While cloud computing provides a reliable means to support remote and ensure an effective on demand access to computing resources, its impromptu management creates significant challenges to scalability, dependability, economical costs and performance of the cloud services. Hence the Service Level Agreement (SLA) and the Quality of Service (QoS) will be the distinguishing element between the cloud computing environments. Conversely, service constituents and the cloud elasticity that bring about cloud computing make the management of cloud significantly challenging. The proposed PhD project intends to investigate novel procedures for scalable and competent provisioning of the cloud, whilst assuring dependability and reducing costs. The project will therefore study a ground-breaking approach that puts together the control theory procedures with the distributed algorithms so as to generate self-elastic clouds. Control laws, novel models and distributed algorithms will consequently be proposed for automatic provisioning, configuration and employment of cloud services so as to meet the requirements for SLA, while dealing with dynamics and scalability issues.
The urgency for data storage has resulted from the need for constant memory or information. The continuous demand for data access has led to the development of numerous methods for managing and organizing it. The advancement from data processing to user utilization has led to substitutions in the way raw data is organized and searched. This explosive increase of unstructured data brings challenges in the scalability, dependability and performance of the wed services. Cloud is up-and-coming as an occurrence and it is booming at the convergence of several tendencies in the software engineering. Service based architecture, internet oriented and virtualization applications have rapidly developed in the past few years. Following this, cloud is proving to be the next major step in this field. This is because; it several tasks can be performed through a specific network services. Within this relatively short time, various web services have been on use for cloud computing. Some of these include; the software as a service, platform as a service, and platform as a service among others.
Nevertheless, debates have been boiling up on how the cloud can be improved so as to ensure quality and cheap services. Service Level Agreement (SLA) is one way of ensuring scalability, dependability and availability of web services to the consumers. The unspoken demand to eradicate the loss and unavailability of data has triggered the development of exceedingly available and reliable storage structures. In addition to this, the need for cheap web services has resulted to the development of advanced devices and systems for storage. With this in mind, there is need to innovate more reliable models and techniques that will improve the scalability and dependability of data.
- Research Question I: Novel models and distributed algorithms have not been efficient in the provisioning, configuration and employment of cloud services
- Research Question II: There have never been algorithms that are scalable for the minimization of web service cost thus data retrieval is relatively expensive.
The storage and retrieval of information is a systematic procedure that encompasses the collection and cataloging of data so that it can easily be located and recovered on request. Advanced information processing techniques have made the speedy and careful retrieval of large amounts of data possible. Database schemes can store data in the form of distinct records that can be explored and retrieved based on the content of their fields. However, the volume of data has swiftly been on constant increase in the past few years. Whilst computer technology has to a greater extent encouraged the growth of information, its rapid growth has posed a challenge on the storage and retrieval. In the past, several kinds of data handling databases have been developed including the relational network models but they still have their own shortages despite the fact that they have been improved. In addition, most of the data is currently in unstructured format thus it can not be properly managed by the ancient models. Therefore there is need for invention of advanced protocols that will work towards scalable and dependable computing in the cloud for unstructured data.
Qian, Tipper & Krishnamurthy (1) have supported the idea that the advanced computer technology has brought about multifaceted problems that have continually posed a threat on the confidentiality, integrity and availability of data. Several researchers have described data confidentiality as the general term that has been used in the prevention of data from exposure to systems or individuals that are not permitted. Integrity on the other hand has been described as protection of data to the extent that it can not be altered without permission while the purpose of availability is to ensure that data is easily accessible when needed (2). Horton & Mugge (3) have provided dependable tips that help IT scholars to deliver convenient hacking and protection architecture solutions as well as how to use effective and advanced security equipment. Tlhong and Reeve in (4) have described a Service Level Agreement (SLA) that is based on Service Level Agreement Language (SLAng). In their work, they have successful analyzed SLAs and provided an inspiration for automating and modeling of SLAs modeling.
Research has been done on the concepts and techniques that have been on use for the retrieval of information that has been hidden in huge data storage devices. Special focus has been put on the issues concerning the convenience, achievability, efficiency and scalability of the data (5). Hill (8), goes ahead into explaining the key objectives of unstructured data protection which include; integrity, confidentiality and availability. Research has shown that machine decipherable SLAs are effective in improving the levels of computerization in the management of SLA thus reducing possible violations (4). Web services have been greatly expanding over the past years and this rate is expected to increase in the coming years and to ensure that unstructured data is safe and well secured, apposite access-control techniques need to be enforced (6). Photopoulos (7) also supports the idea that a multi-tiered move toward security of unstructured data, coupled with appropriate strategies and clear access control is necessary.
Like many other researchers, Qian, Tipper & Krishnamurthy (1) have also acknowledged that the advanced technology has resulted to complex problems that continue to intimidate the availability, integrity, and confidentiality of unstructured information. Their work has also strived in providing an inclusive exposure of security procedures and survivability as well as determining how these two apparatus work together to construct a vigorous and reliable Information Assurance. The prominence of internet-based and rich media applications has led to the sudden increase of unstructured data that in turn require secure and more reliable storage and retrieval solutions (13). Thuraisingham (12) asserts that applications that involve video processing, rendering of animation, audio, complex websites coupled with the aptitude to easily create digital information from hand phones and other devices has greatly contributed to the burdens of storage capacities. Unlike structured data, unstructured data is formless and includes PDF files, slide decks, videos, web pages, images, MP4 and Mp3 files and word documents among others (12). The storage and retrieval of data, both structured and unstructured is the whole process that involves the collection and recording of data so as to make it available to the users on request (9).