Organizations are turning to private clouds due to concerns about security, privacy and administrative control. They are attracted by the flexibility and other advantages of cloud computing but are wary of breaking decades-old institutional practices and procedures. Private Clouds can help to alleviate these concerns by retaining security policies, in-organization ownership and providing increased accountability when compared with public services. This work investigates how it may be possible to develop an energy-aware private cloud system able to adapt workload allocation strategies so that overall energy consumption is reduced without loss of performance or dependability. Current literature focuses on consolidation as a method for improving the energy-efficiency of cloud systems, but if consolidation is undesirable due to the performance penalties, dependability or latency then another approach is required. Given a private cloud in which the machines are constant, with no machines being powered down in response to changing workloads, and a set of virtual machines to run, each with different characteristics and profiles, it is possible to mix the virtual machine placement to reduce energy consumption or improve performance of the VMs. Through a series of experiments this work demonstrates that workload mixes can have an effect on energy consumption and the performance of applications running inside virtual machines. These experiments took the form of measuring the performance and energy usage of applications running inside virtual machines. The arrangement of these virtual machines on their hosts was varied to determine the effect of different workload mixes. The insights from these experiments have been used to create a proof-of- concept custom VM Allocator system for the OpenStack private cloud computing platform. Using CloudMonitor, a lightweight monitoring application to gather data on system performance and energy consumption, the implementation uses a holistic view of the private cloud state to inform workload placement decisions.

Mobile handheld devices such as smartphones are now convenient as they allow users to make calls, reply to emails, find nearby services and many more. The increase in functionality and availability of mobile applications also allow mobile devices to be used in many different everyday situations (for example, while on the move and carrying items). While previous work has investigated the interaction difficulties in walking situations, there is a lack of empirical work in the literature on mobile input when users are physically constrained by other activities. As a result, how users input on touchscreen handheld devices in encumbered and mobile contexts is less well known and deserves more attention to examine the usability issues that are often ignored. This thesis investigates targeting performance on touchscreen mobile phones in one common encumbered situation - when users are carrying everyday objects while on the move. To identify the typical objects held during mobile interactions and define a set of common encumbrance scenarios to evaluate in subsequent user studies, Chapter 3 describes an observational study that examined users in different public locations. The results showed that people carried different types of bags and boxes the most frequently. To measure how much tapping performance on touchscreen mobile phones is affected, Chapter 4 examines a range of encumbrance scenarios, which includes holding a bag in-hand or a box underarm, either on the dominant or non-dominant side, during target selections on a mobile phone. Users are likely to switch to a more effective input posture when encumbered and on the move, so Chapter 5 investigates one- and two- handed encumbered interactions and evaluates situations where both hands are occupied with multiple objects. Touchscreen devices afford various multi-touch input types, so Chapter 6 compares the performance of four main one- and two- finger gesture inputs: tapping, dragging, spreading & pinching and rotating, while walking and encumbered. Several main evaluation approaches have been used in previous walking studies, but more attention is required when the effects of encumbrance is also being examined. Chapter 7 examines the appropriateness of two methods (ground and treadmill walking) for encumbered and walking studies, justifies the need to control walking speed and examines the effects of varying walking speed (i.e. walking slower or faster than normal) on encumbered targeting performance. The studies all showed a reduction in targeting performance when users were walking and encumbered, so Chapter 8 explores two ways to improve target selections. The first approach defines a target size, based on the results collected from earlier studies, to increase tapping accuracy and subsequently, a novel interface arrangement was designed which optimises screen space more effectively. The second approach evaluates a benchmark pointing technique, which has shown to improve the selection of small targets, to see if it is useful in walking and encumbered contexts.

Αντικείμενο της Διπλωματικής εργασίας είναι η μεταφορά ενός παιχνιδιού στρατηγικής, πού έχει αναπτυχθεί με γνώμονα τους κανόνες της Ενισχυτικής Μάθησης (Reinforcement Learning) & των Νευρωνικών δικτύων (Neural Networks), σε πλατφόρμα κινητού τηλεφώνου 6ης γενιάς & μέσω διαδικτύου με σκοπό την εκπαίδευσης του συστήματος από τον απλό χρήστη. Σκοπός της διπλωματικής εργασίας είναι η σχεδίαση, ανάπτυξη του συστήματος λογισμικού βασισμένου σε τεχνολογίες διαδικτύου & κινητής τηλεφωνίας. Η Παρούσα εργασία επικεντρώνεται σε τρεις τομείς: • Έρευνα υφιστάμενων mobile τεχνολογιών • Δυνατότητα επανασχεδιασμού εφαρμογής με χρήση τεχνολογιών mobile internet • Έλεγχος υφιστάμενης δυνατότητας επανασχεδίασης και επαναϋλοποίησης του παιχνιδιού με χρήση άλλων συστημάτων (πλατφόρμων) σε συνδυασμό με χρήση ενισχυτικής μάθησης. / In this paper we review our work on th acquisition of game-playing capabilitiew by a computer , when the only source of knowledge comes from extended self-play and sparsely dispersed human-expert play. We summarily present experiment that showhow a reinforcement learning backbone coupled with neural networks for approximation can indeed serve as amechanism of the acquisition oof game playing skill and we derive game interestingness measures that are inexpensive and strightforward to compute, yert also capture the relative quality of the game playing engine.