225 research outputs found
Can Everybody be Happy in the Cloud? Delay, Profit and Energy-Efficient Scheduling for Cloud Services
No full textThe rapid development of Cloud Computing provides consumers and service providers with a wide range of opportunities and challenges. Considering the substantial infrastructure investments being made by cloud providers, the reduction of operating expenses (OPEX) while maximizing the profit of the provided services is of great importance. One way to achieve this is by maximizing the efficiency of resource utilization. However, profit maximization does not necessarily coincide with the improvement of a user’s Quality of Service (QoS); users generating higher profit for the provider may be scheduled first, causing high delays to low-paying users. Further, the contradictory nature of users’ and providers’ needs also extends to the energy consumption problem, as the minimization of service delays could cause cloud resources to be constantly “on”, leading to high energy consumption, high costs for providers and undue environmental impact. The objective of our work is to analyze this multidimensional trade-off. We first investigate the problem of efficient resource allocation strategies for time-varying traffic, and propose a new algorithm, MinDelay, which aims at achieving the minimum service delay while taking into account provider’s profit. Then, we propose E-MinDelay, an energy-efficient approach for CPU-intensive tasks in cloud systems. Furthermore, we propose an improved version of the Energy Conscious Task Consolidation (ECTC) algorithm, which combines task consolidation and migration techniques with E-MinDelay. Our results demonstrate that energy consumption and service delays corresponding to profit loss can be simultaneously decreased using an efficient scheduling algorithm
Deep learning-based scale-invariant cancer detection from whole slide image
No full textConvential cancer diagnosis methods from whole slide images (WSI) train a deep Convolutional Neural Network (CNN) to make patch level predictions, and then aggregate the image-level predictions to classify a tumour as either benign or malignant. To classify a patch, the CNN extracts features through convolutional layers and then process the feature maps using fully connected layers. The size of the filters used in the convolutional layers defines the receptive field of the network. Small filters are computationally efficient but do not capture a large context. On the other hand, large filters allow learning features that capture a larger context but are very expensive both in terms of computational time and memory requirements. This paper focuses on two main challenges. The First one is how to incorporate a large context while minimizing the computational overhead. The second one is that the cancerous cells can be of arbitrary size, and thus any detection and recognition approach should be scale-invariant. We introduce the Dilated SPP VGG-16 network with different dilation rates applied to every block of the VGG-16 network. The proposed dilated SPP VGG-16 architecture allows to increase the receptive field of the network without increasing the filter size or the depth of the network, and thus are very efficient to train. It also enables the multiscale analysis without changing the architecture of the network and retraining. We tested the proposed approach on the publicly available Camelyon17 dataset. Our experiments show that the proposed CNN achieves comparable or better accuracy than a conventional deep learning method, but with significantly less computational time and memory requirements
Between a Polonaise and a Nocturne (In Greek)
No full textThis short story is an extract from Barbed Wire in Aphrodite's Garden, an unpublished novel by John Bandler. Translated from English to Greek with assistance from Eirini Zacharidis. Thanks also to John Vlachopoulos, Dinos Mavromatis, and Polychronis Koutsakis. See also http://www.bandler.com/venusA well-heeled Greek boy meets a Turkish girl in a remote village in strife-torn British-colonial Cyprus in 1956 and asks her for a date
Scheduling for telemedicine traffic transmission over WLANs
No full textThe major disadvantage of the Enhanced Distributed Channel Access (EDCA, the contention-based channel access function of 802.11e) is that it is unable to guarantee priority access to higher priority traffic in the presence of significant traffic loads from low priority users. This problem is enhanced by the continuously growing number of multimedia applications and the popularity of Wireless Local Area Networks (WLANs). Hence, solutions in scheduling multimedia traffic transmissions need to take into account both the Quality of Service (QoS) requirements and the Quality of Experience (QoE) associated with each application, especially those of urgent traffic, like telemedicine, which carries critical information regarding the patients’ condition. In this work, we propose an easy-to-implement token-based and self policing-based scheduling scheme combined with a mechanism designed to mitigate congestion. Our approach is shown to guarantee priority access to telemedicine traffic, to satisfy its QoS requirements (delay, packet dropping) and to offer high telemedicine video QoE while preventing bursty video nodes from over-using the medium
Efficient Call Admission Control for MPEG-4 wireless videoconference traffic
No full textUser mobility, combined with the rapidly growing number of multimedia applications, form a challenging and yet unresolved problem for the development of Call Admission Control schemes over next generation wireless cellular networks. In this work we propose a new efficient CAC scheme for MPEG-4 videoconference traffic over cellular networks, which uses precomputed traffic scenarios for its decision-making. Our scheme is shown, via an extensive simulation study, to clearly excel in comparison with well-known existing approaches, in terms of quality of service (QoS) provisioning to users receiving videoconference traffic
A new Call Admission and Medium Access Control framework for multimedia traffic over GEO satellite networks
No full textThe burstiness of multimedia traffic and the long propagation delays in Geosynchronous (GEO) satellite systems call for an efficient Medium Access Control (MAC) protocol and an equally efficient Call Admission Control (CAC) scheme, in order to provide acceptable Quality of Service (QoS) to multimedia users. This paper proposes a fair and dynamic CAC and MAC framework, named Fair Predictive Resource Reservation Access (FPRRA), which is based on accurate videoconference traffic prediction and makes decisions after taking into account the provider revenue. The framework¿s performance is evaluated in comparison to other efficient schemes from the literature
Efficient traffic policing for videoconference traffic over wireless cellular networks
No full textThe constant development of new multimedia applications which are ldquogreedyrdquo in terms of bandwidth and quality of service requirements calls for new approaches to the traffic policing problem. In recent work we have modeled the behavior of multiplexed H.263 videoconference traces. Our results lead us to introduce, in this work, a new video traffic model for single H.263 videoconference sources and to propose a new traffic policing scheme for wireless videoconference traffic, based on the model. To the best of our knowledge, this is the first work in the relevant literature where the token generator is based on a traffic model, and not on a fixed rate
Classification of overlapped data with improved regularisation techniques using Fuzzy Deep Neural Networks
No full textThis thesis investigates methods to enhance the performance of a Deep Neural Network (DNN) classifier when dealing with numerical data by the introduction of improved regularisation techniques. In this thesis, three factors are considered for enhancement. The three significant factors that are considered are overlapped data in balanced and imbalanced data environments and the issue of invariance and overfitting.
