1,722,043 research outputs found
Ottimizzazione delle Prestazioni del Cloud Computing: Simulazione di Sistemi di Accodamento Multi-Server a Capacità Finita Optimizing Cloud Computing Performance: Simulation of Finite-Capacity Multi-Server Queueing Systems
No full textIl concetto di cloud computing nasce dall’architettura software distribuita con l’obiettivo di fornire servizi tramite Internet. Ha rivoluzionato l’accesso e l’utilizzo delle risorse computazionali, offrendo notevole flessibilità e scalabilità. Tuttavia, la natura dinamica degli ambienti cloud, accentuata dalle crescenti esigenze dei carichi di lavoro basati sull’Intelligenza Artificiale, che richiedono un vasto numero di risorse e funzionalità, comporta sfide significative nella gestione delle risorse, nella garanzia della qualità del servizio e nell’ottimizzazione delle prestazioni. Questo lavoro affronta le problematiche legate all’allocazione delle risorse dovute alla diversità delle tipologie (es. memoria, CPU, banda, I/O), concentrandosi sulla simulazione e analisi di sistemi di accodamento multi-server a capacità finita all'interno degli ambienti cloud, utilizzando il framework OMNeT++.
La prima parte della tesi ha riguardato lo sviluppo di una libreria di simulazione dettagliata per l’implementazione di sistemi di accodamento multi-server. Il lavoro si è concentrato sugli aspetti pratici della simulazione di ambienti cloud con risorse finite, in particolare attraverso la realizzazione di un simulatore open-source per analizzare modelli di accodamento su OMNeT++. Configurando diverse discipline di accodamento, come le politiche First-In First-Out (FIFO) e quelle basate sulla priorità, è stato possibile modellare dinamicamente l’allocazione delle risorse, simulare la pianificazione dei job e analizzare le metriche prestazionali, fornendo uno strumento solido per l’esame di scenari realistici nel cloud computing. L’importanza del simulatore è giustificata dalla mancanza di risultati analitici generali che permettano di prevedere efficacemente le metriche prestazionali.
Nella seconda parte, il lavoro è stato esteso con uno studio di simulazione approfondito dei sistemi di accodamento multi-server a capacità finita. Questo studio ha esplorato l’impatto delle diverse politiche di accodamento sulle metriche prestazionali in ambiente cloud. I risultati hanno evidenziato che assegnare priorità ai job con tempi di servizio più lunghi e maggiori richieste di risorse può migliorare le prestazioni complessive del sistema, beneficiando sia i job piccoli che quelli grandi.
Come prosecuzione di questa tesi, è stato analizzato l’effetto delle politiche di scheduling, in particolare quelle preemptive e non preemptive. Lo studio ha dimostrato che le politiche preemptive, incluso un approccio greedy in cui le risorse vengono liberate preemptando prima i job più piccoli e proseguendo progressivamente verso quelli più grandi fino al raggiungimento della soglia di risorse necessarie, migliorano le metriche come la probabilità di perdita, il tempo di attesa e l’utilizzo delle risorse. Tuttavia, queste politiche introducono anche sprechi computazionali, poiché i job attivi vengono interrotti durante il servizio. Questo compromesso mette in evidenza la complessità e l’eterogeneità nell’ottimizzazione degli ambienti cloud e la necessità di considerare attentamente sia le prestazioni sia l’efficienza delle risorse.
In generale, questo lavoro contribuisce a una migliore comprensione della relazione tra discipline di accodamento, allocazione delle risorse e prestazioni del sistema nei contesti di cloud computing. I risultati della ricerca sono di grande utilità non solo per i fornitori di servizi cloud, ma anche per i consumatori, per una progettazione ottimale dei sistemi cloud-based, grazie agli strumenti di simulazione sviluppati e alle conoscenze analitiche ottenute dagli esperimenti condotti.The concept of cloud computing has emerged from the distributed software architecture with the aim of providing services over the Internet. It has revolutionized the access and use of computational resources with remarkable flexibility and scalability. However, the dynamic nature of cloud environments, intensified by the growing demands of Artificial Intelligence-driven workloads that require a vast number of resources and capabilities, introduces serious challenges in resource allocation, service quality assurance, and performance optimization. This work addresses the challenges of resource allocation due to diverse resource types (e.g. memory, cores, bandwidth, input/output) by focusing on the simulation and analysis of finite-capacity multi-server job queueing systems within cloud computing environments, using the OMNeT++ framework. Specifically, we investigate the impact of scheduling policies on the performance metrics of a multi-server job system to contribute to a deeper understanding and provide solutions to optimize cloud resource allocation due to diverse resource types required by jobs from different types of sources.
