The main goal of this project is to improve the computation times in data processing for the R and Python programming languages by utilizing parallel programming techniques. The goals include implementing and leveraging various parallel computing strategies. The performance between the two languages are then compared. In this project, we processed three large XML datasets, comprising [datascience (70 MB)], [softwareengineering(1.3 GB)] and [stackoverflow (185 GB)] data. We used R and Python program scripts that implement parallel and distributed programming strategies for computation speed ups. We implemented SAX (Simple API for XML) processing instead of DOM (Document Object Mode) on the humongous XML files in this project because DOM processing requires memory several times the size of XML files. Thus, it is impossible to use DOM processing as the machine used in this project only have 8 GB memory. SAX processing requires almost no memory to process large XML files as it does not load them into memory. Processing time is very important in data analytics. It would be very unproductive to be waiting hours, days or even months for the results. Not to mention time-sensitive analysis. So, we must achieve a reasonable duration for completion of data processing. In this project, we found that parallel and distributed processing improved the processing time of large XML files significantly. We executed multi-threading SAX processing on an exactly same XML file multiple times and saw increase in data processing speed. This experiment is wrongly setup because when we did multi-threaded SAX processing on real dataset (several different XMLs), we saw a decrease in data processing speed. This is because the machine in this project only had single hard disk drive, thus the performance is limited by the single read head as the disk need to spin back and forth to serve multiple read on multiple XML files. In this project, we proved that data processing speed is improved when we did parallel SAX processing on multiple HDD. We analysed large XML files in this project, but depending on the types of data and its source format, the method to use in Big Data analysis must be approached Abstract xviii differently on a case by case basis. We found the methods of processing in this project for our XML can be improved by adding more HDD, parallel process data on separate machines and combine the results upon completion and increasing the bufier size of SAX parser. It is important to note that, as computer science students, our primary focus in this project is on computing techniques and time speed ups. Even though we will be processing big data, the value of the data and its output are of secondary importance.