Article Preview
TopIntroduction
The current economic climate has resulted in businesses adopting models that are distributed and which are dependent on Internet communications. One such model is cloud computing. There is a drive from industry, service providers and government bodies to encourage businesses to realize the potential of cloud computing. The European Union considers Cloud computing as a set of technologies that could revolutionize the way businesses operate (European Commission 2014). Cloud computing refers to a large collection of virtualised services and computing applications that are delivered on a subscription based model. It encapsulates complex back-end processes and data-stores. For example, concepts like Business Intelligence as a Service (BIaaS), allows for processing business related data-sets and precision data mining using processes hosted in the cloud (Chang 2014). The concept of big-data is a result of cloud computing to address the need to host databases that can handle “out-of-the-norm” type of data. These are the type of data that could not be accommodated in traditional Relational Database Management Systems (RDBMS). For example, scientific datasets are so huge that without big-databases it would be almost impossible to store and process them to yield any meaningful information. Therefore, the combination of big-data techniques with cloud computing is a boon for a modern society where ICT is embedded into the way of life. One of the initiatives for adopting cloud computing is the ability to develop organisational IT services and business models that are in accordance with low power consumption policies to tackle climate change (Kasemsap 2015). However, this is not without any negative implications. Hosting big-data services on the cloud will require significant hardware, software and networking resources; all of which contribute to high volumes of energy consumption. Although cloud computing architectures have endeavoured to promote the green theme, there is still a major concern on the amount of green-house gases new ICT systems will contribute to (Mouftah and Kantarci 2013). This problem is not straight-forward as Dastbaz (2015) explains that one of the major challenges ahead is the question of how to accurately measure the energy consumed by power-hungry applications and data stores on the cloud. There is no doubt that clouds will continue to evolve as the dependency on them grows, but it is important to be wary of the environmental impact. It should come as no surprise that the severity of the environmental impact of data-centres has motivated companies like Microsoft, Yahoo and Google to create new data centres on Columbia River to harness renewable energy and cooling from flowing water.
Appreciating and accepting the link between cloud computing and sustainability will allow for research into methods to combat the energy crisis as clouds continue to emerge. These methods should be focussed around components and principles that comprise of the cloud. Some of these are Grid computing, virtualisation, autonomic computing, utility computing etc (Domdouziz, 2015). To address energy efficiency, it is important to acknowledge that the above components which are often in a distributed fashion, process huge amounts of data possibly over large scale networks and/or data centres. Hence sustainability must consider how the “personality” of network traffic can be gathered from various points. This understanding could then form the basis for cloud management algorithms. Saving energy requires using specialist techniques like traffic consolidation, link management etc (Bilal et al, 2014). But his not always easy as there are so many “moving parts” that change rapidly and that have a rippling effect on other entities. This paper looks into some means of gathering network traffic related data and measuring them, based on previous research and related literature.