GPUPeP: Parallel Enzymatic Numerical P System simulator with a Python-based interface

Raghavan, Sindhu; Shanthanu, S.; P, Rohit M; Chandrasekaran, K.

doi:10.1016/j.biosystems.2020.104186

Cited by 7 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GPUs are low-cost, low-power consumption, and high performance concerning conventional multiprocessors. Many current desktop computers have equipped with the GPU enable graphics cards, which can improve the performance of processing without additional costs [28,45]. Thus achieving speed up even around 5× by GPU can be valuable work.…”

Section: Comparison Between Previous and Proposed Methodsmentioning

confidence: 99%

“…However, thousands of threads able to be executed in parallel using one low-cost platform of GPU. Different variants of P systems have been simulated on GPU [27], including enzymatic numerical P system [28], spiking neural P system [3], population dynamics P systems [29], etc. This study used active membrane systems, another variant of the membrane system for GPU simulation with Compute Unified Device Architecture (CUDA).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Representation of Membrane Computing with a Clustering Algorithm on the Graphical Processing Unit

Muniyandi

Maroosi

2020

Processes

View full text Add to dashboard Cite

Long-timescale simulations of biological processes such as photosynthesis or attempts to solve NP-hard problems such as traveling salesman, knapsack, Hamiltonian path, and satisfiability using membrane systems without appropriate parallelization can take hours or days. Graphics processing units (GPU) deliver an immensely parallel mechanism to compute general-purpose computations. Previous studies mapped one membrane to one thread block on GPU. This is disadvantageous given that when the quantity of objects for each membrane is small, the quantity of active thread will also be small, thereby decreasing performance. While each membrane is designated to one thread block, the communication between thread blocks is needed for executing the communication between membranes. Communication between thread blocks is a time-consuming process. Previous approaches have also not addressed the issue of GPU occupancy. This study presents a classification algorithm to manage dependent objects and membranes based on the communication rate associated with the defined weighted network and assign them to sub-matrices. Thus, dependent objects and membranes are allocated to the same threads and thread blocks, thereby decreasing communication between threads and thread blocks and allowing GPUs to maintain the highest occupancy possible. The experimental results indicate that for 48 objects per membrane, the algorithm facilitates a 93-fold increase in processing speed compared to a 1.6-fold increase with previous algorithms.

show abstract

Section: Comparison Between Previous and Proposed Methodsmentioning

confidence: 99%