Ad-hoc aggregate query processing algorithms based on bit-store for query intensive applications in cloud computing

Μαζικά σύνολα δεδομένων με πρωτοφανείς διαστάσεις εμφανίζονται πρόσφατα σε πολλούς τομείς. Αυτά τα σύνολα δεδομένων απαιτούν ειδικά εργαλεία ώστε να αναλύονται αποτελεσματικά και να εξάγονται χρήσιμες πληροφορίες από αυτά. Για την αντιμετώπιση αυτών των προκλήσεων έχουν δημιουργηθεί παράλληλα εργαλεία κατάλληλα για μαζικά δεδομένα. Το σημαντικότερο είναι να δοθεί ιδιαίτερη έμφαση ώστε να μπορούμε να παραλληλίσουμε διεργασίες στον βέλτιστο δυνατό βαθμό. Εν αντιστοιχία,ο σκοπός αυτής της διατριβής είναι να παράσχει μια μέθοδο απάντησης επερωτήσεων χρησιμοποιώντας ένα μόνο γύρο της τεχνικής του MapReduce. Αυτό επιτυγχάνεται δίδοντας ιδιαίτερη έμφαση στα δεδομένα εισόδου,δηλαδή σε δυαδικές ή «παχιές» σχέσεις δεδομένων . Yλοποιούμε έναν αλγόριθμο πολλαπλών συζεύξεων (“multiway join”) στους δύο αυτούς τύπους εισόδων και πραγματοποιούμε συγκρίσεις, όπου διαπιστώνουμε ότι οι «παχιές» σχέσεις δεδομένων επιτυγχάνουν καλύτερο χρόνο επεξεργασίας λόγω της φύσης αυτών των συνόλων δεδομένων. Η περίπτωση των "παχιών"σχέσεων δεδομένων, εξετάζεται στο πλαίσιο της ενοποίησης δεδομένων,συγκεκριμένα προτείνεται μια εφαρμογή για τη διασταύρωση των πληροφοριών στις πηγές των πολλαπλών δεδομένων, όταν υπάρχουν αντικρουόμενες πληροφορίες. Η περίπτωση των δυαδικών σχέσεων δεδομένων μελετάται επίσης σε σχέση με τα σύνολα δεδομένων RDF, τα οποία αποτελούν μια σημαντική εφαρμογή στον τομέα της διαχείρισης των σημασιολογικών δεδομένων ιστού και επιτυγχάνουμε σημαντική αύξηση στην ταχύτητα της επεξεργασίας επερωτημάτων.

Download Full-text

Ad-Hoc Parallel Data Processing on Pay-As-You-Go Clouds with Nephele

Advances in Systems Analysis, Software Engineering, and High Performance Computing - Distributed Computing Innovations for Business, Engineering, and Science ◽

10.4018/978-1-4666-2533-4.ch010 ◽

2013 ◽

pp. 191-218

Author(s):

Daniel Warneke

Keyword(s):

Cloud Computing ◽

Data Processing ◽

Ad Hoc ◽

Cluster Systems ◽

Software Frameworks ◽

Homogeneous Cluster ◽

Processing Cost ◽

Parallel Data ◽

The One ◽

Processing Framework

In recent years, so-called Infrastructure as a Service (IaaS) clouds have become increasingly popular as a flexible and inexpensive platform for ad-hoc parallel data processing. Major players in the cloud computing space like Amazon EC2 have already recognized this trend and started to create special offers which bundle their compute platform with existing software frameworks for these kinds of applications. However, the data processing frameworks which are currently used in these offers have been designed for static, homogeneous cluster systems and do not support the new features which distinguish the cloud platform. This chapter examines the characteristics of IaaS clouds with special regard to massively-parallel data processing. The author highlights use cases which are currently poorly supported by existing parallel data processing frameworks and explains how a tighter integration between the processing framework and the underlying cloud system can help to lower the monetary processing cost for the cloud customer. As a proof of concept, the author presents the parallel data processing framework Nephele, and compares its cost efficiency against the one of the well-known software Hadoop.