The first part of this thesis involved developing a detailed simulation library for implementing multi-server job queueing systems. This work focused on the practical aspects of simulating cloud environments with finite resources, specifically developing an open-source simulator to analyze queueing models on the top of OMNeT++. By configuring various queueing disciplines such as First-In, First-Out and Priority-based scheduling algorithms, we were able to dynamically model resource allocation, simulate job scheduling, and analyze performance metrics, providing a robust tool for examining real-world scenarios in cloud computing. The importance of the simulator is justified by the lack of general analytical results that can predict the performance metrics efficiently.
In the second part, we extended our work by conducting a detailed simulation study of finite-capacity multi-server job systems. This study focused on understanding the impact of different queueing policies on performance metrics in the cloud environment. Our findings revealed that assigning priority to jobs with longer service times and higher resource demands can positively impact both smaller jobs with shorter service times and big jobs, and thus enhance overall system performance.
As a follow-up to this thesis, we further investigate the performance implications of scheduling policies, particularly preemptive and non-preemptive approaches. The study demonstrated that preemptive policies, including a greedy approach in which resources are freed by first preempting smaller jobs and gradually progressing to big ones until the required resource threshold is met, improved performance metrics such as the probability of dropping, waiting time and resource utilization. However, these policies also introduce computational waste as the active jobs are disrupted during their service in this model. This trade-off emphasizes the heterogeneity of optimizing cloud computing environments and the need for careful consideration of both performance and resource efficiency.
In general, our work contributes to a better understanding of the relationship between queueing disciplines, resource allocation, and system performance in cloud computing environments. These research findings are of great help not only to cloud service providers, but also to cloud consumers for the optimal design of cloud-based systems, which is achieved by simulation tools developed during the study and the analytical insight from the experiments conducted as part of the work
Exploring the determinants of methane emissions from a worldwide perspective using panel data and machine learning analyses
Full text linkThis article contributes to the scant literature exploring the determinants of methane emissions. A lot is explored considering CO2 emissions, but fewer studies concentrate on the other most long-lived greenhouse gas (GHG), methane which contributes largely to climate change. For the empirical analysis, a large dataset is used considering 192 countries with data ranging from 1960 up to 2022 and considering a wide set of determinants (total central government debt, domestic credit to the private sector, exports of goods and services, GDP per capita, total unemployment, renewable energy consumption, urban population, Gini Index, and Voice and Accountability). Panel Quantile Regression (PQR) estimates show a non-negligible statistical effect of all the selected variables (except for the Gini Index) over the distribution's quantiles. Moreover, the Simple Regression Tree (SRT) model allows us to observe that the losing countries, located in the poorest world regions, abundant in natural resources, are those expected to curb methane emissions. For that, public interventions like digitalization, green education, green financing, ensuring the increase in Voice and Accountability, and green jobs, would lead losers to be positioned in the winner's rankings and would ensure an effective fight against climate change
Asset pricing in the Middle East’s equity markets
Full text linkYesThis paper undertakes a comparison between five multifactor variants of the capital asset pricing model. These include additional factors based on size, book to market value, momentum, liquidity and a new investor protection metric based on the product of institutional quality in a country and the proportion of free float shares, which captures the impact of controlling block holders. Using monthly returns of 909 blue chip firms from 18 Middle East & North African equity markets for 16 years, we show that a two factor CAPM augmented with a factor mimicking portfolio based on the investor protection metric yields the highest explanatory power. Analysis of Kalman filter time varying investor protection betas reveals investor protection premiums in Egypt, Iraq, Lebanon and Tunisia and corresponding discounts in Israel, Saudi Arabia, Kuwait, Oman, Dubai and Abu Dhabi
Social-aware secret key generation for secure device-to-device communication via trusted and non-trusted relays
No full textPhysical layer security (PLS) is a promising technology in device-to-device (D2D) communications by exploiting reciprocity and randomness of wireless channels, which attracts considerable research attention in the D2D communications community. In this paper, we investigated PLS for secure key generation rate (SKGR) in D2D communications based on cooperative trusted and non-trusted relays. By leveraging social ties, we exploit three social phenomena for secure communications, i.e., trusted scenario (social trust), non-trusted scenario (social reciprocity) and partially trusted scenario (mixed social trust and social reciprocity). The coalition game theory is further utilized to select the optimal relay pairs for improving SKGR. On the basis of social ties, we develop an algorithm for SKGR that protects the keys secret from both eavesdropper and nontrusted selected relays. We incorporate secure relays selection and system wide security for D2D communications. The stability and convergence of the proposed algorithm are also proved in our work. Both numerically and analytically results verify effectiveness and consistency of our proposed scheme, which ensures better SKGR performance in D2D communications.</p
