Mathematical models for network card simulation and their empirical validations

Nurika, Okta; Hassan, Mohd Fadzil; Zakaria, Nordin; Jung, Low Tan

doi:10.1109/ismsc.2015.7594029

Cited by 3 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order for the network card to achieve near-maximum receiving speed, it must be set according to the transmission characteristics that it receives, which include the transmission size and the packet size. The network card setting consists of 3 options, which are active wait, passive wait, and watermark mode as detailed in our paper [7] about the proposed mathematical models to represent the throughput generation of these 3 modes. The active wait triggers kernel interrupt to call the CPU to process every packet that comes.…”

Section: Optimization Casementioning

confidence: 99%

A Study of Fluctuations in Genetic Algorithm Optimized Network in Data Centre

Nurika¹,

Hassan²,

Zakaria³

et al. 2017

Proceedings of the First EAI International Conference on Computer Science and Engineering

Self Cite

View full text Add to dashboard Cite

Study of fluctuation in genetic algorithm has been a sub-objective in genetic algorithm implementations. The reliability of genetic algorithm may vary based on implementation case, hence it is necessary to investigate its performance pattern for each implementation case. The purpose of this study is to observe the reliability of genetic algorithm in our previously simulated network optimization in a data centre. Previous researchers found fluctuation as random occurrence, mainly within small population. This paper's fluctuation observation revolves around our recent optimization of data centre's network. Our findings agree with the nature of genetic algorithm and other researches, where it is found that the fluctuation of fitness values in our case happened randomly in general, but it had higher probability with small population size. However, regardless of fluctuations that in average occurred during early stage of population generation, the near-optimal solutions with near maximum fitness values were able to be generated. This fact has proven the robustness of genetic algorithm itself.

show abstract

Section: Optimization Casementioning

confidence: 99%

A Study of Fluctuations in Genetic Algorithm Optimized Network in Data Centre

Nurika¹,

Hassan²,

Zakaria³

et al. 2017

Proceedings of the First EAI International Conference on Computer Science and Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…This paper is an extension of the previous research about Genetic Algorithm (GA) assisted simultaneous multiple network cards optimization in a data centre [1], which is based on mathematical models described in [2]. The results of the simulations in [1] have been implemented in real physical network cards as published in [3] with the data transmissions sent using Hping3 software.…”

Section: Introductionmentioning

confidence: 99%

“…The throughputs of data transmissions from the optimized network cards and the default version are to be compared as a validation of methods proposed in [2], when implemented in a Hadoop data transfer infrastructure.…”

Section: Introductionmentioning

confidence: 99%

Implementation of Network Cards Optimizations in Hadoop Cluster Data Transmissions

Nurika¹,

Hassan²,

Zakaria³

2017

ICST Transactions on Ubiquitous Environments

Self Cite

View full text Add to dashboard Cite

In this paper, the previously invented new methods of network card optimization are applied in a Hadoop cluster, where data transfers occur from the Master to the slave node. The slave node's network card setting is optimized subjective to the characteristics of the incoming data transmissions, which are indicated by the overall transmission size and packet size. The throughput comparisons between the optimized network card settings and the default setting conclude that the optimized versions always generate higher throughputs. Synchronously, the optimized settings also minimize CPU cycles utilization as they deploy timer-based polling (passive wait mode), in order to process the received data packets. This novel practice within Hadoop cluster may be replicated by other data cluster vendors, thus improving their data transfer's throughput and efficiency.

show abstract