Many classification algorithms, such as DNNs, classify a data item as belonging to either true or false for binary classification. However, in some real-world applications, they may have data items belonging to two classes to a certain degree. Data with similar characteristics appear in the feature space with different degrees of belongings is known as overlapped data. The overlapping class issue is one of the significant factors that lead to poor classification performance. In practice, there are two ways of handling this overlapped data issue. First, is the removal of overlapped instances, and the second is the separating of the overlapped regions and classify them separately. However, there are many drawbacks to these practices. The removal of overlapped instances is not the best option as it may remove essential data items that describe the dataset, especially in an imbalanced dataset.
On the other hand, when the overlapped regions and non-overlapped regions are classified separately, then it is a time-consuming task. Hence, there is a need to consider other techniques to handle overlapped data. Furthermore, a traditional classifier does not consider the underlying overlapping behaviour of the data attributes. However, the underlying overlapping behaviour of the data attributes can be addressed with the use of fuzzy concepts. When a data item belongs to different degrees to different classes, that belongings can be modelled using fuzzy concepts to classify the classes. Therefore, in this research, an overlapped data handling technique using Fuzzy C-Means, fuzzy membership grades, and cluster centre values named as FuzzyDNN is proposed. The results indicated that the proposed FuzzyDNN is capable of addressing the underlying behaviour of the overlapped data when performing classification. FuzzyDNN improves the classification accuracy by 8.89%, 0.88% and 1.24% when compared with the next highest performing technique for the three datasets tested on this thesis.
On the other hand, DNNs tend to overfit due to its ability to extract more features from a given set of data. One of the main problems in the generalisation capability of a DNN classifier is due to a small number of training data with limited variations is used. It is, therefore, vital to present training data with different variations of the domain to a classifier to ensure that the classifier can generalise well. Therefore, if pattern variations are smaller in the training dataset, one cannot expect a good generalisation from the classifier. Hence, in this research, a technique to improve the generalisation capability of DNNs is proposed to address this issue. Generally, the techniques used to improve the generalisation ability is known as regularisation techniques. There are various regularisation techniques in practice to handle different issues that can affect the generalisation capability of the DNN. However, the proposed technique is capable of augmenting numerical dataset to enhance the training dataset by introducing variations in the training of a classifier. In this thesis, the FDA, the proposed data augmentation technique, uses fuzzy concepts. The experimental results indicated that the FDA could enhance the training dataset to assist the DNN classifier to generalise well to the unseen data and act as a proper regularisation technique when compared with some commonly used regularisation techniques.
Finally, in this research, the classification of the overlapped data for an imbalanced dataset, and its generalisation capability are considered concurrently. An imbalanced binary dataset is a dataset with instances of one class predominately higher than the other class. In such scenarios, the traditional classifiers biases towards the majority classes. However, the performance of a classifier degrades heavily when overlapped data also appear in the imbalanced dataset. Given that the issues of invariant of training data for the DNN can also occur at the same time given that the available data could be limited, there is a need to have a suite of techniques working together to address the three issues concurrently. Further, there is a limited amount of work concentrates on numerical data classification with DNNs for imbalanced overlapped data. Therefore, in this research, a model is proposed to handle the overlapped data in an imbalanced dataset using the proposed data augmentation technique to improve the generalisation ability of the DNN classification model. All the algorithms proposed in this thesis was implemented using MATLAB and Python (in an Anaconda Environment)
Call Admission Control for wireless videoconference traffic based on the users' willingness to pay
No full textNetwork designers face a challenging problem when trying to control the traffic entered into a wireless cellular network. The reason is the rapidly growing number of multimedia applications, combined with user mobility. Additionally, the existence of multiple cellular providers leads to strong competition among them. Providers need to use efficient resource management schemes in order to keep existing clients satisfied and attract new customers, so that they can increase their revenue. In this work, we use traffic modeling for H.264 videoconference traffic in the implementation of a new Call Admission Control mechanism which makes decisions on the possible acceptance of a video call into the network not only based on the predicted bandwidth that users will need, but also on the revenue that the provider will make when degrading current users in order to accommodate new ones. Our mechanism is shown, via an extensive simulation study, to provide high Quality of Service (QoS) to wireless H.264 videoconference users
QRP05-1: A new model for multiplexed VBR H.263 videoconference traffic
No full textDue to the burstiness of video traffic, video modeling is very important in order to evaluate the performance of future wired and wireless networks. In this paper we build a discrete autoregressive (DAR) model to capture the behavior of multiplexed H.263 videoconference movies from VBR coders. Our results show that the model provides great accuracy in its prediction of the behavior of the multiplexed sources
- …
