PARALLEL PROCESSING PROBLEM AND SOLUTION - A CASE STUDY ON MATLAB PARALLEL COMPUTING TOOLBOX DATASTORES

Nur Sabrina Zulkifli #1, Mohamed Faidz Mohammed Said #2

# Faculty of Computer & Mathematical Sciences. Universiti Teknologi MARA
70300 Seremban, Negeri Sembilan, MALAYSIA

1 sabrienazulkifli@gmail.com
2 faidzms@ieee.org


Abstract—Datastore in MATLAB is composed of a gigantic majority of the data. The datastore is a warehouse for collections of data that are too bulky to fit in memory. Numerous calculations are escalated in MATLAB applications and thus more works can benefit by having quicker execution if parallelism is provided by MATLAB. This particular tool provides solution for the demand of scrutinizing a particular record or a social event of reports. It goes about as a hotspot for data that has a comparable structure and is orchestrating it. Furthermore, accessing the big data of datastore would also deliver various advantages to the users. It can easily specify data set like single text file or collection of text files. It can incrementally read subsets of the data. Overall, datastore is used for many key reasons. In terms of data characteristics, the text data should have n files stored in the Hadoop Distributed File System (HDFS). The compute platform is used for the desktop applications. Lastly, for the analysis characteristics, it can facilitate users by supporting the load, analysing and also discarding the workflows.

Keyword: datastores, toolbox, parallel computing, MATLAB



REFERENCES

[1]	M. Paluszek and S. Thomas, MATLAB Machine Learning. Apress, 2016.

[2]	Y. Chen and S. F. Tan, "Matlab* g: A grid-based parallel Matlab," 2004.

[3]	 C. Politis, "5G-enabled Emergency Networks," in WWRF 39th meeting, Barcelona, Spain, 18/10/2017 2017: Kingston University London. 

[4]	S. Zulkifli, "Reasearch Paper Progress NSZ ", ed, 2017.

[5]	A. R. González. Matlab - Data Handling and Processing. Available: http://www.cs.bath.ac.uk/dm4t/w2-presentations/alfonso-matlab.pdf.

[6]	K. De Gussem, J. De Gelder, P. Vandenabeele, and L. Moens, "The Biodata toolbox for MATLAB," Chemometrics and Intelligent Laboratory Systems, vol. 95, no. 1, pp. 49-52, 2009.

[7]	 J. Lee, "MATLAB New Features Interface for Big Data Analytics," in MATLAB Expo 2015, Korea. 

[8]	G. Gopal, R. Kumar, and N. Kumar, "Genetic Algorithm: Simple to Parallel Implementation using MapReduce."

[9]	D. Willingham, "Big Data & Predictive Analytics."

[10]	M. P. Kulkarni, "Dissemination of Information in Multi Sensor Data Fusion in Vehicular Safety Applications," Vol. No. 2 Issue No. 2 August 2014, p. 63, 2014.

[11]	TC Toolbox for MATLAB Programmer Guide 2015.

[12]	 J. Krupa, A. Procházka, V. Hanta, and R. Háva, "Technical Computing Using Sybase Database for Biomedical Signal Analysis," in Proceedings of the Conference on Technical Computing. MathWorks & Humusoft, 2009. 

[13]	 D. Willingham, "Big Data Analysis and Analytics with MATLAB," in 15th Int. Conf. on Accelerator and Large Experimental Physics Control Systems (ICALEPCS'15), Melbourne, Australia, 17-23 October 2015, 2015: JACOW, Geneva, Switzerland, pp. 656-659. 

[14]	N. Suri et al., "Peer-to-peer Communications for Tactical Environments: Observations, Requirements, and Experiences," IEEE Communications Magazine, vol. 48, no. 10, pp. 60-69, 2010, doi: 10.1109/MCOM.2010.5594678.

[15]	M. A. Keyvanrad and M. M. Homayounpour, "A brief survey on deep belief networks and introducing a new object oriented toolbox (DeeBNet)," arXiv preprint arXiv:1408.3264, 2014.