell cable and an isolated single cell. This technology relies on dataflow paradigm, in which the data processing elements, called pipelines, operate concurrently and are connected in a way that the output of one element is the input for the following/next one. In my work I am additionally comparing the advantages and disadvantages of the proposed approach with the current state-of-the-art based on CPU and GPU regarding performance and resource utilization.