The fourth industrial revolution and environmental efficiency: The role of fintech industry
No full textIn the recent years, fintech industry of the fourth industrial revolution has grown multifold, which raised the concerns of scholars over the excessive usage of electricity. This paper places contribution to the existing literature by analyzing the impact of fintech industry on environmental efficiency across selected EU countries. We also utilized indicators like high-tech industry and e-commerce along with fintech industry to better understand the relationship between fourth industrial revolution and environmental efficiency. This study used Data Envelopment Analysis (DEA) to evaluate environmental efficiency using two different techniques i.e., Slack-Base Measure (SBM) and Epsilon-Based Measure (EBM). Method of Moments Quantile (MMQ) regression is employed as a basic regression technique, while instrumental variables Generalized method of Moments (IV-GMM) is used for robust analysis. The results show that, the overall environmental efficiency of EU countries have improved over the years. As the indicators of the fourth industrial revolution, fintech industry and e-commerce exert a positive effect and improve environmental efficiency; however, high-tech industry reduces environmental efficiency. The results further show that, economic growth and green finance investment promote environmental efficiency, while industrialization and R&D deteriorates it. The results can be of special interest for the policy makers of technological world
Many-objective optimization based intrusion detection for in-vehicle network security
Full text linkIn-vehicle network security plays a vital role in ensuring the secure information transfer between vehicle and Internet. The existing research is still facing great difficulties in balancing the conflicting factors for the in-vehicle network security and hence to improve intrusion detection performance. To challenge this issue, we construct a many-objective intrusion detection model by including information entropy, accuracy, false positive rate and response time of anomaly detection as the four objectives, which represent the key factors influencing intrusion detection performance. We then design an improved intrusion detection algorithm based on many-objective optimization to optimize the detection model parameters. The designed algorithm has double evolutionary selections. Specifically, an improved differential evolutionary operator produces new offspring of the internal population, and a spherical pruning mechanism selects the excellent internal solutions to form the selected pool of the external archive. The second evolutionary selection then produces new offspring of the archive, and an archive selection mechanism of the external archive selects and stores the optimal solutions in the whole detection process. An experiment is performed using a real-world in-vehicle network data set to verify the performance of our proposed model and algorithm. Experimental results obtained demonstrate that our algorithm can respond quickly to attacks and achieve high entropy and detection accuracy as well as very low false positive rate with a good trade-off in the conflicting objective landscape
LCDMA: lightweight cross-domain mutual identity authentication scheme for internet of things
No full textWith the widespread popularity of mobile terminals in the Internet of Things (IoT), the demand for cross-domain access of mobile terminals between different regions has also increased significantly. The nature of wireless communication media makes mobile terminals vulnerable to security threats in cross-domain access. Identity authentication is a prerequisite for secure data transmission in the cross-domain, and it is also the first step to guarantee the credibility of data sources. Most existing authentication schemes are based on bilinear pairing or public-key encryption and decryption with high computation overhead, which are not suitable for the resource-limited mobile IoT terminals. Moreover, these schemes have some security drawbacks and cannot meet the security requirements of cross-domain access. In this article, we propose a lightweight cross-domain mutual identity authentication (LCDMA) for the mobile IoT environment. LCDMA uses a symmetric polynomial instead of high-complexity bilinear pairing in the traditional schemes. We theoretically analyze the security performance under the random oracle model. Our results show that LCDMA not only resists common attacks but also preserves secure traceability while guaranteeing anonymity. Performance evaluation further demonstrates that our scheme has better performance in terms of computation and communication overhead, compared with other existing representative schemes.</p
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
SLIM: a secure and lightweight multi-authority attribute-based signcryption scheme for IoT
Full text linkAlthough attribute-based signcryption (ABSC) offers a promising technology to ensure the security of IoT data sharing, it faces a two-fold challenge in practical implementation, namely, the linearly increasing computation and communication costs and the heavy load of single authority based key management. To this end, we propose a Secure and Lightweight Multi-authority ABSC scheme called SLIM in this paper. The signcryption and de-signcryption costs of devices are reduced to a small constant by offloading most of the computation to the edge server. To minimize communication and storage costs, a short and constant-size ciphertext is designed. Moreover, we adopt a hierarchical multi-authority architecture, setting up multiple attribute authorities that manage keys independently to prevent the bottleneck. Rigorous security analysis proves that the SLIM scheme can resist adaptive chosen ciphertext attacks and adaptive chosen message attacks under the standard model. Simulation experiments demonstrate the correctness of our theoretical derivations and the cost reduction of the SLIM scheme in computation, communication and storage
- …
