An Overview of Hardware Implementations of P Systems
Implementing the P systems on parallel hardware is a research highlight in bio-inspired computing since the membrane computing is a large-scale parallel computing paradigm which have a potential to tremendously speed up the computation. Field-programmable gate arrays (FPGAs) and CUDA-enabled GPUs ar...
| Autores: | , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión enviada para evaluación y publicación |
| Fecha de publicación: | 2020 |
| País: | España |
| Institución: | Universidad de Sevilla (US) |
| Repositorio: | idUS. Depósito de Investigación de la Universidad de Sevilla |
| OAI Identifier: | oai:idus.us.es:11441/107616 |
| Acceso en línea: | https://hdl.handle.net/11441/107616 |
| Access Level: | acceso abierto |
| Palabra clave: | Membrane Computing P systems Hardware implementations FPGA CUDA |
| Sumario: | Implementing the P systems on parallel hardware is a research highlight in bio-inspired computing since the membrane computing is a large-scale parallel computing paradigm which have a potential to tremendously speed up the computation. Field-programmable gate arrays (FPGAs) and CUDA-enabled GPUs are the primary hardware which is employed to implement P systems. FPGA-based hardware implementations use different strategies considering regions or evolution rules as processing units. This implies the existence of several parallel architectures for FPGAs specially designed to implement P systems. In contrast, the CUDA-enabled GPUs are a pre-defined parallel platform and numerous types of P systems are directly implemented on it. The object distribution problem (choosing which rules will be applied) is the core problem of all hardware implementations. This problem is particularly difficult, because in the general case the model of P systems is non-deterministic and maximally parallel, hence the corresponding problem is NP-hard. Several heuristics were proposed in order to accelerate the process of the computation of the corresponding ruleset. In this article we overview different approaches and designs for hardware implementations of P systems as well as corresponding solutions to the object assignment problem. |
|---|