Article Preview
TopIntroduction
The term ‘big data’ has gained huge popularity in recent years among IT professionals and decision makers. A recent survey, conducted by IDG Enterprise (2014) amongst more than 750 IT decision-makers, has shown the interest in big data continues to rise, as nearly half of the respondents (50%) are implementing or planning to implement big data projects within their organizations. Data is increasing massively nowadays. Easily, it is possible to find data in every aspect of our life. Data is getting bigger because it is being generated from more devices and sources. According to a report published by cisco in 2015, around 11 billion devices will be in use by 2019. Obviously, huge volume of data would be generated. Furthermore, it is interesting to see how data increased in the past decade. A study by Intel showed that humankind has generated five Exabytes until 2003. Data has increased to reach 2.7 Zettabytes from 2003 to 2013, (i.e. 500x more data). Data will continue to increase to three times bigger by 2015. In same context, Das & Kumar (2013) stated that nowadays five Exabytes is being generated every two days. This will continue to reach eight Zettabytes by 2015 (i.e. equivalent of 18 million libraries of Congress).
It is obvious how the volume of data is increasing enormously. But in spite of that, a very small fraction of this data has been exploited; the rest is not tapped yet. This shed light on the types of data. There are two types of data, structured data which can be sorted and organized in ordinary database engines. At the other hand, unstructured data does not have pre-defined data models that allow sorting in the ordinary databases such as videos, photos, images, emails, text documents and blogs. Das & Kumar, (2013) argued that unstructured data would account for 90% of data in the next decade. They stated that analyzing this massive amount of data would expose new improvements that were impossible to determine previously. Hence, organizations would seek for technologies that allow them to deal with this data. Here, big data and analytics appear in the horizon to achieve that.
Big data is defined as “Large pools of data that can be captured, communicated, aggregated, stored, and analyzed” (Manyika et al. 2011, p. 4). However, Laney & Taylor (2013) spoke on conditions of big data. They conditioned the data growth to be three-dimensional, high volume where data is very huge in size, high velocity where data is increasing very quickly in a pace that we can’t follow, and high variety where data is coming from variety of sources. For that, Woo et al. (2011) combined the above in their definition. They characterized big data as the new generation of technology and architecture that is designed to extract value from a very large volume of data that is coming from variety of resources by enabling high-velocity capture, discovery and/or analysis.
With the availability of the data, organizations seek to extract value and gain business intelligence needed to achieve their goals. Das & Kumar (2013) argued that big data analytics as a technology-enabled strategy would help organizations in gaining richer, deeper, and more accurate insights into customers, partners and the business and ultimately gaining competitive advantage.
This research aims to develop a model of factors that will affect the successful implementation of big data. In order to identify those factors, it is important to study why organizations are implementing big data, and identify the challenges that hinder the successful implementation. This is done by studying the relevant literature. The result obtained from the literature will be used to develop an initial model of factors that effect on implementation of big data. This will be followed by primary data collection through semi-structured interviews. The themes obtained from the literature review results will be considered as the basis to build the interviews’ questions. Furthermore, the findings obtained from the primary data will be used to extend and consolidate the initial model.