Download Full-text

Vehicular Cloud Computing

Mobile Networks and Cloud Computing Convergence for Progressive Services and Applications - Advances in Wireless Technologies and Telecommunication ◽

10.4018/978-1-4666-4781-7.ch014 ◽

2014 ◽

pp. 262-274 ◽

Cited By ~ 2

Author(s):

Kayhan Zrar Ghafoor ◽

Marwan Aziz Mohammed ◽

Kamalrulnizam Abu Bakar ◽

Ali Safa Sadiq ◽

Jaime Lloret

Keyword(s):

Cloud Computing ◽

Vehicular Networks ◽

Vehicular Ad Hoc Networks ◽

Ad Hoc ◽

Vehicular Cloud Computing ◽

Vehicular Cloud ◽

Performance Improvements ◽

Potential Applications ◽

Research Outcome ◽

Hoc Networks

Recently, Vehicular Ad Hoc Networks (VANET) have attracted the attention of research communities, leading car manufacturers, and governments due to their potential applications and specific characteristics. Their research outcome was started with awareness between vehicles for collision avoidance and Internet access and then expanded to vehicular multimedia communications. Moreover, vehicles’ high computation, communication, and storage resources set a ground for vehicular networks to deploy these applications in the near future. Nevertheless, on-board resources in vehicles are mostly underutilized. Vehicular Cloud Computing (VCC) is developed to utilize the VANET resources efficiently and provide subscribers safe infotainment services. In this chapter, the authors perform a survey of state-of-the-art vehicular cloud computing as well as the existing techniques that utilize cloud computing for performance improvements in VANET. The authors then classify the VCC based on the applications, service types, and vehicular cloud organization. They present the detail for each VCC application and formation. Lastly, the authors discuss the open issues and research directions related to VANET cloud computing.

Download Full-text

Security in Ad Hoc Network and Computing Paradigms

Cloud Security ◽

10.4018/978-1-5225-8176-5.ch030 ◽

2019 ◽

pp. 592-620

Author(s):

Poonam Saini ◽

Awadhesh Kumar Singh

Keyword(s):

Cloud Computing ◽

Distributed Computing ◽

Ad Hoc Networks ◽

Resource Sharing ◽

Vehicular Ad Hoc Networks ◽

Ad Hoc ◽

Dominant Role ◽

Portable Computing ◽

Secure Protocols ◽

Hoc Networks

Resource sharing is the most attractive feature of distributed computing. Information is also a kind of resource. The portable computing devices and wireless networks are playing a dominant role in enhancing the information sharing and thus in the advent of many new variants of distributed computing viz. ubiquitous, grid, cloud, pervasive and mobile. However, the open and distributed nature of Mobile Ad Hoc Networks (MANETs), Vehicular Ad Hoc Networks (VANETs) and cloud computing systems, pose a threat to information that may be coupled from one user (or program) to another. The chapter illustrates the general characteristics of ad hoc networks and computing models that make obligatory to design secure protocols in such environments. Further, we present a generic classification of various threats and attacks. In the end, we describe the security in MANETs, VANETs and cloud computing. The chapter concludes with a description of tools that are popularly used to analyze and access the performance of various security protocols.

Download Full-text

Conducting Research in the Cloud

Cloud Computing for Teaching and Learning ◽

10.4018/978-1-4666-0957-0.ch004 ◽

2012 ◽

pp. 50-63

Author(s):

Bay Arinze ◽

Cheickna Sylla

Keyword(s):

Cloud Computing ◽

Web 2.0 ◽

Ad Hoc ◽

Knowledge Generation ◽

Future Research ◽

Geographically Dispersed ◽

Computer Services ◽

New Ideas ◽

Conducting Research ◽

New Research

Web 2.0 research is a term for research that uses Web platforms and tools for collaboration, communication, and knowledge generation by researchers who may be geographically dispersed. The new tools allow additional forms of synergistic collaborations between ad-hoc groups of researchers, crowdsourcing of new ideas, and to represent innovative platforms for sharing knowledge more rapidly. In parallel with these new research developments, cloud computing has emerged as a new way to provision and use IT resources to all types of computer users. With cloud computing, computer services are accessed over the Internet in a scalable fashion, and users are abstracted from the actual hardware and software, paying only for resources they use. This chapter discusses how current and future research will make use of cloud computing and how Web 2.0-based research models are transforming how research is conducted globally. It examines these new IT infrastructure models and explores how they can be deployed by organizations and individuals. It then discusses the benefits of cloud computing to the research enterprise and future directions for cloud computing-based research.

Download Full-text