Project 3: UNN-UNESCO/HP Brain Gain Initiative Project on Grid Computing Funded by: HP-UNESCO Brain Gain Initiative, Grant No. 3250042751, 2012, MPI - Collins N. Udanor

A project funded under the UNN-UNESCO/HP Brain Gain Initiative. The Grid Computing Project was on Sustaining the Research and Grid Computing Components of the University of Nigeria’s UNESCO-HP Brain Gain Project. Grid Computing is a distributed computing infrastructure where powerful computers across the globe are interconnected to jointly perform tasks using specialized application programs in what is known as a virtual organization (VO). We designed and implemented the University of Nigeria Grid Computing Infrastructure – the LIONGRID UNN the first and only grid computing infrastructure in Nigeria.

The UNN-UNESCO-HP Brain Gain Initiative was a project aimed at building a high-performance computing infrastructure using grid computing technology. we made use of 2 numbers of HP multi-core Xeon computing servers and 3 numbers of HP Z-workstations to build the grid computing infrastructure. All the computing systems were housed in the University’s Data Centre and connected to the University’s network with a dedicated Cisco switching router. Dedicated Internet Protocol (IP) addresses were assigned to all 5 systems. we then used the grid computing middleware (a specialized software) known as gLite version 4.3 to configure all the 5 systems, ensuring that each one communicates with one another. Middleware refers to the software that sits between the application layer and the underlying hardware infrastructure and enables the various components of the grid to communicate and coordinate with each other. Middleware can include a wide range of technologies, such as job scheduling software, resource management tools, and data management systems, which all work together to enable the efficient and effective distribution of computing tasks across a network of computers.

This work is significant in making available powerful high-performance HP ProLiant multi-core computing servers that enabled researchers to submit their computation job for processing. We installed a number of simulation software in the server such as the OpenFoam, which used by some of our Ph.D. students for 2D Fluid Mechanics simulation. It made their work much easier, and result was delivered in record time.

We also developed an application software for predicting plant tissue culture growth. Plant tissue culture is a method of micropropagation of plant cells, tissues, or organs under sterile conditions on a nutrient culture medium of known composition. The application we developed using Python programming language was deployed in the Lion Grid to predict the combination and quantity of media such as auxin and cytokinin that would be needed to grow plant tissues know as ex-plants. We Used laboratory data from the cocoyam plant to develop and train the application. It predicted the number of roots, shoots, and heights of the ex-plants. The purpose of this application is to reduce the number of trials a scientist has to do in the lab to grow plants from plant tissues. Using this application the scientist knows the right combination and quantity of media that would give the best yield before going to the lab. This way, time and money is saved in numerous trials.

The results obtained from the simulation showed over 67% prediction accuracy as compared to the laboratory experiments.

The infrastructure was also significant in enabling us to collect big data from different applications in real-time since the infrastructure was made available to global research works like the Search for Extraterrestrial Intelligence (SETI) research (seti.org) whenever the machines were idle.

This project is a high-performance computing project which is built on the principle of distributed computing to be able to handle large workloads that individual computers may not be able to handle. Grid computing enabled us to have access to high-speed computers for research applications such as big data applications, parallel computing, etc.

In our current research we are leveraging on the skills gained from this project to design big IoT data infrastructure for machine learning tasks. These skills include setting up a high-performance server from the scratch, setting up the operating systems such as the Linux servers, configuring networks, and providing network security using firewalls, among others.

Project publications

  1. Akaneme, F. I., Udanor, C. N., Nwachukwu, J., Ugwuoke, C., Okezie, C. E. A. and Ogwo, B. (2014). A Grid-enabled Application for the Simulation of Plant Tissue Culture Experiments. International Journal of Advanced Computer Science and Information Technology 3, (3), 227 – 242. – published journal paper
  2. N. Udanor, F.I. Akaneme, Okezie, CEA, Ogwo B, Aneke S, Ogbuokiri B.O, Ezugwu A.O, Ugwuishiwu C.H. Deployment of an e-Infrastructure for Academic Research. The 7th EAI International Conference on e‐Infrastructure and e‐Services for Developing Countries (AFRICOMM 2015). Springer Lecture Notes of the Institute of Computer Sciences, Social Informatics and Telecommunications Engineering. Volume 171. – published conference paper